rjones (Fri, 30 Aug 2019 15:59:42 GMT):
joshuafernandes
rjones (Fri, 30 Aug 2019 16:00:08 GMT):
User User_1 added by rjones.
Khaled.MH (Fri, 30 Aug 2019 16:12:40 GMT):
Has joined the channel.
VipinB (Fri, 30 Aug 2019 18:55:48 GMT):
Has joined the channel.
VipinB (Fri, 30 Aug 2019 18:56:46 GMT):
Welcome #besu
lucassaldanha (Sat, 31 Aug 2019 15:24:42 GMT):
Has joined the channel.
rjones (Sun, 01 Sep 2019 14:57:23 GMT):
Has left the channel.
shemnon (Wed, 04 Sep 2019 15:34:01 GMT):
Has joined the channel.
rjones (Wed, 04 Sep 2019 15:57:27 GMT):
Has joined the channel.
ajsutton (Wed, 04 Sep 2019 22:34:05 GMT):
Has joined the channel.
VijayMichalik (Wed, 04 Sep 2019 22:41:02 GMT):
Has joined the channel.
bretthenderson (Wed, 04 Sep 2019 22:45:02 GMT):
Has joined the channel.
EdJoJob (Wed, 04 Sep 2019 22:51:42 GMT):
Has joined the channel.
shahankhatch (Wed, 04 Sep 2019 23:14:56 GMT):
Has joined the channel.
bretthenderson (Thu, 05 Sep 2019 00:59:48 GMT):
@rjones after the migration of the code from PegaSys to HLF repos we have a 1.3 Release Candidate and then Release due soon after (RC 25/26 Sept, Release 2/3 Oct) Given these will be the first "releases" from the Besu code base are there any requirements we will need to meet (like Foundation approval) or do we execute the releases as we would have prior to the migration?
rjones (Thu, 05 Sep 2019 01:00:43 GMT):
@bretthenderson I think there is an email thread with @grace.hartley and @dhuseby on that very topic
grace.hartley (Thu, 05 Sep 2019 01:00:43 GMT):
Has joined the channel.
rjones (Thu, 05 Sep 2019 01:01:31 GMT):
@dhuseby is really the point person there on process
bretthenderson (Thu, 05 Sep 2019 04:36:41 GMT):
Thanks Ry.
faraggi (Thu, 05 Sep 2019 07:37:02 GMT):
Has joined the channel.
AnilSingh (Fri, 06 Sep 2019 16:58:17 GMT):
Has joined the channel.
madelinemurray (Sun, 08 Sep 2019 19:38:18 GMT):
Has joined the channel.
mbaxter (Mon, 09 Sep 2019 13:29:31 GMT):
Has joined the channel.
macfarla (Mon, 09 Sep 2019 20:03:53 GMT):
Has joined the channel.
NicolasMassart (Wed, 11 Sep 2019 22:14:17 GMT):
Has joined the channel.
dhuseby (Thu, 12 Sep 2019 15:28:25 GMT):
@bretthenderson Grace and I met earlier this week to discuss this
dhuseby (Thu, 12 Sep 2019 15:28:37 GMT):
I got the security team contact information
dhuseby (Thu, 12 Sep 2019 15:29:01 GMT):
and I sent her in the right direction on the CII badge stuff https://www.coreinfrastructure.org/programs/badge-program/
dhuseby (Thu, 12 Sep 2019 15:29:34 GMT):
and we are putting together a checklist of criteria for security audits so that we can send out RFPs to get bids from some auditors
dhuseby (Thu, 12 Sep 2019 15:30:12 GMT):
@bretthenderson that's pretty much it for now...we have a plan for migrating from your JIRA to our JIRA but that will happen *after* the release
ajsutton (Thu, 12 Sep 2019 15:30:48 GMT):
Should mention this here too:
We plan to migrate the PegaSysEng/pantheon repo to hyperledger/besu and do the tech.pegasys.pantheon to org.hyperledger.besu rename on 16 Sept AEST (UTC+10). There will be a code freeze from 1600 on 15 Sept AEST (UTC+10) until the migration is completed. We'll notify this channel when that is complete. Any work not on the master branch of the PegaSysEng/pantheon repo by the time of the code freeze will need to be manually migrated to the new namespace and applied to a fresh clone of the new hyperledger/besu repo.
faisal.khan (Thu, 12 Sep 2019 15:59:06 GMT):
Has joined the channel.
shemnon (Thu, 12 Sep 2019 16:26:50 GMT):
We are starting our #HyperledgerBesu contributor calls next week with an APAC/AMER friendly time, 0100 UTC on 17 Sep (or 11am Brisbane Tues 17 Sep, 9pm SF Mon 16 Sep) - https://wiki.hyperledger.org/display/BESU/2019-09-17+Besu+Contributor+Call
Our next call will be an EMEA/AMER friendly time at 1500 UTC on Tues 1 Oct (or 5pm Paris/Berlin, 11am New York, no agenda yet). Calls will be bi-weekly and will alternate APAC/EMEA friendliness.
bretthenderson (Thu, 12 Sep 2019 22:13:37 GMT):
Thanks Dave. Will work with you to sort out how we migrate our Jira tickets over as you say, after release.
usmansaleem (Fri, 13 Sep 2019 00:21:01 GMT):
Has joined the channel.
drinkcoffee (Fri, 13 Sep 2019 01:30:08 GMT):
Has joined the channel.
antonydenyer (Fri, 13 Sep 2019 08:07:28 GMT):
Has joined the channel.
josh-richardson (Fri, 13 Sep 2019 15:25:54 GMT):
Has joined the channel.
iikirilov (Fri, 13 Sep 2019 15:26:55 GMT):
Has joined the channel.
dhuseby (Fri, 13 Sep 2019 20:54:45 GMT):
roger
pinges (Mon, 16 Sep 2019 00:27:02 GMT):
Has joined the channel.
GinaRubino (Mon, 16 Sep 2019 11:21:48 GMT):
Has joined the channel.
jakehaugen (Mon, 16 Sep 2019 14:17:59 GMT):
Has joined the channel.
timbeiko (Mon, 16 Sep 2019 15:14:10 GMT):
Has joined the channel.
rjones (Mon, 16 Sep 2019 15:33:39 GMT):
Has left the channel.
carlosfaria (Mon, 16 Sep 2019 21:32:46 GMT):
Has joined the channel.
EdJoJob (Tue, 17 Sep 2019 03:19:08 GMT):
Hi all, a _very_ basic and not-at-all comprehensive guide to getting started with contributing to Hyperledger Besu is at https://wiki.hyperledger.org/display/BESU/How+to+Contribute
daisuke1983 (Tue, 17 Sep 2019 03:49:11 GMT):
Has joined the channel.
rjones (Tue, 17 Sep 2019 18:48:07 GMT):
Has joined the channel.
rjones (Tue, 17 Sep 2019 18:48:52 GMT):
hi - could we have the DCO discussion here in public please? @shemnon @rai
rai (Tue, 17 Sep 2019 18:48:52 GMT):
Has joined the channel.
rjones (Tue, 17 Sep 2019 18:49:40 GMT):
It is arguable if the CAB line matters for DCO compliance or not.
rai (Tue, 17 Sep 2019 18:50:23 GMT):
Who can we reach out to about that?
rai (Tue, 17 Sep 2019 18:50:32 GMT):
To check if it does or not
rjones (Tue, 17 Sep 2019 18:51:25 GMT):
@dhuseby is this a discussion you're having with Dolan right now?
rjones (Tue, 17 Sep 2019 18:51:42 GMT):
@rai in the past we've used multiple SoB lines.
shemnon (Tue, 17 Sep 2019 18:52:12 GMT):
I'm trying to understand the objection, it wasn't clear from my first reading. You want more than just the email corrected, but also have it changed to SOB. That's fine.
rjones (Tue, 17 Sep 2019 18:52:38 GMT):
For now, for today.
rjones (Tue, 17 Sep 2019 18:52:54 GMT):
since I know that is A-OK and there is no objection from LF Legal to that.
rjones (Tue, 17 Sep 2019 18:53:17 GMT):
There is a parallel conversation that I think @dhuseby is having with LF Legal about this.
rjones (Tue, 17 Sep 2019 18:54:01 GMT):
The bug in todo project that I linked to is basically all of LF Legal and the big project management (CNCF, etc) saying this is an example of a problem
shemnon (Tue, 17 Sep 2019 18:54:02 GMT):
Without the change to SOB part I was confused as to why it even mattered. But if we require SOB by all claimed authors and skip the Co-Authored-by statement I think that will make it clear it was a multi-authored commit.
rjones (Tue, 17 Sep 2019 18:54:19 GMT):
yes, it does.
rjones (Tue, 17 Sep 2019 18:55:19 GMT):
I cannot wait until I have said the final word that I ever say about DCO compliance, but I fear my headstone will read `Signed-off-by: Ry Jones
rai (Tue, 17 Sep 2019 18:56:01 GMT):
LMAO
rjones (Tue, 17 Sep 2019 18:56:02 GMT):
Your earlier point about the tool itself not checking is correct. However, the tools are a first line of defense. The people that may be checking someday are lawyers
rai (Tue, 17 Sep 2019 18:57:23 GMT):
Ok, so two sign off lines for now, I'll let chris hare know
rjones (Tue, 17 Sep 2019 18:57:46 GMT):
OK. You, personally, can fix the commit message and be done with it on this PR
rjones (Tue, 17 Sep 2019 18:57:58 GMT):
here is another tool: https://gist.github.com/ryjones/07ac650dcc5e83c91e8308ec98bacda4 that I've used
rai (Tue, 17 Sep 2019 19:00:53 GMT):
Ah, ok, I thought it had to be done by the person for enforcement reasons. But what about that error about the email? Does he need to provide a repliable email?
rjones (Tue, 17 Sep 2019 19:01:19 GMT):
yes. I added the corrected thingy in a comment.
rai (Tue, 17 Sep 2019 19:01:20 GMT):
I could sign off if I just use the current github email that's there
rai (Tue, 17 Sep 2019 19:01:27 GMT):
ahh
rai (Tue, 17 Sep 2019 19:01:30 GMT):
let me check
rai (Tue, 17 Sep 2019 19:01:33 GMT):
I think I missed that
rjones (Tue, 17 Sep 2019 19:01:50 GMT):
https://github.com/hyperledger/besu/pull/19#issuecomment-532334992
rjones (Tue, 17 Sep 2019 19:02:12 GMT):
since that is the SoB line he's used on other commits in the Besu codebase
rai (Tue, 17 Sep 2019 19:02:35 GMT):
Wait, so the sign off doesn't have to be by the authors? That's Adrian's email and we're talking about a commit by Chris Hare and I
rjones (Tue, 17 Sep 2019 19:03:11 GMT):
OK I looked up the wrong person then :)
rjones (Tue, 17 Sep 2019 19:03:39 GMT):
go here: https://github.com/hyperledger/besu/commits?author=CjHare
rjones (Tue, 17 Sep 2019 19:04:05 GMT):
I was copying the SoB line from that log, sorry for the confusion
rjones (Tue, 17 Sep 2019 19:04:05 GMT):
I was copying the SoB line from that log, sorry for _my_ confusion
sagregg (Tue, 17 Sep 2019 19:06:48 GMT):
Has joined the channel.
rjones (Tue, 17 Sep 2019 19:21:45 GMT):
```Rys-iMac:pantheon ry$ git log -1 --format=fuller
12:20:24.738798 exec-cmd.c:139 trace: resolved executable path from Darwin stack: /Library/Developer/CommandLineTools/usr/bin/git
12:20:24.739410 exec-cmd.c:236 trace: resolved executable dir: /Library/Developer/CommandLineTools/usr/bin
12:20:24.739792 git.c:419 trace: built-in: git log -1 --format=fuller
12:20:24.741980 run-command.c:643 trace: run_command: unset GIT_PAGER_IN_USE; LESS=FRX LV=-c less
commit 44bd845a6a7f811f6751caf15f2eee879d7690ad (HEAD -> master)
Author: CJ Hare
rjones (Tue, 17 Sep 2019 19:21:45 GMT):
```Rys-iMac:pantheon ry$ git log -1 --format=fuller
12:20:24.738798 exec-cmd.c:139 trace: resolved executable path from Darwin stack: /Library/Developer/CommandLineTools/usr/bin/git
12:20:24.739410 exec-cmd.c:236 trace: resolved executable dir: /Library/Developer/CommandLineTools/usr/bin
12:20:24.739792 git.c:419 trace: built-in: git log -1 --format=fuller
12:20:24.741980 run-command.c:643 trace: run_command: unset GIT_PAGER_IN_USE; LESS=FRX LV=-c less
commit 44bd845a6a7f811f6751caf15f2eee879d7690ad (HEAD -> master)
Author: CJ Hare
shemnon (Tue, 17 Sep 2019 19:25:38 GMT):
So is it safe to say that contributor-licence-agreements are not good enough for LF?
shemnon (Tue, 17 Sep 2019 20:41:51 GMT):
Why is `Dismiss stale pull request approvals when new commits are pushed` enabled on the repo? This is a significant break to our workflow. Consider that we require a clean Jenkins build for each PR on top of master that means whenever someone else's PR goes in and we have to merge from master that we need to get a new approval every time.
shemnon (Tue, 17 Sep 2019 20:43:17 GMT):
This is cause because `Require branches to be up to date before merging` is also selected, and we then have a status check afterwards.
rai (Tue, 17 Sep 2019 20:45:40 GMT):
Additionally, PRs can be approved with understanding that the comments are uncontroversial and will be addressed. This is especially important when collaborating with folks on the other side of the world
shemnon (Tue, 17 Sep 2019 20:55:24 GMT):
This may also go with approve with conditions, "Approved assuming the CRLF issue is fixed"
ajsutton (Tue, 17 Sep 2019 21:05:07 GMT):
I've unchecked `Dismiss stale pull request approvals when new commits are pushed`.
EdJoJob (Tue, 17 Sep 2019 22:36:44 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=CDARpfoLz5ZpGWuRB) this ultimately points at a bash script, which we have embedded (with modifications) into our CI pipelines. So now we can have _at most_ 1 non-DCO commit on master
ajsutton (Tue, 17 Sep 2019 22:42:47 GMT):
Yes unfortunately there's no way with GitHub to check the DCO on a squash & merge commit before it hits master. Hopefully we can get into the habit of keeping sign off lines but otherwise we may need to look at doing the squash on the branch and then a rebase merge. We'll see how it goes.
EdJoJob (Tue, 17 Sep 2019 23:49:48 GMT):
@rjones could we please have a homebrew-besu repository
rjones (Wed, 18 Sep 2019 04:45:20 GMT):
@EdJoJob https://github.com/hyperledger/homebrew-besu
faraggi (Wed, 18 Sep 2019 09:34:32 GMT):
I'm wondering how we will enforce this on/educate community contributors. @rjones Do you have any info on how much of a barrier this is to contributors?
faraggi (Wed, 18 Sep 2019 09:36:31 GMT):
@EdJoJob https://github.com/hyperledger/homebrew-besu
faraggi (Wed, 18 Sep 2019 09:37:18 GMT):
The only thing in there is the apache license. Such a OSS thing. 🙂
Thanks.
rjones (Wed, 18 Sep 2019 13:46:06 GMT):
@faraggi I usually initialize new repos with just the default license file, since it's a requirement for our projects.
rjones (Wed, 18 Sep 2019 15:06:57 GMT):
@faraggi here is a counter example: https://github.com/hyperledger/hl-landscape where I'm waiting for a repo to come in from CNCF
faraggi (Wed, 18 Sep 2019 16:09:32 GMT):
Oh, to be clear: I think its great to do so! :thumbsup:
rjones (Wed, 18 Sep 2019 16:37:17 GMT):
I shipped some Besu poker chips to Australia, so you'll all need to meet with @EdJoJob to get one
shemnon (Wed, 18 Sep 2019 18:39:15 GMT):
Besu's codebase has a code of conduct - https://github.com/hyperledger/besu/blob/master/CODE-OF-CONDUCT.md
Hyperledger Project has a code of conduct - https://wiki.hyperledger.org/display/HYP/Hyperledger+Code+of+Conduct
Can these live side by side or do we need to update the CoC in besu or do we need to mention that both apply? @rjones
rjones (Wed, 18 Sep 2019 18:49:06 GMT):
@shemnon I'm not sure?
rjones (Wed, 18 Sep 2019 18:49:06 GMT):
@shemnon I'm not sure - could you please send email to community-architects@hyperledger.org ?
shemnon (Wed, 18 Sep 2019 21:33:55 GMT):
I just committed the fork block numbers for Goerli and Ropsten. Since the next release is supposed to be 1.3-RC perhaps we should consider a 1.2.4 release so the ropsten fork is off a client not labeled "RC"
arash009 (Wed, 18 Sep 2019 21:49:21 GMT):
Has joined the channel.
timbeiko (Wed, 18 Sep 2019 21:53:14 GMT):
Good point. We can discuss this in the product meeting tomorrow. Basically we need to figure out if the 1.2.4 replaces the RC, is the RC or if we want a RC one week after
arash009 (Wed, 18 Sep 2019 21:56:13 GMT):
I'm not sure we want to re-schedule the RC at such short notice, there are probably a few other pieces that need the extra week to be finished off for actual RC, but I dont see why not do a 1.2.4 this week and keep RC planned for next as normal
shemnon (Wed, 18 Sep 2019 21:59:16 GMT):
I was thinking of an out of cycle release, just for the fork numbers.
shemnon (Wed, 18 Sep 2019 21:59:29 GMT):
Still do 1.3-RC as scheduled and 1.2.4 relatively "soon"
timbeiko (Wed, 18 Sep 2019 22:02:01 GMT):
That sounds reasonable! The only PRs we need are the fork block ones, right?
timbeiko (Wed, 18 Sep 2019 22:02:09 GMT):
And release notes
shemnon (Wed, 18 Sep 2019 23:09:39 GMT):
The question is if we do a CD style "here it is" or do a fork off of 1.2.3. There are a few more PRs (including the besu renaming) than just the fork block, but nothing worse than what was done 1.2.2->1.2.3 with the database stuff
shemnon (Wed, 18 Sep 2019 23:13:52 GMT):
I do think we need to wait for Geth and Parity to +1 the testnet forks.
RobDawson (Wed, 18 Sep 2019 23:25:33 GMT):
Has joined the channel.
RobDawson (Wed, 18 Sep 2019 23:25:34 GMT):
I'd like to suggest we go ahead and do a 1.2.4 release this week.
RobDawson (Wed, 18 Sep 2019 23:26:33 GMT):
@EdJoJob and team I think are working to coordinate this happening.
ajsutton (Wed, 18 Sep 2019 23:31:14 GMT):
I'm strongly in favour of sticking with our continuous delivery process and 1.2.4 comes off master as usual. The only thing to check is that pruning is still off by default (which I think it is).
bretthenderson (Thu, 19 Sep 2019 00:24:57 GMT):
Current thinking is we do a 1.2.4 release Friday AM (Aust)/Thursday PM (US)
shemnon (Thu, 19 Sep 2019 03:58:07 GMT):
Seriously, lets hold off until Geth and Parity signal it is OK. May a bit paranoid.
shemnon (Thu, 19 Sep 2019 03:58:11 GMT):
may not be
shemnon (Thu, 19 Sep 2019 03:58:27 GMT):
There is an all core devs call this week.
shemnon (Thu, 19 Sep 2019 03:59:05 GMT):
So first thing monday AUS looks like a better target.
RobDawson (Thu, 19 Sep 2019 06:09:52 GMT):
sounds reasonable. We should check in tomorrow?
RobDawson (Thu, 19 Sep 2019 06:10:54 GMT):
given the most pressing reason for the release was to get the hardfork numbers out, holding until the signals from geth and parity makes sense.
NicolasMassart (Thu, 19 Sep 2019 08:49:23 GMT):
Would 1.2.4 include changes from https://pegasys1.atlassian.net/browse/PAN-3140 and https://pegasys1.atlassian.net/browse/PAN-3155 that are already in master?
kiichi (Thu, 19 Sep 2019 10:03:07 GMT):
Has joined the channel.
shemnon (Thu, 19 Sep 2019 13:58:29 GMT):
So Hudson Jamison chimed in on the AllCoreDevs gitter last night and made the fork numbers official. We can release as soon as the release notes are ready.
shemnon (Thu, 19 Sep 2019 14:58:36 GMT):
Also, Rinkeby set their fork block, for 13 Nov. PR is out for review.
shemnon (Thu, 19 Sep 2019 15:54:16 GMT):
PR is merged. The next stop is a mainnet block number.
shemnon (Thu, 19 Sep 2019 15:54:30 GMT):
And that won't be for at least 6 weeks.
ajsutton (Thu, 19 Sep 2019 19:53:11 GMT):
Yes, as usual we'd take the release from current master. And for the benefit of those new to besu, we follow trunk based development (https://trunkbaseddevelopment.com is a good explanation and I'd recommend the book `Continuous Delivery` by Humble and Farley as well).
timbeiko (Thu, 19 Sep 2019 21:28:54 GMT):
OK, so final schedule for the next few releases is as follows, using Australia/USA dates:
23/24 Sept: Besu 1.2.4
6/7 Oct: Besu 1.3
timbeiko (Fri, 20 Sep 2019 12:51:48 GMT):
For clarity, after a lot of back and forth :sweat_smile:, here are the dates for the next couple Besu releases for Australia/Americas:
1.2.4: 23/24 Sept
1.3: 3/4 Oct
joshuafernandes (Mon, 23 Sep 2019 00:17:53 GMT):
1.2.4 currently slated in for 0800 AEST 24/9
cjhare (Mon, 23 Sep 2019 22:02:15 GMT):
Has joined the channel.
EdJoJob (Mon, 23 Sep 2019 22:05:25 GMT):
@channel - We're beginning the release process for besu, please don't merge anything to master until further notice
EdJoJob (Tue, 24 Sep 2019 02:14:35 GMT):
It's now safe to merge into master again.
shemnon (Tue, 24 Sep 2019 19:36:16 GMT):
We have a contributor's call next week. If there is anything you would like discussed please add it to the agenda - https://wiki.hyperledger.org/display/BESU/2019-10-01+Besu+Contributor+Call
circlespainter (Wed, 25 Sep 2019 08:44:10 GMT):
Has joined the channel.
shemnon (Wed, 25 Sep 2019 17:35:35 GMT):
FYI we just did a DCO related master rollback, from commit b5a88e96777529e97321b9e9fe8132c6b8f788a8 rolling back to 497e825919415e618567d61b6aa0241033c0ebf4.
bretthenderson (Fri, 27 Sep 2019 01:08:47 GMT):
*ALL* Besu 1.3 release activities are happening next week so make sure your PRs are ready for inclusion.
*Besu 1.3.0 RC* is planned for *Tuesday 1st October* starting at 8am AEST
*Besu 1.3.0* release is planned for *Friday 4th October* (also at 8am AEST)
rjones (Fri, 27 Sep 2019 08:07:36 GMT):
https://lgtm.com/projects/g/hyperledger/besu/alerts?mode=list
NicolasMassart (Fri, 27 Sep 2019 08:32:45 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=unfyjkh87kNG74yPk)
Clipboard - September 27, 2019 10:32 AM
rjones (Fri, 27 Sep 2019 09:46:48 GMT):
@NicolasMassart considering the size of the code base, and the nature of the outstanding alerts, I think it's awesome
rjones (Fri, 27 Sep 2019 09:47:29 GMT):
I mean... https://lgtm.com/projects/g/hyperledger/indy-plenum/?mode=list
rjones (Fri, 27 Sep 2019 09:48:25 GMT):
the drill down tool is neat, too: https://lgtm.com/projects/g/hyperledger/indy-plenum/latest/files/common/version.py?sort=name&dir=ASC&mode=heatmap
NicolasMassart (Fri, 27 Sep 2019 11:13:35 GMT):
Maybe we should add the badge to our readme.
joshuafernandes (Sun, 29 Sep 2019 22:17:02 GMT):
@here: *Besu 1.3 RC* slated for release this time tomorrow ie *0800 AEST 1/10*
cjhare (Sun, 29 Sep 2019 22:30:27 GMT):
Well badges are pretty neat!
cjhare (Sun, 29 Sep 2019 22:30:27 GMT):
Well badges are pretty neat
cjhare (Sun, 29 Sep 2019 22:30:27 GMT):
Well badges are pretty neat, especially when they nice things about your project :slight_smile:
shemnon (Sun, 29 Sep 2019 23:39:20 GMT):
Is LGTM something being added to all relevant Hyperledger repos? @rjones would you be willing to speak to it on our contributors call tuesday?
rjones (Mon, 30 Sep 2019 00:55:52 GMT):
@shemnon I will probably be on a plane - but I did add it to every repo the had a codebase in a language LGTM supports
rjones (Mon, 30 Sep 2019 00:59:15 GMT):
@shemnon that would be pretty late for Tokyo :)
NicolasMassart (Mon, 30 Sep 2019 09:34:33 GMT):
https://github.com/hyperledger/besu/pull/68
iikirilov (Mon, 30 Sep 2019 16:13:41 GMT):
https://github.com/hyperledger/besu/pull/62 makes all future PRs automatically request reviews from everyone specified in the CODEOWNDERS file. Is this intentional?
shemnon (Mon, 30 Sep 2019 17:18:39 GMT):
Yea, I'm not liking that PR either. Unless we are going to push down code owners to specific directories and make it a subset of the maintainers it looks like Github's CODEOWNERS is a net negative as it makes requrested reviewers hard to use and adds a whole lot of noise.
shemnon (Mon, 30 Sep 2019 17:19:58 GMT):
Furthermore, perhaps these repo wide change should be discussed in rocketchat first?
rjones (Mon, 30 Sep 2019 18:48:46 GMT):
@iikirilov I don't have strong feelings about how code ownership is split up. If that PR isn't to your liking, that's fine.
shemnon (Mon, 30 Sep 2019 19:39:47 GMT):
Here's the PR for the rollback: https://github.com/hyperledger/besu/pull/69
The mechanics of what it does to reviewers are what is prompting me here. Otherwise I would be neutral.
EdJoJob (Mon, 30 Sep 2019 22:34:58 GMT):
We're beginning the release process for besu, please don't merge anything to master until further notice
EdJoJob (Mon, 30 Sep 2019 23:08:09 GMT):
It's now safe to merge into master again.
shemnon (Tue, 01 Oct 2019 14:56:25 GMT):
besu contributor call in 4 min: https://consensys.zoom.us/j/725448960
jframe (Wed, 02 Oct 2019 01:00:34 GMT):
Has joined the channel.
faraggi (Wed, 02 Oct 2019 08:37:20 GMT):
Yesterday's contributor call notes have been added to the wiki:
https://wiki.hyperledger.org/display/BESU/2019-10-01+Contributor+call
lucassaldanha (Wed, 02 Oct 2019 22:13:36 GMT):
We have a proposal for a new Besu project maintainer!
https://github.com/hyperledger/besu/pull/83
Maintainers, please vote within two weeks.
To learn more about this process, check https://github.com/hyperledger/besu/blob/master/MAINTAINERS.md#becoming-a-maintainer
arash009 (Fri, 04 Oct 2019 00:07:24 GMT):
Besu 1.3 has been released!! Get your copy here: https://bintray.com/hyperledger-org/besu-repo/besu/1.2.4#
Change log available at: https://github.com/hyperledger/besu/releases/tag/1.3.0
arash009 (Fri, 04 Oct 2019 00:07:24 GMT):
Besu 1.3 has been released!! Get your copy here: https://bintray.com/hyperledger-org/besu-repo/besu/1.3.0#
Change log available at: https://github.com/hyperledger/besu/releases/tag/1.3.0
NicolasMassart (Thu, 10 Oct 2019 10:35:50 GMT):
As we have a maintainers list on Besu, I propose to have a list also on Besu-doc and make them having different people given that writing doc is not the same as writing code: https://github.com/hyperledger/besu-docs/pull/47
NicolasMassart (Thu, 10 Oct 2019 19:42:08 GMT):
Hi Besu contributors, I'm working on https://github.com/hyperledger/besu/pull/36 and it's becoming bigger every day... so I think it will be hard to review if I provide this in one big PR. But merging this unfinished work would require CLI options to be disabled as we probably don't want to have user test it until it's more advanced. So I don't know how to handle this big long lasting thing. Any advice? Thanks.
arash009 (Mon, 14 Oct 2019 21:08:16 GMT):
Hi all. *Besu 1.3.1* release is planned for 16 October. Change log will be distributed as part of the release. Please reach out if you have any questions.
arash009 (Tue, 15 Oct 2019 00:06:49 GMT):
Hi all, Besu 1.3.1 is scheduled for release 16 October. Change log will be provided with the release. Any questions please reach out!
lucassaldanha (Tue, 15 Oct 2019 01:27:03 GMT):
Hi all! Ivaylo has just been accepted as a Besu maintainer after the proposal reached absolute majority. Welcome Ivaylo!
https://github.com/hyperledger/besu/pull/83
EdJoJob (Tue, 15 Oct 2019 22:29:26 GMT):
We're beginning the release process for besu, please don't merge anything to master until further notice
EdJoJob (Tue, 15 Oct 2019 23:33:06 GMT):
It's now safe to merge into master again.
hiron (Wed, 16 Oct 2019 14:19:30 GMT):
Has joined the channel.
shemnon (Fri, 18 Oct 2019 16:49:39 GMT):
When we migrated the old pantheon repo we missed the govenance document (which no longer applies). All maintainers please review this PR - https://github.com/hyperledger/besu/pull/120
bretthenderson (Mon, 21 Oct 2019 00:42:37 GMT):
Hi all, Besu 1.3.2 is being planned for release this week, Tuesday 22nd (Wednesday 23 AEST). Please raise any questions or concerns asap.
joshuafernandes (Tue, 22 Oct 2019 22:12:15 GMT):
@channel - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
joshuafernandes (Tue, 22 Oct 2019 22:12:40 GMT):
@all - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
bretthenderson (Wed, 23 Oct 2019 05:27:43 GMT):
Hi All, It's now safe to merge into master for hyperledger/besu again.
rjones (Thu, 24 Oct 2019 16:25:15 GMT):
please add github ID `ryjones` as an owner/admin on https://readthedocs.org/projects/hyperledger-besu/ thank you
madelinemurray (Thu, 24 Oct 2019 19:54:51 GMT):
Done :-)
rjones (Thu, 24 Oct 2019 20:00:01 GMT):
thanks
SteveMieskoski (Sun, 27 Oct 2019 01:02:05 GMT):
Has joined the channel.
joshuafernandes (Mon, 28 Oct 2019 23:26:33 GMT):
Hello @all, we are looking to move CI of Besu over to CircleCI over the next few weeks. In the interim we will use Jenkins & Circle simultaneously; of which only Jenkins will count towards builds being merge-able on PRs and the like.
You may see some oddities in the Circle build checks during this time - please disregard these for the moment.
If you have questions or concerns please feel free to reach out.
shemnon (Tue, 29 Oct 2019 14:14:23 GMT):
Besu contributor call in 45 minutes. Agenda - https://wiki.hyperledger.org/display/BESU/2019-10-29+Besu+Contributor+Call
NicolasMassart (Wed, 30 Oct 2019 16:43:13 GMT):
https://github.com/hyperledger/besu-docs/pull/47 met the acceptance criteria so I merged the new Besu doc repos maintainers list in master.
Who is responsible to make the actual repos maintainers match this list? I'm not able to change this as I'm not a repos admin.
Thanks!
rjones (Wed, 30 Oct 2019 17:00:20 GMT):
@NicolasMassart here are the people that can do it: https://github.com/orgs/hyperledger/teams/besu-admin/members
NicolasMassart (Wed, 30 Oct 2019 17:05:16 GMT):
Thanks!
NicolasMassart (Wed, 30 Oct 2019 17:06:27 GMT):
@shemnon @mbaxter @joshuafernandes @EdJoJob @ajsutton can one of you update the Besu doc repos maintainer list? thanks.
mbaxter (Wed, 30 Oct 2019 17:12:35 GMT):
I can take a look
mbaxter (Wed, 30 Oct 2019 17:16:55 GMT):
There's a docs maintainers team, but it's missing steven gregg, and I'm not seeing a way to add him ... https://github.com/orgs/hyperledger/teams/besu-docs-maintainers/members
rjones (Wed, 30 Oct 2019 17:17:14 GMT):
@mbaxter what is his GitHub id
rjones (Wed, 30 Oct 2019 17:17:14 GMT):
@mbaxter what is his GitHub ID
mbaxter (Wed, 30 Oct 2019 17:17:27 GMT):
I sgregglives
mbaxter (Wed, 30 Oct 2019 17:17:27 GMT):
sgregglives
rjones (Wed, 30 Oct 2019 17:18:59 GMT):
Invite sent. Once they accept, you can add them. https://github.com/hyperledger is where it can be accepted
mbaxter (Wed, 30 Oct 2019 17:20:43 GMT):
thanks @rjones !
mbaxter (Wed, 30 Oct 2019 17:21:53 GMT):
@NicolasMassart - I gave the docs maintainers team write access to the docs repo, I can remove write access for the larger Besu Contributors team whenever makes sense. Should we wait for steven to accept?
rjones (Wed, 30 Oct 2019 17:24:28 GMT):
Personally, I wouldn't. I'm not on the Besu team, though :u6307:
mbaxter (Wed, 30 Oct 2019 17:26:34 GMT):
yeah, I think I'll just go ahead and update. @sagregg - let me know if you run into any issues getting access
mbaxter (Wed, 30 Oct 2019 17:27:44 GMT):
updated
rjones (Wed, 30 Oct 2019 17:33:04 GMT):
@mbaxter should edjojob be removed? he isn't in the list
mbaxter (Wed, 30 Oct 2019 17:34:25 GMT):
probably - although he's on the list of admins so it makes no functional difference at this point
rjones (Wed, 30 Oct 2019 17:37:15 GMT):
the functional difference is someone can request a review from @besu-docs-maintainers and it goes to that list. I really don't care - he has the power to remove himself, as an admin
rjones (Wed, 30 Oct 2019 17:37:19 GMT):
and it's your project.
NicolasMassart (Wed, 30 Oct 2019 17:37:20 GMT):
yes, I did not put admins in the maintainers list as they are already admins and have all access.
rjones (Wed, 30 Oct 2019 17:40:30 GMT):
The GitHub name is `@hyperledger/besu-docs-maintainers` if you want to test it out
rjones (Wed, 30 Oct 2019 17:40:41 GMT):
(on a CR)
mbaxter (Wed, 30 Oct 2019 17:40:42 GMT):
I'm not see any UI for adding / removing team members. Should I have permissions to edit the teams?
rjones (Wed, 30 Oct 2019 17:42:28 GMT):
I added you to the maintainer role of the parent team - try now?
rjones (Wed, 30 Oct 2019 17:43:00 GMT):
I hate to say "I long for the days of Active Directory ACL management" but to me GitHub's team and ACL management is opaque
mbaxter (Wed, 30 Oct 2019 17:43:16 GMT):
:joy:
mbaxter (Wed, 30 Oct 2019 17:43:36 GMT):
Still not seeing any editing UI here: https://github.com/orgs/hyperledger/teams/besu-docs-maintainers/members
rjones (Wed, 30 Oct 2019 17:43:56 GMT):
I don't know what else to do. You're an admin of the parent team and the project.
NicolasMassart (Wed, 30 Oct 2019 17:44:27 GMT):
I see one. And it seems that I can do the changes...
rjones (Wed, 30 Oct 2019 17:44:54 GMT):
I guess you have to be a member of that team to make edits?
rjones (Wed, 30 Oct 2019 17:44:57 GMT):
seems subpar
NicolasMassart (Wed, 30 Oct 2019 17:45:31 GMT):
Clipboard - October 30, 2019 6:45 PM
rjones (Wed, 30 Oct 2019 17:46:02 GMT):
@NicolasMassart correct - I already invited him; once he accepts, you can add him.
NicolasMassart (Wed, 30 Oct 2019 21:17:38 GMT):
Done thanks!
kziemianek (Sat, 02 Nov 2019 09:21:44 GMT):
Has joined the channel.
drinkcoffee (Sun, 03 Nov 2019 01:55:54 GMT):
Uncle Block question: In Ethereum, what block number and block timestamp does an uncle block receive? Is it the values which reflect when it was originally mined, or when it was included as an uncle block? To save asking related questions again, it would be great if you could supply a link to where in Besu does this processing.
ajsutton (Sun, 03 Nov 2019 02:46:22 GMT):
Besu doesn't currently find Ommer blocks to include when mining. However, ommer's are just headers of blocks that are not on the canonical chain. So the block number and everything else about the block are the same as when it was first mined.
drinkcoffee (Sun, 03 Nov 2019 06:27:02 GMT):
Thanks @ajsutton !
xds2000 (Sun, 03 Nov 2019 16:46:04 GMT):
Has joined the channel.
xds2000 (Sun, 03 Nov 2019 16:46:05 GMT):
hi team, i wonder to know what is the mainnet use-case for enterprises
xds2000 (Sun, 03 Nov 2019 16:46:35 GMT):
i quote from here: Judging Criteria
The bounty will be awarded to the best submission based on the impact of the mainnet use-case for enterprises, compliance with Contribution Guidelines, and the Submission Requirements above, according to PegaSys.
xds2000 (Sun, 03 Nov 2019 16:46:43 GMT):
https://gitcoin.co/issue/PegaSysEng/BountiedWork/21/3627
xds2000 (Mon, 04 Nov 2019 15:31:21 GMT):
anyone in here?
NicolasMassart (Mon, 04 Nov 2019 15:42:22 GMT):
Hi, yes sorry. I was waiting for the person who pushed this bounty, and because of timezones he's not available yet. But as soon as he is, I will ask him to have a answer your question. Thanks for your patience.
timbeiko (Mon, 04 Nov 2019 15:48:23 GMT):
Hi there,
This bounty is reserved for EthWaterloo participants. The bounty is left open for participants to showcase their creativity :smile: Basically, anything that helps enterprises using Ethereum. Two ways we think about this are either using mainnet as a way to bridge various private networks (even if it's as simple as posting data from a private network to mainnet and retrieving it from another network) or to make things easier for enterprises using Besu on mainnet (think integrations, reliability, perfomance, etc.).
Hopefully that helps! happy to answer any follow up questions.
oyincode (Tue, 05 Nov 2019 20:11:02 GMT):
Has joined the channel.
joshuafernandes (Tue, 05 Nov 2019 22:07:03 GMT):
@all - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
xds2000 (Wed, 06 Nov 2019 00:12:27 GMT):
Thanks a lot
joshuafernandes (Wed, 06 Nov 2019 00:28:01 GMT):
@all besu 1.3.3 release, its safe to merge back to master now
xds2000 (Wed, 06 Nov 2019 00:54:10 GMT):
cool
joshuafernandes (Wed, 06 Nov 2019 04:08:15 GMT):
@channel - We're beginning the release process for hyperledger/besu 1.3.4, please don't merge anything to master until further notice
joshuafernandes (Wed, 06 Nov 2019 04:08:15 GMT):
@all - We're beginning the release process for hyperledger/besu 1.3.4, please don't merge anything to master until further notice
arash009 (Wed, 06 Nov 2019 04:09:57 GMT):
For clarity, we've identified an issue with Besu 1.3.3 so we are in the process of releasing Besu 1.3.4 which will address this issue.
joshuafernandes (Wed, 06 Nov 2019 05:56:50 GMT):
@all besu 1.3.4 release done, its safe to merge back to master now
ericglau (Wed, 06 Nov 2019 06:45:40 GMT):
Has joined the channel.
oyincode (Wed, 06 Nov 2019 09:24:16 GMT):
hey guys I'm trying to setup the besu project on my local machine and its giving me this error,
Each time I run the gradlew build or gradlew installDist, i get this in my console.
What went wrong:
Execution failed for task ':buildSrc:compileGroovy'.
> BUG! exception in phase 'semantic analysis' in source unit 'C:\Users\onuem\Desktop\besu\buildSrc\src\main\groovy\CheckSpdxHeader.groovy' Unsupported class file major version 57
Try:
Run with --stacktrace option to get the stack trace. Run with --info or --debug option to get more log output. Run with --scan to get full insights.
Get more help at https://help.gradle.org
BUILD FAILED in 7s
oyincode (Wed, 06 Nov 2019 09:24:19 GMT):
I'm using Java 13.0.1
oyincode (Wed, 06 Nov 2019 09:24:33 GMT):
Please I'd appreciate any help
NicolasMassart (Wed, 06 Nov 2019 10:48:16 GMT):
Hi @oyincode please find the answer by @faraggi on https://chat.hyperledger.org/channel/besu?msg=Mep8Ea3TZ357gXquk
oyincode (Wed, 06 Nov 2019 10:52:06 GMT):
yes thanks i've seen the answer @NicolasMassart
shemnon (Wed, 06 Nov 2019 14:34:38 GMT):
Looks like you are using Java 13. Gradle still doesn't support it. If you use Java 8, 9, 10, 11, or 12 things should work smoothly.
rachidchami (Wed, 06 Nov 2019 18:35:44 GMT):
Has joined the channel.
mackcom (Wed, 06 Nov 2019 20:46:52 GMT):
Has joined the channel.
GregTheGreek (Wed, 06 Nov 2019 21:08:28 GMT):
Has joined the channel.
mojo (Wed, 06 Nov 2019 22:37:18 GMT):
Has joined the channel.
madelinemurray (Thu, 07 Nov 2019 02:04:39 GMT):
Thinking about what documentation is needed for the plugin API and at what point we consider plugin APIs stable enough to be considered as no longer changing/shoud be doc'd. My understanding is that we need two things:
1. Some general information about how get started building a plugin
2. There's JavaDoc in the code
Re 1. Who's a good person to talk to about this?
Re 2. Do we want to publish this outside the code?
madelinemurray (Thu, 07 Nov 2019 04:32:41 GMT):
@timbeiko - did we end up with a JI for making the `--pruning-enabled` and `--privacy-enabled` options mutually exclusive?
madelinemurray (Thu, 07 Nov 2019 04:32:41 GMT):
@timbeiko - did we end up with a JI for making the `--pruning-enabled` and `--privacy-enabled` options mutually exclusive? I wanted to add that fast-sync and privacy also need to be mutually exclusive but now I can't find the other JI.
hmijail (Thu, 07 Nov 2019 05:31:00 GMT):
Has joined the channel.
timbeiko (Thu, 07 Nov 2019 11:24:47 GMT):
Yes: https://pegasys1.atlassian.net/browse/PAN-3275
shemnon (Thu, 07 Nov 2019 15:24:18 GMT):
The technical steering committee is chancing how they are calendaring meetings. (https://wiki.hyperledger.org/display/TSC/2019+11+07+TSC+Agenda). They need to hear from people on projects who may need access to scheduling meetings. I think that would be me, Bret and Felipe? Any others? For those maintainers who want to schedule calendar items (like the recurring contributor call) we need to email David Huseby (email in the link) soon.
Nabilel 1 (Thu, 07 Nov 2019 15:31:24 GMT):
Has joined the channel.
cjhare (Fri, 08 Nov 2019 05:56:18 GMT):
Hello Besu contributors!
Prompted by recent the experience of discovering a defect immediately following the Besu 1.3.3 release (resulting in the Besu 1.3.4 release), a question was asked that I figured we could do with converting from being tribal knowledge into a written down process (e.g. a HLF Besu Wiki page).
*When do we remove the Besu binary release artifact from Bintray?*
_(context: a bug being discovered in that version of Besu)_
As a release tag is created as part of the release process, in theory anyone could build a binary of the defected version, so removal of the binary from Bintray does not prevent that version from ever existing. Perhaps never removing the binary a sensible option?
Maybe when a catastrophic defect is found (data corruption, security breach, product break), then would warrant removing the binary?
Should we potentially also be deleting the release tag in addition to removing the binary?
What are peoples thoughts on this matter?
cjhare (Fri, 08 Nov 2019 05:56:18 GMT):
Hello Besu contributors!
Prompted by recent the experience of discovering a defect immediately following the Besu 1.3.3 release (resulting in the Besu 1.3.4 release), a question was asked that I figured we could do with converting from being tribal knowledge into a written down process (e.g. a HLF Besu Wiki page).
*When do we remove the Besu binary release artifact from Bintray?*
_(context: a bug being discovered in that version of Besu)_
As a release tag is created as part of the release process, in theory anyone could build a binary of the defected version, so removal of the binary from Bintray does not prevent that version from ever existing. Perhaps never removing the binary a sensible option?
Maybe when a catastrophic defect is found (data corruption, security breach, product break), then would warrant removing the binary?
Should we potentially also be deleting the release tag in addition to removing the binary?
What are peoples thoughts on this matter?
cjhare (Fri, 08 Nov 2019 05:56:18 GMT):
Hello Besu contributors!
Prompted by recent the experience of discovering a defect immediately following the Besu 1.3.3 release (resulting in the Besu 1.3.4 release), a question was asked that I figured we could do with converting from being tribal knowledge into a written down process (e.g. a HLF Besu Wiki page).
*When do we remove the Besu binary release artifact from Bintray?*
_(context: defect discovered in that version of Besu)_
As a release tag is created as part of the release process, in theory anyone could build a binary of the defected version, so removal of the binary from Bintray does not prevent that version from ever existing. Perhaps never removing the binary a sensible option?
Maybe when a catastrophic defect is found (data corruption, security breach, product break), then would warrant removing the binary?
Should we potentially also be deleting the release tag in addition to removing the binary?
What are peoples thoughts on this matter?
cjhare (Fri, 08 Nov 2019 05:56:18 GMT):
Hello Besu contributors!
Prompted by recent the experience of discovering a defect immediately following the Besu 1.3.3 release (resulting in the Besu 1.3.4 release), a question was asked that I figured we could do with converting from being tribal knowledge into a written down process (e.g. a HLF Besu Wiki page).
*When do we remove the Besu binary release artifact from Bintray?*
_(context: defect discovered in that version of Besu)_
As a release tag is created as part of the release process, in theory anyone could build a binary of the defected version from source, so removal of the binary from Bintray does not prevent that version from ever existing. Perhaps never removing the binary a sensible option?
Maybe when a catastrophic defect is found (data corruption, security breach, product break), then that would warrant removing the binary?
Should we potentially also be deleting the release tag in addition to removing the binary?
What are peoples thoughts on this matter?
ajsutton (Fri, 08 Nov 2019 06:50:22 GMT):
I'd say remove the binary if there's data corruption in common cases or significant security breach. I wouldn't remove it for failing to start. Repositories work best when things remain reliably availalbe so it's a big deal to remove something. I wouldn't ever remove the release tag in source code - it's important to have a historical record.
hiteshjoshi1 (Fri, 08 Nov 2019 07:48:55 GMT):
Has joined the channel.
NicolasMassart (Fri, 08 Nov 2019 09:05:59 GMT):
putting the question another way: why keeping it? What's the use of a faulty release? What's the use of tagging a non working repos state?
ALso in my experience, users sometimes do upgrate going through all intermetidate versions and test at each step. So going from 1.3.2 to 1.3.4 passes by 1.3.3 which will block them. Why giving a stick to be beaten with?
ajsutton (Sat, 09 Nov 2019 01:38:28 GMT):
Because repositories work best when you can count on them to continue to have the artefacts and not have things mysteriously disappear. Otherwise you wind up having to mirror the repo and maintain your own. Similarly the github repository is an important record of history, documenting what has happened. Rewriting that history to pretend the bad release just creates confusion about where it went or what was in the actual release.
shemnon (Sat, 09 Nov 2019 19:04:13 GMT):
So that is something I need to flesh out I think. For the 1.4 release I think it's time to start making some APIs fixed and open. And we need a plugin "how to" for some plugin types. But some are quite involved (Key Value Storage) so for now the tutorial would be for something simpler.
timbeiko (Sun, 10 Nov 2019 00:53:56 GMT):
@ajsutton @EdJoJob I'm at EthWaterloo with a hacker and they are trying to run EthSigner but it crashes right after starting, displaying the `--help` info, which is odd. Not sure if any of you can help him?
timbeiko (Sun, 10 Nov 2019 00:53:56 GMT):
@ajsutton @EdJoJob @lucassaldanha I'm at EthWaterloo with a hacker and they are trying to run EthSigner but it crashes right after starting, displaying the `--help` info, which is odd. Not sure if any of you can help him?
timbeiko (Sun, 10 Nov 2019 00:54:05 GMT):
He's about to post here, setting up his LFID :-)
CengizSirlan (Sun, 10 Nov 2019 00:55:49 GMT):
Has joined the channel.
CengizSirlan (Sun, 10 Nov 2019 00:55:51 GMT):
c:\ bin\ethsigner --chain-id=2018 --downstream-http-port=8590 file-based-signer --key-file="C:\Users\Cengiz Sirlan\Desktop\ETH\keyFile.txt" --password-file="C:\Users\Cengiz Sirlan\Desktop\ETH\passwordFile.txt"
Setting logging level to INFO
2019-11-09 19:52:08.140-05:00 | main | INFO | SignerSubCommand | Version = ethsigner/v0.3.0/windows-x86_64/oracle-java-13
Usage:
ethsigner [OPTIONS] [COMMAND]
Description:
This command runs the EthSigner.
Options:
--chain-id=
CengizSirlan (Sun, 10 Nov 2019 00:56:10 GMT):
C:\Users\Cengiz Sirlan\Desktop\ETH\ethsigner-0.3.0>bin\ethsigner --chain-id=2018 --downstream-http-port=8590 file-based-signer --key-file=../keyFile.txt --password-file=../passwordFile.txt
Setting logging level to INFO
2019-11-09 19:18:54.643-05:00 | main | INFO | SignerSubCommand | Version = ethsigner/v0.3.0/windows-x86_64/oracle-java-13
Usage:
ethsigner [OPTIONS] [COMMAND]
Description:
This command runs the EthSigner.
Options:
--chain-id=
CengizSirlan (Sun, 10 Nov 2019 00:56:25 GMT):
some examples of what I tried and the results
cjhare (Mon, 11 Nov 2019 05:07:25 GMT):
> What's the use of tagging a non working repos state?
A challenge with altering the git history comes from git being a distributed version control system, not a centralised version control system. As each client mirrors the repo, unless you have access to all client, then you cannot have full confidence of removing the buggy release tag.
https://git-scm.com/book/en/v2/Getting-Started-About-Version-Control
ericglau (Mon, 11 Nov 2019 16:35:18 GMT):
Is there any plan to publish Besu releases on Maven Central? It would be useful for builds.
shemnon (Mon, 11 Nov 2019 16:58:22 GMT):
No active plans. Right now we don't have a stable extrenal API we are willing to stick to, so that would make external builds unstable. Anything with 'internal' in the GAV is subject to sudden and unplanned changes.
shemnon (Mon, 11 Nov 2019 22:00:30 GMT):
We have a Besu Contributor Call today at 0100 UTC (11am Brisbane, 5pm SF, 8pm NY, middle of night Eurooe). Adrian Sutton will be the host. If you have anything to discuss please add it to the agenda: https://wiki.hyperledger.org/display/BESU/2019-11-12+Besu+Contributor+Call
shemnon (Mon, 11 Nov 2019 22:00:30 GMT):
We have a Besu Contributor Call today at 0100 UTC (11am Brisbane, 5pm SF, 8pm NY, middle of night Europe). Adrian Sutton will be the host. If you have anything to discuss please add it to the agenda: https://wiki.hyperledger.org/display/BESU/2019-11-12+Besu+Contributor+Call
matkt (Tue, 12 Nov 2019 21:04:37 GMT):
Has joined the channel.
shemnon (Tue, 12 Nov 2019 21:36:28 GMT):
@mackcom @GregTheGreek How long did the sync to ETC chainhead take? Was it archival or fast sync?
mackcom (Wed, 13 Nov 2019 00:30:01 GMT):
I'm not sure exactly how long since I was syncing and stopping to refactor code, ect. It's 380 GB, it took at least 3 days. It was full archival.
shemnon (Wed, 13 Nov 2019 04:48:47 GMT):
Do the other ETC clients support ETH/63 fast sync?
cjhare (Wed, 13 Nov 2019 06:05:25 GMT):
FYI - The release remediation page has now been created on the Wiki https://wiki.hyperledger.org/display/BESU/Release+Remediation
GregTheGreek (Wed, 13 Nov 2019 14:38:48 GMT):
We have Parity (outright) and a fork of Geth (multi-geth) - so only if they support it upstream?
shemnon (Wed, 13 Nov 2019 15:46:42 GMT):
IIRC parity would have warpsync (which is currently broken on mainnet) and I'm not sure if multi-geth keeps up with warp-sync. However they are also light client RLPx calls so if ETC has mainnet style light clients it would support it too. I'll find out here shortly.
joshuafernandes (Thu, 14 Nov 2019 00:47:29 GMT):
Hello @all
We have been using CircleCI and Jenkins for the last couple of weeks and that has been working well. We are at the next stage where we are going to turn on Circle build checks as we progress to using CircleCI only and will eventually discontinue the use of Jenkins. If you have any questions or concerns please feel free to reach out.
Radhi (Fri, 15 Nov 2019 14:42:15 GMT):
Has joined the channel.
shemnon (Fri, 15 Nov 2019 15:24:17 GMT):
ETC support up to atlantis has been merged, so fire up your nodes and sync `--network=classic` and `--network=kotti`! (dev code only, it won't ship in a release until next week).
shemnon (Fri, 15 Nov 2019 15:24:38 GMT):
thanks to @mackcom and @GregTheGreek for their contribution.
shemnon (Fri, 15 Nov 2019 15:26:43 GMT):
Also, Antoine Tolume's contribution of Stratum support merged this week as well, so you can mine with GPUs off your besu nodes.
shemnon (Fri, 15 Nov 2019 22:56:42 GMT):
So to close the loop on this. It appears all clients have fast sync. "getc" which I presume is Geth Classic, Parity, and Multigeth all advertise eth/63, which is the ethereum subprotocol that supports fast sync.
shemnon (Fri, 15 Nov 2019 22:57:14 GMT):
Kotti synced in about 12 minutes, almost all the time was spent downloading blocks over a residential internet connection.
shemnon (Sat, 16 Nov 2019 04:53:14 GMT):
Classic support has issues with fast sync - https://jira.hyperledger.org/browse/BESU-116
shemnon (Sat, 16 Nov 2019 14:25:25 GMT):
Classic fast synced in 4h19m on my mac pro last night. 28GB database.
GregTheGreek (Sat, 16 Nov 2019 15:11:28 GMT):
incredible!
madelinemurray (Tue, 19 Nov 2019 10:53:28 GMT):
Trying this
sweexordious (Tue, 19 Nov 2019 12:02:35 GMT):
Has joined the channel.
sweexordious (Tue, 19 Nov 2019 12:02:40 GMT):
Hello,
sweexordious (Tue, 19 Nov 2019 12:04:46 GMT):
Ive been trying to to contribute to a documentation issue but i am stuck on these links as i cant find them on the new documentation website:
- https://besu.hyperledger.org/en/latest/Getting-Started/Run-Docker-Image/#custom-genesis-file
- https://besu.hyperledger.org/en/latest/Getting-Started/Run-Docker-Image/#data-directory
- https://besu.hyperledger.org/en/latest/Getting-Started/Run-Docker-Image/#data-directory
- https://besu.hyperledger.org/en/latest/Getting-Started/Run-Docker-Image/#custom-configuration-file
- https://besu.hyperledger.org/en/latest/EthStats/Overview/
- https://besu.hyperledger.org/en/latest/Reference/JSON-RPC-API-Methods/#txpool_pantheontransactions
- https://github.com/PegaSysEng/pantheon/blob/master/PRIVACYROADMAP.MD
Any help or redirection is welcome,
Thank you.
faraggi (Tue, 19 Nov 2019 15:36:26 GMT):
Hi @sweexordious Where did you find those links?
@NicolasMassart @madelinemurray
NicolasMassart (Tue, 19 Nov 2019 20:36:54 GMT):
Did you find these links in a Jira issue?
sweexordious (Tue, 19 Nov 2019 21:12:43 GMT):
I found them on this issue:
https://jira.hyperledger.org/projects/BESU/issues/BESU-110?filter=addedrecently
NicolasMassart (Tue, 19 Nov 2019 21:20:20 GMT):
I don't think I understand your point as this ticket is about a list of broken links in the changelog, so yes they have to be fixed. All links in the list are expected to be broken as this issue is still in the "todo" state. You should not look in this list for working links.
NicolasMassart (Tue, 19 Nov 2019 21:21:20 GMT):
Oh, ok, I understood your issue. Sorry for being a bit slow on this...
sweexordious (Tue, 19 Nov 2019 21:24:12 GMT):
No problem, Thank you for your time
NicolasMassart (Tue, 19 Nov 2019 21:24:47 GMT):
So the first about docker are pointing to content that does not exist anymore in the doc we migrated to HL. Removing the link is probably the best thing to do.
NicolasMassart (Tue, 19 Nov 2019 21:25:45 GMT):
Ethstats link should now point to https://besu.hyperledger.org/HowTo/Deploy/Lite-Network-Monitor/ you can use the internal doc site search engine to find the content that was moved in another place in the structure.
NicolasMassart (Tue, 19 Nov 2019 21:27:08 GMT):
txpool_pantheontransactions method was renamed as it container pantheon in it. Again, you could use the search feature and find that it was replaced by https://besu.hyperledger.org/en/stable/Reference/API-Methods/#txpool_besutransactions
NicolasMassart (Tue, 19 Nov 2019 21:30:02 GMT):
And for the last one, privacy info was merged in the common roadmap https://github.com/hyperledger/besu/blob/master/ROADMAP.md
NicolasMassart (Tue, 19 Nov 2019 21:33:35 GMT):
But to me, all broken links to specific articles or pages before Besu 1.2.1 should be simply pointing to https://besu.hyperledger.org/
Otherwise you will end up with anachronic disaster trying to make old links point to new content. SO just making them point to Besu doc site root is the simple way as people could still go back in the changelog commits history to see a previous version it they want to se the previous link.
NicolasMassart (Tue, 19 Nov 2019 21:34:48 GMT):
Please let me know if you create a PR for that, so I could review it.
sweexordious (Tue, 19 Nov 2019 23:23:53 GMT):
Thank you so much for your help.
I will make a PR as soon as possible.
sweexordious (Wed, 20 Nov 2019 08:24:05 GMT):
Hello,
I have tried making a pull request about these documentation changes but im stuck with the DCO probot of github.
It is PR #203
sweexordious (Wed, 20 Nov 2019 08:24:27 GMT):
The problem is described in the pull comments
NicolasMassart (Wed, 20 Nov 2019 10:13:25 GMT):
I will have a look a it. Will let you know
sweexordious (Wed, 20 Nov 2019 10:15:49 GMT):
Thank you
NicolasMassart (Wed, 20 Nov 2019 13:37:29 GMT):
Hi, looking at you PR, it seems that you work on your fork on the master branch but you did not clean your fork master after the first PR was merged, so you continue pushing commits from your master that contains DCO issues, and you can see that your PR commits list contains far more commits than just the links related ones.
What you should do is create a topic branch as indicated at step 2 of contribution workflow doc in https://github.com/hyperledger/besu/blob/master/CONTRIBUTING.md#contribution-workflow
Then you only have the expected commits in the PR and when we merge the PR you can update your master and get rid of the topic branch.
I'm looking at the best way to fix this situation.
sweexordious (Wed, 20 Nov 2019 15:03:33 GMT):
Exactly yeah, i see commits that aren't mine.
Following your steps, i recloned my repo, created a new topic called BESU-110, switched to it, pushed the results. So, i can see the new branch on my github repo, however when i create a PR from that branch, i still have that DCO problem.
I think what i will do, save the changes i made, delete my fork, fork it again and then, create a new topic, make the changes, push them and then PR ? do you think this will work ?
Sorry for the late replies as i am in school and i dont have permanent access to my computer.
sweexordious (Wed, 20 Nov 2019 15:04:46 GMT):
The newest PR i have made from my newly created branch is PR #204
NicolasMassart (Wed, 20 Nov 2019 15:57:14 GMT):
the issue is that you may have pushed faulty commits to your master branch on your fork. No need to delete your fork (it would break links with some previous PRs, I don't know if you have some, but still).
sweexordious (Wed, 20 Nov 2019 22:00:47 GMT):
I have one previous PR but i guess i can sacrifice that.. I guess ill just have to remake the changes again on a new fork.
NicolasMassart (Wed, 20 Nov 2019 22:01:49 GMT):
Wha you could do,You can just rename you master, pull the fresh one from upstream, push it to your fork and go from there.
NicolasMassart (Wed, 20 Nov 2019 22:02:11 GMT):
then remove the old renamed master
sweexordious (Wed, 20 Nov 2019 22:15:45 GMT):
I will look into that and let you know the results.
Thank you so much for your help.
EdJoJob (Wed, 20 Nov 2019 22:27:30 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
EdJoJob (Wed, 20 Nov 2019 23:21:40 GMT):
It's now safe to merge into master for hyperledger/besu again.
drinkcoffee (Thu, 21 Nov 2019 06:16:44 GMT):
I am writing a JSON RPC API that returns a complex data structure (it is a map of BigIntegers to ints). Is there an existing JSON RPC API that I should follow that does something similar? Is this sort of thing a good idea? Would I be better off trying to return smaller parts of this?
I was going to put in some sort of check to make sure the blob being returned wasn't too big. Is there any other checks you suggest?
NicolasMassart (Thu, 21 Nov 2019 11:42:13 GMT):
How does besu answer to websocket ping as described in https://tools.ietf.org/html/rfc6455#section-5.5.2 ? Because I have an issue with web3j using websocket that really coincides with the connection lost detection (every 60s by default).
My guess, but I did not explore the code enough to tell, is that this ping is handled as a request and of course is not a valid one. Hence having these two logs:
On the besu node side:
```
2019-11-21 11:14:49.089+00:00 | vert.x-eventloop-thread-8 | DEBUG | WebSocket08FrameDecoder | Decoding WebSocket Frame opCode=9
2019-11-21 11:14:49.090+00:00 | vert.x-eventloop-thread-8 | DEBUG | WebSocket08FrameDecoder | Decoding WebSocket Frame length=0
2019-11-21 11:14:49.093+00:00 | vert.x-eventloop-thread-8 | DEBUG | WebSocket08FrameEncoder | Encoding WebSocket Frame opCode=10 length=0
2019-11-21 11:14:49.094+00:00 | vert.x-eventloop-thread-8 | DEBUG | WebSocketService | Received Websocket request (host=172.17.0.1, port=49722)
2019-11-21 11:14:49.114+00:00 | vert.x-worker-thread-6 | DEBUG | WebSocketRequestHandler | Error mapping json to WebSocketRpcRequest
io.vertx.core.json.DecodeException: Failed to decode:No content to map due to end-of-input
at [Source: (io.netty.buffer.ByteBufInputStream); line: 1, column: 0]
at io.vertx.core.json.Json.decodeValue(Json.java:222) ~[vertx-core-3.8.0.jar:3.8.0]
at io.vertx.core.json.JsonObject.fromBuffer(JsonObject.java:975) ~[vertx-core-3.8.0.jar:3.8.0]
at io.vertx.core.json.JsonObject.
NicolasMassart (Thu, 21 Nov 2019 11:43:14 GMT):
and on the java app side:
```
> Task :Main.main()
12:30:11.159 [WebSocketConnectReadThread-15] DEBUG o.w.p.websocket.WebSocketClient - Opened WebSocket connection to ws://localhost:7546
12:31:11.170 [WebSocketConnectReadThread-15] DEBUG o.w.p.websocket.WebSocketClient - Received message {"jsonrpc":"2.0","id":null,"error":{"code":-32600,"message":"Invalid Request"}} from server ws://localhost:7546
12:31:11.196 [WebSocketConnectReadThread-15] ERROR o.w.p.websocket.WebSocketClient - Failed to process message '{"jsonrpc":"2.0","id":null,"error":{"code":-32600,"message":"Invalid Request"}}' from server ws://localhost:7546
java.io.IOException: 'id' expected to be long, but it is: 'null'
at org.web3j.protocol.websocket.WebSocketService.getReplyId(WebSocketService.java:334)
at org.web3j.protocol.websocket.WebSocketService.processRequestReply(WebSocketService.java:206)
at org.web3j.protocol.websocket.WebSocketService.onWebSocketMessage(WebSocketService.java:197)
at org.web3j.protocol.websocket.WebSocketService$1.onMessage(WebSocketService.java:122)
at org.web3j.protocol.websocket.WebSocketClient.lambda$onMessage$0(WebSocketClient.java:52)
at java.base/java.util.Optional.ifPresent(Optional.java:183)
at org.web3j.protocol.websocket.WebSocketClient.onMessage(WebSocketClient.java:49)
at org.java_websocket.client.WebSocketClient.onWebsocketMessage(WebSocketClient.java:462)
at org.java_websocket.drafts.Draft_6455.processFrame(Draft_6455.java:696)
at org.java_websocket.WebSocketImpl.decodeFrames(WebSocketImpl.java:367)
at org.java_websocket.WebSocketImpl.decode(WebSocketImpl.java:212)
at org.java_websocket.client.WebSocketClient.run(WebSocketClient.java:389)
at java.base/java.lang.Thread.run(Thread.java:835)
```
because nodes response includes a null id.
NicolasMassart (Thu, 21 Nov 2019 11:45:12 GMT):
a very simple java code to test (given that you have imported web3j as a dependency, see http://docs.web3j.io/getting_started/#gradle) is:
```
import java.net.ConnectException;
import org.web3j.protocol.Web3j;
import org.web3j.protocol.websocket.WebSocketService;
public class Main {
public static void main(String[] args) {
final WebSocketService wss = new WebSocketService("ws://localhost:7546", true);
try {
wss.connect();
} catch (final ConnectException e) {
System.out.println("ConnectException: " + e.getMessage());
}
Web3j.build(wss);
}
}
```
NicolasMassart (Thu, 21 Nov 2019 11:45:49 GMT):
Replace the endpoint by yours in the code of course.
NicolasMassart (Thu, 21 Nov 2019 11:46:10 GMT):
SO if anyone could confirm the way Besu handles ping requests, thanks.
NicolasMassart (Thu, 21 Nov 2019 11:46:10 GMT):
So if anyone could confirm the way Besu handles ping requests, tests were made with hyperledger/besu:1.3.4 docker image. Thanks.
NicolasMassart (Thu, 21 Nov 2019 14:02:35 GMT):
For what I read in Besu code, the websocket handler passes the buffer directly and tries to unserialize it as JSON, which fails in case of a ping frame. Also the debug log shows a line `DEBUG | WebSocketService | Received Websocket request (host=172.17.0.1, port=49722)`, notice the double space where the buffer.toString() should be displayed after the word "request", which matches the idea that ping frame containing an opcode 0x9 which is not a printable caracter.
NicolasMassart (Thu, 21 Nov 2019 14:04:38 GMT):
For the debug log, see https://github.com/hyperledger/besu/blob/master/ethereum/api/src/main/java/org/hyperledger/besu/ethereum/api/jsonrpc/websocket/WebSocketService.java#L119
shemnon (Thu, 21 Nov 2019 14:29:49 GMT):
We let vertx handle our websocket requests. My guess w/o looking at code is we are shunting all websocket work down the "this is a request" pipe and should be special casing pings and pongs.
NicolasMassart (Thu, 21 Nov 2019 14:30:42 GMT):
Is that something we can do in Besu? Do you want me to create a Jira ticket?
shemnon (Thu, 21 Nov 2019 14:31:15 GMT):
Make a JI, I think we can handle it, jsut need to examine the vertx apis. There are several vertx bugs related to it so I think it's totally doable.
NicolasMassart (Thu, 21 Nov 2019 14:31:32 GMT):
Great thanks.
shemnon (Thu, 21 Nov 2019 14:32:02 GMT):
I think we already have a JI howeber...
shemnon (Thu, 21 Nov 2019 14:33:24 GMT):
It's in the pre-migration DB, we don't have a BESU- JI
shemnon (Thu, 21 Nov 2019 14:34:51 GMT):
PAN-3279 - we still got people going to the old DB so this one will need to be migrated by hand. PAN- is on the "to decommission" list.
NicolasMassart (Thu, 21 Nov 2019 14:35:09 GMT):
Ok, I can create it based in the old DB then.
NicolasMassart (Thu, 21 Nov 2019 14:36:17 GMT):
But more than just being about web3j only it's basically the websocket RFC that asks for handling ping/pong
shemnon (Thu, 21 Nov 2019 14:36:38 GMT):
Yep, web3j correctly implements the RFC and is a good test example.
shemnon (Thu, 21 Nov 2019 14:37:01 GMT):
The bug is at the websockets level, web3j is the means to reproduce.
NicolasMassart (Thu, 21 Nov 2019 14:37:12 GMT):
Thanks
NicolasMassart (Thu, 21 Nov 2019 15:28:31 GMT):
I created https://jira.hyperledger.org/browse/BESU-128 let me know if you need me to do something else.
lucassaldanha (Thu, 21 Nov 2019 20:39:12 GMT):
Just a quick thing. Vertx WS has got a specific handler for control frames (ping, pong, etc). It is the `websocket.frameHandler`.
I don't know why this isn't handled by Vertx automatically, but if we want to do it in Besu it should be easy.
shemnon (Thu, 21 Nov 2019 20:39:36 GMT):
I'll hand it off to you then.
shemnon (Thu, 21 Nov 2019 20:39:56 GMT):
From what I've seen we get the ping, start handling the poing, then get sent into our handler code when it should be skipped.
lucassaldanha (Thu, 21 Nov 2019 20:39:57 GMT):
Me and my big mouth...
lucassaldanha (Thu, 21 Nov 2019 20:39:58 GMT):
hahaha
lucassaldanha (Thu, 21 Nov 2019 20:40:15 GMT):
I'll take a look!
shemnon (Thu, 21 Nov 2019 20:40:24 GMT):
I tried chanign to `handleBinaryMessage` but that fails tests
shemnon (Thu, 21 Nov 2019 20:40:24 GMT):
I tried chaning to `handleBinaryMessage` but that fails tests
shemnon (Thu, 21 Nov 2019 20:41:28 GMT):
WebSocketService line 116
cjhare (Thu, 21 Nov 2019 22:55:19 GMT):
`org.hyperledger.besu.ethereum.api.jsonrpc.internal.results.TransactionReceiptResult` is a return JSON object containing a nested object (another JSON object).
The idea with complex datastructures with jackson annotations are they become an daisy-chain of annotated classes, when encoding/decoding occurs then Jackson takes care of the rest
shemnon (Fri, 22 Nov 2019 01:14:23 GMT):
`Map
drinkcoffee (Fri, 22 Nov 2019 01:31:25 GMT):
Great. Thanks
NicolasMassart (Fri, 22 Nov 2019 10:23:30 GMT):
I did not forget you, I'm trying to find someone who could help you.
sweexordious (Fri, 22 Nov 2019 17:26:48 GMT):
Hello,
I managed to make it work as you pointed out.
Thank you so much for your help.
NicolasMassart (Mon, 25 Nov 2019 08:27:22 GMT):
test
madelinemurray (Mon, 25 Nov 2019 08:30:08 GMT):
Thanks for letting us know
NicolasMassart (Mon, 25 Nov 2019 08:30:29 GMT):
Thanks
shemnon (Mon, 25 Nov 2019 15:43:36 GMT):
Tomorrow (Tuesday EMEA/AMER) we have a Besu contributor/community call in EMEA/AMER friendly times - https://wiki.hyperledger.org/display/BESU/2019-11-26+Besu+Contributor+Call
shemnon (Tue, 26 Nov 2019 02:28:34 GMT):
I've proposed to add Edward Mack as a maintainer for Besu - https://github.com/hyperledger/besu/pull/219 - All maintainer please read and comment.
faraggi (Tue, 26 Nov 2019 14:50:34 GMT):
Reminder that the bi-weekly Besu Contributor call is going to be held in 10 minutes: https://consensys.zoom.us/j/725448960
Agenda can be found here: https://wiki.hyperledger.org/display/BESU/2019-11-26+Besu+Contributor+Call
shemnon (Wed, 27 Nov 2019 21:39:06 GMT):
I'd like to propsoe increasing the 1.4 RC window by two to four weeks "to the left" to accomidates Antoine's changes to have Besu start using Tuweni. Here's a wiki doc for the proposal - https://wiki.hyperledger.org/pages/viewpage.action?pageId=24777394
sweexordious (Wed, 27 Nov 2019 22:42:19 GMT):
Hello,
I have been playing with the json-rpc methods and i noticed a field ID in every method call. it means what ? when i delete it i get no answer, and when i change its value to arbitrary values, i still get the results with that ID i send.
Example:
curl -X POST --data '{"jsonrpc":"2.","method":"eth_mining","params":[], "id": "something"}' http://127.0.0.1:8545
I get the following response:
{
"jsonrpc" : "2.0",
"id" : "something",
"result" : true
}
shemnon (Wed, 27 Nov 2019 22:47:06 GMT):
See https://www.jsonrpc.org/specification#notification
shemnon (Wed, 27 Nov 2019 22:47:06 GMT):
The ID is needed in multiplexed situations. Over WebSockets or if you pass in an array of RPCs instead of a single RPC.
shemnon (Wed, 27 Nov 2019 22:47:06 GMT):
https://www.jsonrpc.org/specification#notification
sweexordious (Wed, 27 Nov 2019 22:48:22 GMT):
Thank you
shemnon (Wed, 27 Nov 2019 22:50:01 GMT):
The ID is needed in multiplexed situations. Over WebSockets or if you pass in an array of RPCs instead of a single RPC.
sweexordious (Wed, 27 Nov 2019 22:50:41 GMT):
I guess i need to go over json-rpc a little bit more
shemnon (Wed, 27 Nov 2019 22:51:59 GMT):
This is just a deep corner case that most devs never need to worry about.
ajsutton (Thu, 28 Nov 2019 22:41:48 GMT):
FYI: You can get email notifications of Besu wiki changes by going to the wiki (https://wiki.hyperledger.org/display/BESU/Hyperledger+Besu) and in the top right hand corner click on "Watching" and select "Watch all content in this space". There's not a lot of activity so shouldn't overwhelm you but will make it a lot easier to follow proposals and discussions there. You can just watch individual pages as well but I'd recommend at least trying watching the whole space first.
cjhare (Fri, 29 Nov 2019 01:57:46 GMT):
FYI there is now an initial Defect Prioritisation Policy on the Wiki: https://wiki.hyperledger.org/display/BESU/Defect+Prioritisation+Policy
Initially it'll be trialled by the PegaSys POs when triaging (prioristing new issues raised) and will be reviewed/iterated on as needed ...but everyone is also feel free to triage bugs raise in JIRA against Besu and suggest improvements/alterations :slight_smile:
EdJoJob (Sun, 01 Dec 2019 22:59:46 GMT):
With having needed to help some people through fixing their repos post a DCO issue on the master, I have added some instructions for the announcement and fix to local repos to https://wiki.hyperledger.org/pages/viewpage.action?pageId=24772914
EdJoJob (Sun, 01 Dec 2019 23:00:05 GMT):
(Fixing DCO on master/release)
sweexordious (Mon, 02 Dec 2019 08:33:28 GMT):
Hello,
I am trying with the BESU-54 https://jira.hyperledger.org/browse/BESU-54 . It's about checking number of parameters before calling the json-rpc method.
So far, i found where the code of each function is. Now im wondering, whether i will have to add to all of the methods code a parameters number check or there is a better way to do it ?
Thank you
sloops77 (Mon, 02 Dec 2019 12:58:20 GMT):
Has joined the channel.
cjhare (Mon, 02 Dec 2019 23:34:41 GMT):
...it depends :thinking:
cjhare (Mon, 02 Dec 2019 23:34:41 GMT):
...it depends :thinking:
I've added two clarifying questions on the JI, the responses would impact the recommendation (as there are a few to satisfy BESU-54)
sweexordious (Mon, 02 Dec 2019 23:44:10 GMT):
Thank you for your help :D
shemnon (Tue, 03 Dec 2019 16:49:04 GMT):
CircleCI question - One of my PRs is failing with memory issues (integerationTEsts, exit 137) - Is it OK to change the CI to make integration tests an XL container?
rjones (Tue, 03 Dec 2019 19:02:36 GMT):
@shemnon I have no issue with it
rjones (Tue, 03 Dec 2019 19:03:13 GMT):
@shemnon is that something I need to do, or is it something you can do in the codebase?
shemnon (Tue, 03 Dec 2019 19:27:24 GMT):
It's in my current PR, circleci config is part of the code repo.
EdJoJob (Wed, 04 Dec 2019 22:11:02 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
EdJoJob (Wed, 04 Dec 2019 23:08:15 GMT):
It's now safe to merge into master for hyperledger/besu again.
madelinemurray (Thu, 05 Dec 2019 04:40:49 GMT):
Besu v1.3.6 has been released. Download from https://pegasys.tech/solutions/hyperledger-besu/#downloads
Changelog https://github.com/hyperledger/besu/blob/master/CHANGELOG.md
NicolasMassart (Thu, 05 Dec 2019 11:08:00 GMT):
Message from @carlosfaria in #besu https://chat.hyperledger.org/channel/besu?msg=7gWzr2oJm2pjWQKfu
> Hey guys, How can I use the IbftProtocolSchedule to leverage a change in blockPeriod on a certain block number, defined on genesis forks?
NicolasMassart (Thu, 05 Dec 2019 11:08:00 GMT):
Message from @carlosfaria in #besu https://chat.hyperledger.org/channel/besu?msg=7gWzr2oJm2pjWQKfu
NicolasMassart (Thu, 05 Dec 2019 11:08:00 GMT):
Message from @carlosfaria in #besu https://chat.hyperledger.org/channel/besu?msg=7gWzr2oJm2pjWQKfu that mey be something contributing devs can answer.
NicolasMassart (Thu, 05 Dec 2019 11:08:00 GMT):
Message from @carlosfaria in #besu https://chat.hyperledger.org/channel/besu?msg=7gWzr2oJm2pjWQKfu that may be something contributing devs here can answer.
shemnon (Thu, 05 Dec 2019 18:05:50 GMT):
Design review - Adding RPCEndpointService to the plugins framework. Please comment -https://wiki.hyperledger.org/display/BESU/RPC+Endpoint+Service
RealDeanZhao (Fri, 06 Dec 2019 05:16:24 GMT):
Has joined the channel.
drinkcoffee (Fri, 06 Dec 2019 05:20:36 GMT):
Ommer questions:
- Who gets the transaction cost reward (gas etc) for the Ommer? The original miner, the new miner, or no one?
- If a Ommer is included in a block, and the Ommer includes a transaction that was valid, but is no longer valid (maybe there is a state update due to another transaction included in the heaviest chain ) is there any transaction cost?
- The block rewards related to Ommers gets smaller as the included Ommer gets older (that is Ommer reward for a block that shares a common Grand Parent gets more than for a Great-Great-Grand Parent). Are the transaction cost rewards similarly affected?
shemnon (Fri, 06 Dec 2019 05:28:24 GMT):
- The miner of the ommer (not the miner of the block with the ommer references
- Transactions in ommers don't matter. They can be totally invalid. Only the header matters, and only the ommer reward is processed.
- Transactions in ommers don't generate rewards.
shemnon (Fri, 06 Dec 2019 05:28:24 GMT):
- The miner of the ommer (not the miner of the block with the ommer references)
- Transactions in ommers don't matter. They can be totally invalid. Only the header matters, and only the ommer reward is processed.
- Transactions in ommers don't generate rewards.
shemnon (Fri, 06 Dec 2019 05:28:24 GMT):
- The miner of the ommer (not the miner of the block with the ommer references)
- Transactions in ommers don't matter. They can be totally invalid. Only the header matters, and only the ommer reward is processed.
- Transactions in ommers don't get processed and don't pay gas fees.
ajsutton (Fri, 06 Dec 2019 07:32:35 GMT):
As Danno said no transaction fees are paid for ommers. The transaction fees charged when/if the ommer was imported as a non-canonical block have no effect on the canonical world state (because it's off on a non-canonical fork, not the canonical chain so it's exactly as if those transactions never happened).
The only thing ommers affect is the block reward payments: The original miner gets a bonus (of blockReward / 32) for each ommer they include and the ommer miner gets a small payment (blockReward * (block.number - ommer.number) / 8) when the block is included as an ommer.
You may find https://www.symphonious.net/2019/11/01/exploring-ethereum-ommers-vs-non-canoncial-blocks/ and https://www.symphonious.net/2019/11/27/exploring-ethereum-what-happens-to-transactions-in-non-canonical-blocks/ useful.
drinkcoffee (Fri, 06 Dec 2019 10:36:42 GMT):
So, the Ommer reward is just for rewarding the other miner for doing work, but the submitter of the transactions will have the transactions put back into the transaction pool, because they will effectivey not have been mined.
drinkcoffee (Fri, 06 Dec 2019 10:36:42 GMT):
So, the Ommer reward is just for rewarding the other miner for doing work, and rewarding the successful block miner for including the block, but the submitter of the transactions will have the transactions put back into the transaction pool, because they will effectivey not have been mined.
carlosfaria (Fri, 06 Dec 2019 12:16:28 GMT):
Hey guys, How can I use the `IbftProtocolSchedule` to leverage a change in `blockPeriod` on a certain block number, defined on genesis forks?
Sam_Devs (Fri, 06 Dec 2019 14:47:01 GMT):
Has joined the channel.
drinkcoffee (Fri, 06 Dec 2019 23:05:00 GMT):
@ajsutton , thanks for the very informative artcles. Very helpful!
drinkcoffee (Sat, 07 Dec 2019 02:18:54 GMT):
@ajsutton Looking at equation 163 of the yellow paper, it looks like the equation for Ommer reward is: (1 + 1/8 * (Ui - BHi)) * Rblock; which is: (1 + 1/8 * (Ommer Block Number - Canonical Block Number)* Block Reward. For example, looking at this block (https://etherscan.io/uncle/0x88a638f806782a6cd3d7e9960beb27aad607c7acaec34a96a8c3f8ee71cd7b27), the Uncle / Ommer block number was 9058659, the Canonical block number was 9058660, the block reward is 2 Ether, so the Ommer reward is: (1 + 1/8*(9058659 - 9058660)) * 2 = 1.75 Ether
ajsutton (Sun, 08 Dec 2019 20:11:46 GMT):
Yes I misread the code, it should definitely be less than one blockReward. The code in besu is:
```
final long distance = header.getNumber() - ommerHeader.getNumber();
final Wei ommerReward = blockReward.minus(blockReward.times(distance).dividedBy(8));
ommerCoinbase.incrementBalance(ommerReward);
```
bretthenderson (Tue, 10 Dec 2019 01:00:39 GMT):
All, the Besu Contributor call is starting. - https://consensys.zoom.us/j/199741148
madelinemurray (Wed, 11 Dec 2019 09:14:25 GMT):
Added the process for updating the changlog and what should be included - https://wiki.hyperledger.org/display/BESU/Changelog
This is a change in process for maintainers - the new process means updating the CHANGELOG.md file as part of development for PRs to be included in the changelog. Let me know if anything isn't clear and it would be great if reviewers could check this is being done while we get used to the new process.
cjhare (Wed, 11 Dec 2019 23:43:38 GMT):
For clarity, must the update to the CHANGELOG.md be part of the contributors PR, or is a seperate PR acceptable?
ajsutton (Thu, 12 Dec 2019 03:40:26 GMT):
It can be a separate PR and not all PRs need a changelog entry.
rjones (Thu, 12 Dec 2019 13:21:27 GMT):
I know Fabric has a script to generate a list automagically which can be pruned.
ajsutton (Thu, 12 Dec 2019 21:27:47 GMT):
We now have three upcoming or recent upgrades noted in our changelog (Istanbul, Muir Glacier and Agharta). I'm wondering if we should just have a table showing forks and minimum releases at the top of the changelog. Still only list the most recent upgrade and any upcoming ones but as a permanent part of the changelog format.
EdJoJob (Thu, 12 Dec 2019 21:45:39 GMT):
Putting myself in the shoes of someone running Besu on a public network, I would like this idea.
shemnon (Thu, 12 Dec 2019 22:19:11 GMT):
ooohh.... I like that alot. There are two categories: mainnet and classic, and 7 networks for those categories (mainnet, ropsten, rinkeby, goerli, classic, mordor, and kotti) But three different levels of support for each persistent network: unsupported, supported (code exists), and scheduled (code is turned on at a block number). And then, the question of what is the order of the blocks?
shemnon (Thu, 12 Dec 2019 22:19:46 GMT):
And on the wiki page perhaps a complete table?
ajsutton (Thu, 12 Dec 2019 23:07:33 GMT):
I feel like you've thought about this more than I have already. :) I was just going to have
Istanbul | 1.x.x
Muir Glacier | 1.x.x
Agharta | 1.x.x
But no objections to having more information if it's useful.
shemnon (Thu, 12 Dec 2019 23:12:24 GMT):
But since testnet block numbers roll out later than the impls there is a slow trickle of updates.
ajsutton (Thu, 12 Dec 2019 23:14:11 GMT):
True. It probably does make sense to have more info. I was just going based off of the notices we currently have in the changelog (but it's at a nice simple point where testnets and mainnet all have blocks assigned).
shubhamv 3 (Fri, 13 Dec 2019 11:17:44 GMT):
Has joined the channel.
shemnon (Sat, 14 Dec 2019 00:35:46 GMT):
Due to a mainnet consensus issue we will be pushing the 1.3.7 release shortly.
shemnon (Sat, 14 Dec 2019 00:35:50 GMT):
I propose we do a 1.3.8 release in the latter half of the week.
EdJoJob (Sat, 14 Dec 2019 00:36:24 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
EdJoJob (Sat, 14 Dec 2019 01:21:40 GMT):
It's now safe to merge into master for hyperledger/besu again.
juancc (Sun, 15 Dec 2019 18:23:23 GMT):
Has joined the channel.
timbeiko (Mon, 16 Dec 2019 19:27:55 GMT):
IMPORTANT: Besu experienced a mainnet consensus issue on Friday. 1.3.7 was promptly released to address it. More information about the issue and mitigation for both public and private networks is now available at: https://wiki.hyperledger.org/display/BESU/Mainnet+Consensus+Bug+Identified+and+Resolved+in+Hyperledger+Besu
timbeiko (Mon, 16 Dec 2019 19:27:55 GMT):
*IMPORTANT*: Besu experienced a mainnet consensus issue on Friday. 1.3.7 was promptly released to address it. More information about the issue and mitigation for both public and private networks is now available at: https://wiki.hyperledger.org/display/BESU/Mainnet+Consensus+Bug+Identified+and+Resolved+in+Hyperledger+Besu
atoulme (Mon, 16 Dec 2019 21:23:44 GMT):
Has joined the channel.
atoulme (Mon, 16 Dec 2019 23:21:09 GMT):
I finally made it to this chat!
atoulme (Mon, 16 Dec 2019 23:21:32 GMT):
Let me know when to revamp 215 with the latest again
shemnon (Mon, 16 Dec 2019 23:25:40 GMT):
Once I get my RPCs for TX logs indexing I think we will cut 1.3.8 and then 1.4 beta will open on master.
hyperledger-bot (Wed, 18 Dec 2019 23:55:10 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/4026`
shemnon (Thu, 19 Dec 2019 00:50:19 GMT):
Looks to be a flakey tests. Re-running the workflow
hyperledger-bot (Thu, 19 Dec 2019 00:51:25 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/4037`
shemnon (Thu, 19 Dec 2019 02:15:16 GMT):
It was a flakey test. All good now.
shemnon (Thu, 19 Dec 2019 19:27:38 GMT):
So where were we suppoed to put our changelog notes again?
madelinemurray (Thu, 19 Dec 2019 19:53:44 GMT):
Is this for the generate logs CLI option and RPC?
madelinemurray (Thu, 19 Dec 2019 19:54:21 GMT):
If so, I did them yesterday and they're in the changelog PR
shemnon (Thu, 19 Dec 2019 20:10:37 GMT):
yes, thanks
EdJoJob (Thu, 19 Dec 2019 22:16:25 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
joshuafernandes (Thu, 19 Dec 2019 22:17:03 GMT):
@all ^
joshuafernandes (Thu, 19 Dec 2019 22:17:27 GMT):
EdJoJob
joshuafernandes (Thu, 19 Dec 2019 22:17:39 GMT):
bretthenderson
EdJoJob (Thu, 19 Dec 2019 23:33:28 GMT):
@all - you can merge to master on hyperledger/besu again
atoulme (Thu, 19 Dec 2019 23:33:38 GMT):
yay
atoulme (Thu, 19 Dec 2019 23:53:13 GMT):
updating my PR with the latest.
shemnon (Fri, 20 Dec 2019 00:34:11 GMT):
so per our release plan the master branch is now ready for 1.4. beta work. IIRC the plan is if we need 1.3.9 we will do one-off patches off of the 1.3.8 release. If we don't need it we won't.
ajsutton (Fri, 20 Dec 2019 00:46:03 GMT):
That's also my understanding. So we should get the giant bytes library PR merged as soon as it's ready now.
atoulme (Fri, 20 Dec 2019 00:57:48 GMT):
just merged with latest
atoulme (Fri, 20 Dec 2019 00:58:12 GMT):
I'm relying on circleci to tell me if I missed a spot fwiw
atoulme (Fri, 20 Dec 2019 01:13:53 GMT):
@ajsutton https://github.com/hyperledger/besu/pull/215
atoulme (Fri, 20 Dec 2019 01:13:55 GMT):
ready for you
atoulme (Fri, 20 Dec 2019 01:26:23 GMT):
all checks have passed. Please squash?
ajsutton (Fri, 20 Dec 2019 02:05:21 GMT):
Merged. WCPGW?
atoulme (Fri, 20 Dec 2019 02:06:36 GMT):
Let's find out I guess :D
atoulme (Fri, 20 Dec 2019 05:10:01 GMT):
ok, as discussed in the PR that was just merged, there is some more cleanup we can do.
atoulme (Fri, 20 Dec 2019 05:10:04 GMT):
Here is one: https://github.com/hyperledger/besu/pull/274
atoulme (Fri, 20 Dec 2019 17:38:20 GMT):
@shemnon would you please review https://github.com/hyperledger/besu/pull/275 ?
shemnon (Fri, 20 Dec 2019 17:38:47 GMT):
I'll add it to my before lunch queue.
atoulme (Fri, 20 Dec 2019 17:39:00 GMT):
While tests don't pass (reference tests throwing the towel at the point of git clone for some reason :/) I think it does the job of making things simpler.
atoulme (Fri, 20 Dec 2019 17:39:02 GMT):
Thanks!
shemnon (Fri, 20 Dec 2019 18:06:56 GMT):
the .gitsubmodule PR from a couple of weeks ago isn't working as hoped.
shemnon (Fri, 20 Dec 2019 18:07:07 GMT):
I pushed the new refspec.
atoulme (Fri, 20 Dec 2019 18:07:08 GMT):
no it's really not.
atoulme (Fri, 20 Dec 2019 18:07:16 GMT):
sorry about that.
shemnon (Fri, 20 Dec 2019 18:10:04 GMT):
It's a general problem. It's either deal with the occasional issue twice a year when we update reference tests or copy the whole test suite into the code base. Neither is ideal.
atoulme (Fri, 20 Dec 2019 18:13:09 GMT):
maybe I can help there.
atoulme (Fri, 20 Dec 2019 18:13:34 GMT):
For Tuweni I had to move the test files I cared about into the repository. I was given a lot of trouble by some software the ASF uses to check content.
atoulme (Fri, 20 Dec 2019 18:14:01 GMT):
So what I can do is package the reference tests into a jar, and distribute it as part of Apache Tuweni.
atoulme (Fri, 20 Dec 2019 18:14:33 GMT):
Would that be useful? Obviously that means Apache Tuweni needs to keep up with reference tests, which is never ideal.
shemnon (Fri, 20 Dec 2019 18:19:13 GMT):
My concern with that approach is we can't see the reference tests easily in the IDE.
atoulme (Fri, 20 Dec 2019 18:54:02 GMT):
to go back to the PR - it all made sense right?
atoulme (Fri, 20 Dec 2019 18:54:22 GMT):
and with that change, I think we now have a stable code base again without too many compromises.
shemnon (Fri, 20 Dec 2019 18:58:27 GMT):
Yea, I approved it. I think I unwedged it and it's building now.
shemnon (Fri, 20 Dec 2019 19:26:18 GMT):
And... it's merged.
atoulme (Fri, 20 Dec 2019 23:14:18 GMT):
thank you!
atoulme (Fri, 20 Dec 2019 23:14:25 GMT):
feeling like doing one more? https://github.com/hyperledger/besu/pull/277
atoulme (Sat, 21 Dec 2019 06:05:22 GMT):
I got a few more open, mostly deleting stuff.
shemnon (Sat, 21 Dec 2019 15:05:20 GMT):
Since it's a holiday week we may not get to review these in our usual SLA.
adityasingh177 (Mon, 23 Dec 2019 15:31:06 GMT):
Has joined the channel.
shemnon (Mon, 23 Dec 2019 16:02:26 GMT):
@atoulme pleas run a `git submodule update` - all your reference test modules are submitting with reverts back to the old reference test.
shemnon (Mon, 23 Dec 2019 16:02:26 GMT):
@atoulme please run a `git submodule update --init --recursive` - all your reference test modules are submitting with reverts back to the old reference test.
atoulme (Mon, 23 Dec 2019 17:20:45 GMT):
ok
atoulme (Mon, 23 Dec 2019 17:22:44 GMT):
that doesn't do anything. No bits marked dirty :/
atoulme (Mon, 23 Dec 2019 17:23:41 GMT):
I tried deleting the folder and run the command - same outcome.
shemnon (Mon, 23 Dec 2019 17:50:46 GMT):
one other way to reduce this is to make sure that the `ethereum/referencetests/src/test/resources` directory is not in the change set unless you are explicitly intending to change it.
shemnon (Mon, 23 Dec 2019 17:51:35 GMT):
IntelliJ does a good job keeping this out, not sure if VS code surfaces that you are updating that submodule.
atoulme (Mon, 23 Dec 2019 18:46:45 GMT):
that might be a good way out for the existing PRs - I'll rebase them and edit the last commit to remove the change to referencetests.
atoulme (Mon, 23 Dec 2019 18:49:13 GMT):
oh you merged them. Thank you!
StefanPingel (Fri, 27 Dec 2019 02:41:27 GMT):
Has joined the channel.
faraggi (Mon, 06 Jan 2020 14:53:29 GMT):
*PSA: Reminder that tomorrow is the EMEA/AMER Contributor call*
https://wiki.hyperledger.org/display/BESU/2020-01-07+Besu+Contributor+Call
:mega:
shemnon (Mon, 06 Jan 2020 15:45:02 GMT):
Review request - fix (most) Gradle 7 compatibility issues - https://github.com/hyperledger/besu/pull/288
atoulme (Mon, 06 Jan 2020 17:58:22 GMT):
reviewed
faraggi (Tue, 07 Jan 2020 15:00:45 GMT):
Contributor call starting now.
Agenda:
https://wiki.hyperledger.org/display/BESU/2020-01-07+Besu+Contributor+Call
Zoom:
https://consensys.zoom.us/j/725448960
hyperledger-bot (Tue, 07 Jan 2020 16:14:50 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/5093`
hyperledger-bot (Tue, 07 Jan 2020 16:19:25 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/5099`
shemnon (Wed, 08 Jan 2020 02:31:11 GMT):
For review - add Tuweni bytes to Plugin APIs - https://github.com/hyperledger/besu/pull/295
matkt (Wed, 08 Jan 2020 13:23:42 GMT):
Hello ! I published a PR but the pipeline remained blocked and does not launch the tests. Do you have any idea where this may come from? https://github.com/hyperledger/besu/pull/294
faraggi (Wed, 08 Jan 2020 13:35:45 GMT):
That's quite a big PR you've made there, @matkt !
I'm not sure why tests aren't running. @joshuafernandes @shemnon ? @abdelhamidbakhta
matkt (Wed, 08 Jan 2020 20:19:52 GMT):
Hi! I created a new PR to test and the new one does not launch the tests either. Could this not be linked to a permission problem. Will the circleci not only work with a PR created by a maintainer or something like that?
matkt (Wed, 08 Jan 2020 20:19:52 GMT):
Hi! I created a new PR to test and the new one does not launch the tests either. Could this not be linked to a permission problem? Will the circleci not only work with a PR created by a maintainer or something like that?
hyperledger-bot (Thu, 09 Jan 2020 09:54:56 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/5398`
faraggi (Thu, 09 Jan 2020 09:59:54 GMT):
@matkt the issue is being looked at, we'll keep you posted
matkt (Thu, 09 Jan 2020 20:55:37 GMT):
I don't know why but the circleci seems to work. Maybe an issue related to my email. I will verify during my next PR if it's always ok. Thanks
VipinB (Tue, 14 Jan 2020 18:07:37 GMT):
We are creating a proposal in the Capital Markets SIG (CMSIG) to workshop a simple Token using TTF (Token Taxonomy Framework) towards implementation in Besu. This is at the suggestion of Mani Pillai CEO of swapshub. It would be great to have some participation from the Besu community.
The TTF has active membership from Accenture, IBM and others who are prominent in Hyperledger.
The CMSIG is the ideal venue for this, since we have a mix of members, both business and technical -we may be able to attract more since we are completely open. The actual effort can be housed in Hyperledger labs.
Would be good to know whether there are any comparable initiatives out there.
We can rapidly bootstrap a basic token definition and make plans for a BESU implementation.
Our meeting is tomorrow (Jan 15) at 10 am EST (15 UTC) join us at https://zoom.us/my/hyperledger.community.backup The agenda is here
https://wiki.hyperledger.org/display/CMSIG/2020-01-15
reddy (Tue, 14 Jan 2020 23:41:38 GMT):
Has joined the channel.
reddy (Tue, 14 Jan 2020 23:47:16 GMT):
Has left the channel.
reddy (Wed, 15 Jan 2020 01:06:21 GMT):
Has joined the channel.
hyperledger-bot (Wed, 15 Jan 2020 19:53:09 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/5732`
EdJoJob (Wed, 15 Jan 2020 23:09:09 GMT):
@all - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
EdJoJob (Wed, 15 Jan 2020 23:44:39 GMT):
@all - We have completed the 1.4.0-beta1 release for hyperledger/besu
rjones (Thu, 16 Jan 2020 18:36:35 GMT):
WRT https://pegasys.tech/introducing-plugin-apis-in-hyperlegder-besu/: `Hyperledger` not `Hyperlegder`
bretthenderson (Thu, 16 Jan 2020 21:05:02 GMT):
Hi All, quick reminder next week is the Contributor Call. If you have any topcis for dicussion please update the agenda
https://wiki.hyperledger.org/display/BESU/2020-01-21+Contributor+Call
sweexordious (Fri, 17 Jan 2020 15:36:50 GMT):
Hello everyone,
I want start first by wishing you and the teams a happy new year. I hope you enjoyed your holidays and spent quality time with your families and loved ones.
I am writing to ask for help about besu issues.. I tried my best to understand the code etc, but i still don't know how to contribute. The issues are still hard for me to work on.
Also, i dont have a great coding background to start dealing with all the libraries used and understand how they interact with each other and be able to solve issues..
So, if there is a way to fast forward me, guidelines, books, technologies or anything that can help me find a path so that, one day, i can be able to contribute to the real coding issues and not just documentation PRs, i will be more than grateful.
Thank you so much for your time.
faraggi (Fri, 17 Jan 2020 16:20:12 GMT):
Hi @sweexordious .
First of all, thanks for _wanting_ to contribute. We welcome all types of help with open arms!
A good place to start is in our 'Getting started Issues': https://jira.hyperledger.org/browse/BESU-69?jql=project%20%3D%20BESU%20AND%20resolution%20%3D%20Unresolved%20AND%20labels%20in%20(good-first-issue%2C%20help-wanted)%20ORDER%20BY%20priority%20DESC%2C%20updated%20DESC
You can choose any issue from there and comment on it to signal you want to work on it.
Feel free to ask here or in #besu for help understanding the issue or resolving it. (don't hesitate to tag me in the message as well)
faraggi (Fri, 17 Jan 2020 16:21:41 GMT):
You can also participate in the contributor calls in which (if we have time left) will go over contributor questions about specific issues.
sweexordious (Fri, 17 Jan 2020 17:58:35 GMT):
Thank you so much for your answer.
In fact, i tried couple of 'good first issue' and i made the PRs. However, i wanna go deeper if possible, and have better understanding of the code base..
faraggi (Mon, 20 Jan 2020 11:14:26 GMT):
Lets have a quick call and I'll guide you further. I
faraggi (Mon, 20 Jan 2020 11:14:26 GMT):
Lets have a quick call and I'll guide you further. I'll PM you
faraggi (Mon, 20 Jan 2020 13:39:03 GMT):
Thank you to our lastest open contributors!
@SteveMieskoski @matkt
And of course, thanks to those that contribute daily to the codebase and to this Rocket chat!
faraggi (Mon, 20 Jan 2020 13:39:03 GMT):
Thank you to our lastest open contributors!
@SteveMieskoski @matkt @marlow
And of course, thanks to those that contribute daily to the codebase and to this Rocket chat!
faraggi (Mon, 20 Jan 2020 13:39:08 GMT):
:thanks:
EdJoJob (Mon, 20 Jan 2020 23:57:14 GMT):
@all - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
EdJoJob (Tue, 21 Jan 2020 00:48:50 GMT):
@all - It's now safe to merge into master for hyperledger/besu again.
sweexordious (Tue, 21 Jan 2020 08:38:44 GMT):
Thank you so much
raedbensaid (Tue, 21 Jan 2020 13:14:19 GMT):
Has joined the channel.
EdJoJob (Wed, 22 Jan 2020 00:17:14 GMT):
Could the person with the recording of yesterdays contributor call please add the link to https://wiki.hyperledger.org/display/BESU/2020-01-21+Contributor+Call
EdJoJob (Wed, 22 Jan 2020 00:20:33 GMT):
( @shemnon @bretthenderson probably ↑ )
bmg (Wed, 22 Jan 2020 14:55:09 GMT):
Has joined the channel.
PaulHarris (Thu, 23 Jan 2020 01:11:31 GMT):
Has joined the channel.
PaulHarris (Thu, 23 Jan 2020 01:14:41 GMT):
Hi all - we're considering updating picocli in besu from 3.9.6 to 4.1.4.
By making this change we're intending on making use of ArgGroups to save us from writing a bunch of code.
It's reportedly a backwards compatible change so we're not expecting any technical issues in this change, but does anyone have a reason not to make this change?
usmansaleem (Thu, 23 Jan 2020 01:32:57 GMT):
There is cli validation code and dependent options checks which might become obsolete and/or require changes. Other than that, we should upgrade to get benefit of new features provided by picocli
rjones (Thu, 23 Jan 2020 01:46:11 GMT):
may I edit your username to remove the space?
PaulHarris (Thu, 23 Jan 2020 01:53:26 GMT):
sure, that's what got generated - sorry
GabrieleOcyan (Fri, 24 Jan 2020 16:33:38 GMT):
Has joined the channel.
ldoyle (Fri, 24 Jan 2020 22:05:04 GMT):
Has joined the channel.
rjones (Wed, 29 Jan 2020 05:19:12 GMT):
The Fabric team has started using Mergify to help with merges - an example [here](https://github.com/hyperledger/fabric/pull/573). The config file they're using is [here](https://github.com/hyperledger/fabric/blob/master/.mergify.yml). If it's of interest, I can enable it on whatever repos you like.
Piyush.Sharma (Wed, 29 Jan 2020 05:45:40 GMT):
Has joined the channel.
Piyush.Sharma (Wed, 29 Jan 2020 05:45:41 GMT):
Hello everyone, I am Piyush Sharma , 2nd yr ECE student, would like to be a part of Hyperledger internship program 2020. I have a sound knowledge of Solidity, Java, C , C++ and blockchain as well. I am new to this eco-system , so any helping hand to get me this internship will be very beneficial .
hyperledger-bot (Wed, 29 Jan 2020 07:45:14 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/6506`
shemnon (Wed, 29 Jan 2020 17:50:04 GMT):
This was a flakey test. A re-run fixed it.
hyperledger-bot (Wed, 29 Jan 2020 17:50:58 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/6547`
hyperledger-bot (Wed, 29 Jan 2020 18:11:04 GMT):
unitTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/6550`
hyperledger-bot (Wed, 29 Jan 2020 18:11:04 GMT):
acceptanceTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/6553`
hyperledger-bot (Wed, 29 Jan 2020 18:11:04 GMT):
referenceTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/6551`
hyperledger-bot (Wed, 29 Jan 2020 18:11:04 GMT):
integrationTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/6552`
shemnon (Wed, 29 Jan 2020 18:33:54 GMT):
Besu is back to building normal.
EdJoJob (Wed, 29 Jan 2020 21:27:38 GMT):
Planning a 1.4.0-beta3 release in 24h. If there are any concerns over this, or any particular PR that should be merged in that time that we want to ensure is in, please let me know
matkt (Wed, 29 Jan 2020 22:50:18 GMT):
Hi ! I just created a PR to fix a bug concerning getSignerMetrics methods (BESU-179) . Do not hesitate to check it. https://github.com/hyperledger/besu/pull/343
faraggi (Thu, 30 Jan 2020 09:30:14 GMT):
Its been merged. thanks for the contribution @matkt !!
akuanti (Thu, 30 Jan 2020 14:39:06 GMT):
Has joined the channel.
hyperledger-bot (Thu, 30 Jan 2020 15:09:14 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/6690`
rjones (Thu, 30 Jan 2020 15:09:57 GMT):
https://chat.hyperledger.org/channel/community-architects?msg=TnYxXXfsvqNxTR2B3
rjones (Thu, 30 Jan 2020 15:09:57 GMT):
https://chat.hyperledger.org/channel/tsc?msg=FJ9iMNCaYhFdow3Kf
rjones (Thu, 30 Jan 2020 15:10:27 GMT):
Is this something Besu needs or wants?
shemnon (Thu, 30 Jan 2020 15:24:00 GMT):
The request is coming from a Besu Maintainer. It's a real nice Java profiler that offers free licenses to OSS projects in exchange for a shout out.
rjones (Thu, 30 Jan 2020 15:24:42 GMT):
OK, and the edit goes into your readme?
rjones (Thu, 30 Jan 2020 15:25:48 GMT):
As in: the links are added here? https://besu.hyperledger.org/en/stable/
rjones (Thu, 30 Jan 2020 15:27:15 GMT):
If I read the yourkit webpage correctly, I don't think you need anything from staff.
shemnon (Thu, 30 Jan 2020 16:14:58 GMT):
fixed
GregTheGreek (Thu, 30 Jan 2020 19:43:41 GMT):
when is the next besu release?
GregTheGreek (Thu, 30 Jan 2020 19:43:57 GMT):
An etc testnet just forked a week early
GregTheGreek (Thu, 30 Jan 2020 19:44:00 GMT):
lol...
GregTheGreek (Thu, 30 Jan 2020 19:44:11 GMT):
we might need to cut a release manually on the chainsafe repo
shemnon (Thu, 30 Jan 2020 19:51:35 GMT):
5 hours. Then two weeks for hte RC
shemnon (Thu, 30 Jan 2020 19:51:41 GMT):
It's beta 3
shemnon (Thu, 30 Jan 2020 19:51:46 GMT):
two weeks is RC 2.
shemnon (Thu, 30 Jan 2020 19:52:06 GMT):
If you want to implement it over 1.3 we can cut a 1.3.9 where the ETC fork data is the only delta.
shemnon (Thu, 30 Jan 2020 19:54:16 GMT):
So I think the ting to do is to make a pr off of 1.3.8 (https://github.com/hyperledger/besu/commit/45178419ec4ede37f8ae2b3688d50a0a80b9300f) and then when you are ready we can make a 1.3 branch and merge it into there. I believe we can do a release off of a branch but I'd have to ask the folks who do the release how hard that is.
GregTheGreek (Thu, 30 Jan 2020 20:03:31 GMT):
release is in 5 hours?
shemnon (Thu, 30 Jan 2020 20:06:09 GMT):
Beta 3.
shemnon (Thu, 30 Jan 2020 20:06:27 GMT):
I think getting the change into a 1.3.9 is a more prudent course.
GregTheGreek (Thu, 30 Jan 2020 20:06:38 GMT):
ok
GregTheGreek (Thu, 30 Jan 2020 20:06:47 GMT):
Well we can open the PR our changes are ready i believe
shemnon (Thu, 30 Jan 2020 20:07:34 GMT):
a pr into beta 3 would be good too.
shemnon (Thu, 30 Jan 2020 20:07:40 GMT):
which is the current master
ajsutton (Thu, 30 Jan 2020 20:11:02 GMT):
Yeah, PR into master first and then backport to the branch is the usual approach (not that we do branches much).
EdJoJob (Thu, 30 Jan 2020 22:51:34 GMT):
@all - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
EdJoJob (Fri, 31 Jan 2020 00:42:19 GMT):
@all - It's now safe to merge into master for hyperledger/besu again.
rjones (Fri, 31 Jan 2020 15:28:17 GMT):
@EdJoJob I could create a group like `besu release engineers` who could use the GitHub interface to set permissions for the other groups from whatever to `read` if you like?
rjones (Fri, 31 Jan 2020 16:16:14 GMT):
@EdJoJob I think [besu-admin](https://github.com/hyperledger/besu/settings/access) could go to this page, drop "besu-contributors" to "Read"
hyperledger-bot (Sat, 01 Feb 2020 07:30:34 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/6869`
EdJoJob (Sun, 02 Feb 2020 20:50:04 GMT):
that is another option that we didn't consider. I will look into that
shemnon (Mon, 03 Feb 2020 15:38:11 GMT):
This was a flakey test that went away on the next build
shemnon (Mon, 03 Feb 2020 22:09:17 GMT):
Besu contributor call tomorrow/today at 1500 UTC - https://consensys.zoom.us/j/199741148
shemnon (Mon, 03 Feb 2020 22:09:25 GMT):
Add your agenda items here - https://wiki.hyperledger.org/display/BESU/2020-02-04+Besu+Contributor+Call
faraggi (Tue, 04 Feb 2020 10:05:13 GMT):
I have some questions referencing the rollback of the latest PR that modified the ALL-CAPS.md files in the repo root:
(https://github.com/hyperledger/besu/pull/321)
1. If we have a roadmap in the repo do we want duplicate content on the wiki? (same goes for other files in the list)
2. can those files exist, yet point to contents in the wiki?
3. ROADMAP.md isn't mentioned in these links: https://wiki.hyperledger.org/display/TSC/Common+Repo+structure and https://wiki.hyperledger.org/display/TF/Repository+Structure+Task+Force
4. I'm still unsure of NOTICES file content.
5. I'm still unsure on what hard rules have been established and what is just still being discussed regarding the ALL-CAPS files.
@Silona @rjones @shemnon @ajsutton @lehors @ChristopherFerris ?
Silona (Tue, 04 Feb 2020 10:05:13 GMT):
Has joined the channel.
lehors (Tue, 04 Feb 2020 10:05:13 GMT):
Has joined the channel.
rjones (Tue, 04 Feb 2020 17:38:57 GMT):
faraggi
lehors (Wed, 05 Feb 2020 07:59:28 GMT):
as of now no decision has been made yet, what we have on the wiki is a proposal and feedback and questions to improve the proposal are welcome! I suggest you take this to the wiki though, where the task force people will more surely see it and can address it. Thank you.
lehors (Wed, 05 Feb 2020 07:59:28 GMT):
as of now no decision has been made yet, what we have on the wiki is a proposal and feedback and questions to improve the proposal are welcome! I suggest you take this to the wiki though, where the task force people will more surely see it and can address it. Thank you. @cbf
timbeiko (Wed, 05 Feb 2020 14:06:49 GMT):
As mentioned on yesterday's contributor call, I've added a template to the Besu Wiki for feature proposals. Feel free to discuss any ideas for Besu here informally, but if you want to submit a more fleshed out proposal, you can use this template to add it to the Wiki: https://wiki.hyperledger.org/display/BESU/Feature+Proposal+Template
shemnon (Wed, 05 Feb 2020 20:52:40 GMT):
The ETC Aztlan changes are in master. And Greg has said they are all that is needed. Should we cut a 1.3.9 with just those changes cherry picked?
Silona (Thu, 06 Feb 2020 01:54:32 GMT):
here is the Link @faraggi https://wiki.hyperledger.org/display/TSC/Common+Repo+structure
faraggi (Thu, 06 Feb 2020 14:36:29 GMT):
@jframe someone's asking a question about multitenancy implementation on the #besu channel. You might be best for answering this.
shemnon (Thu, 06 Feb 2020 15:40:45 GMT):
Here's a run of the proposed standards against BESU: https://gist.github.com/christo4ferris/13df399dff3613d405e8a60f23847385#file-gistfile1-txt-L7426
Silona (Thu, 06 Feb 2020 15:42:21 GMT):
https://wiki.hyperledger.org/display/TF/Repository+Structure+Task+Force
faraggi (Thu, 06 Feb 2020 15:56:28 GMT):
Thanks, much clearer now.
faraggi (Thu, 06 Feb 2020 15:56:50 GMT):
Why is there a github Issue template if most (all?) HL projects use Jira?
shemnon (Thu, 06 Feb 2020 15:57:10 GMT):
That's good feedback for the task force.
shemnon (Thu, 06 Feb 2020 15:57:26 GMT):
They can update the config to turn it off.
faraggi (Thu, 06 Feb 2020 15:57:41 GMT):
I'll write a comment on the corresponding wiki page.
faraggi (Thu, 06 Feb 2020 16:01:51 GMT):
https://gist.github.com/christo4ferris/13df399dff3613d405e8a60f23847385#file-gistfile1-txt-L7446
I believe these lines here will always clash with our tests because of how truffle migration files are structured.
faraggi (Thu, 06 Feb 2020 16:01:51 GMT):
___
https://gist.github.com/christo4ferris/13df399dff3613d405e8a60f23847385#file-gistfile1-txt-L7446
I believe these lines here will always clash with our tests because of how truffle migration files are structured.
shemnon (Thu, 06 Feb 2020 16:02:34 GMT):
Yea, I think not requiring copyright headers in javascripts is another tweak they should consider.
Silona (Thu, 06 Feb 2020 16:02:39 GMT):
please add MANY comments on the task force so Chris and TSC will see them. this is the reason Danno asked that this be run by the maintainers before voting
lehors (Thu, 06 Feb 2020 16:03:14 GMT):
s/MANY/ANY/ :-)
lehors (Thu, 06 Feb 2020 16:03:27 GMT):
it doesn't have to be many
Silona (Thu, 06 Feb 2020 16:08:00 GMT):
ok - comments that are here and that you have need to go there. rather than getting lost to the sands of chat
matkt (Thu, 06 Feb 2020 20:03:19 GMT):
I just published a PR for the bounty (https://github.com/PegaSysEng/BountiedWork/issues/3). Do not hesitate to tell me what you think of the proposed implementation (https://github.com/hyperledger/besu/pull/368).
GregTheGreek (Fri, 07 Feb 2020 00:13:36 GMT):
hey release team! We're ready for this PR: https://github.com/hyperledger/besu/pull/371
GregTheGreek (Fri, 07 Feb 2020 00:13:46 GMT):
to get a 1.3.9 release :D
GregTheGreek (Fri, 07 Feb 2020 00:18:08 GMT):
cc: @shemnon
ajsutton (Fri, 07 Feb 2020 00:49:02 GMT):
@joshuafernandes @EdJoJob :point_up:
EdJoJob (Fri, 07 Feb 2020 00:50:02 GMT):
No worries'
shemnon (Fri, 07 Feb 2020 00:59:25 GMT):
Refernce tests are failing because of CI issues:
> Submodule 'eth-ref-tests' (https://github.com/ethereum/tests.git) registered for path 'ethereum/referencetests/src/test/resources'
> Cloning into '/home/circleci/project/ethereum/referencetests/src/test/resources'...
> Fetched in submodule path 'ethereum/referencetests/src/test/resources', but it did not contain 48142b4ee7a42d081c6eda7b2a57d9b2c958bc81. Direct fetching of that commit failed.
joshuafernandes (Fri, 07 Feb 2020 01:16:38 GMT):
:thumbsup:
EdJoJob (Fri, 07 Feb 2020 04:49:08 GMT):
It took a little massaging, but Hyperledger Besu 1.3.9 is now live
EdJoJob (Fri, 07 Feb 2020 04:49:32 GMT):
@GregTheGreek ↑
GregTheGreek (Fri, 07 Feb 2020 15:39:26 GMT):
Amazing thank you
arash009 (Mon, 10 Feb 2020 02:08:37 GMT):
Hi all, we have been receiving some feedback over the past little while around ease of use of Jira vs Github for raising issues. The community response points to Github being a more familiar tool, as well as general access being already available, making it a lower barrier to entry to log issues, view progress of work, and engage with other contributors to the project. Based on this, we are considering switching the repository of Besu tickets from Jira to Github. Are there other considerations or other feedback that we should think about? This topic will also be tabled at the next few Besu contributor calls for discussion as well.
EdJoJob (Wed, 12 Feb 2020 02:15:10 GMT):
We are in the process of formalising the `plugin-api`s of Besu, as such we would like to publish the javadoc of the `plugin-api` jar. The standard is to have this accessed through javadocs.io, which would require us publishing the Besu repository on bintray to JCenter. Are there any objections to this action?
atoulme (Wed, 12 Feb 2020 17:32:14 GMT):
javadocs.io?
atoulme (Wed, 12 Feb 2020 17:33:14 GMT):
JCenter is fine. Maven Central is fine too. There is relatively no difference, except JCenter makes that sorta automated for you.
rjones (Wed, 12 Feb 2020 17:48:16 GMT):
@EdJoJob we are paying for BinTray and Artifactory today, if that informs your decision
NicolasMassart (Wed, 12 Feb 2020 18:45:55 GMT):
Issues with NAT.txt
NicolasMassart (Wed, 12 Feb 2020 18:48:44 GMT):
I also had:
```
2020-02-12 19:33:14.106+01:00 | pool-9-thread-1 | ERROR | RocksDBColumnarKeyValueStorage | Attempting to use a closed RocksDbKeyValueStorage
2020-02-12 19:33:14.106+01:00 | pool-9-thread-1 | ERROR | BlockMiner | Block mining threw an unhandled exception.
java.lang.IllegalStateException: Storage has been closed
at org.hyperledger.besu.plugin.services.storage.rocksdb.segmented.RocksDBColumnarKeyValueStorage.throwIfClosed(RocksDBColumnarKeyValueStorage.java:235) ~[besu-plugin-rocksdb-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.plugin.services.storage.rocksdb.segmented.RocksDBColumnarKeyValueStorage.get(RocksDBColumnarKeyValueStorage.java:150) ~[besu-plugin-rocksdb-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.plugin.services.storage.rocksdb.segmented.RocksDBColumnarKeyValueStorage.get(RocksDBColumnarKeyValueStorage.java:60) ~[besu-plugin-rocksdb-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.services.kvstore.SegmentedKeyValueStorageAdapter.get(SegmentedKeyValueStorageAdapter.java:50) ~[besu-kvstore-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.storage.keyvalue.WorldStateKeyValueStorage.getTrieNode(WorldStateKeyValueStorage.java:64) ~[besu-core-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.storage.keyvalue.WorldStateKeyValueStorage.getAccountStateTrieNode(WorldStateKeyValueStorage.java:52) ~[besu-core-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.storage.keyvalue.WorldStateKeyValueStorage.isWorldStateAvailable(WorldStateKeyValueStorage.java:81) ~[besu-core-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.worldstate.WorldStateArchive.getMutable(WorldStateArchive.java:54) ~[besu-core-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.blockcreation.AbstractBlockCreator.duplicateWorldStateAtParent(AbstractBlockCreator.java:229) ~[besu-blockcreation-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.blockcreation.AbstractBlockCreator.createBlock(AbstractBlockCreator.java:137) ~[besu-blockcreation-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.blockcreation.AbstractBlockCreator.createBlock(AbstractBlockCreator.java:119) ~[besu-blockcreation-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.blockcreation.BlockMiner.mineBlock(BlockMiner.java:123) ~[besu-blockcreation-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at org.hyperledger.besu.ethereum.blockcreation.BlockMiner.run(BlockMiner.java:82) [besu-blockcreation-1.4.0-beta4-SNAPSHOT.jar:1.4.0-beta4-dev-b6e25567]
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128) [?:?]
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628) [?:?]
at java.lang.Thread.run(Thread.java:835) [?:?]
```
Once when trying to shut down the node, with the same run command. I copied only one here, but had thousand and had to keep pressing Ctrl+C before it shut down after a while.
shemnon (Wed, 12 Feb 2020 19:12:07 GMT):
For the nat errors, the particular errors indicate to me the device advertising upnp services doesn't fully comply with the UPNP spec. jUPNP is fairly strict in that regard.
hyperledger-bot (Wed, 12 Feb 2020 22:52:41 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/8048`
ajsutton (Wed, 12 Feb 2020 22:55:35 GMT):
I believe our current release schedule is 1.4-RC1 today and then the full 1.4 release at the end of the month. There's some ongoing performance work that makes it more efficient to query logs over large numbers of blocks (https://github.com/hyperledger/besu/pull/367 is part of that, I believe there's more to come). Unfortunately that's not going to make today's RC1.
For ETH2 clients like Teku that optimisation is pretty essential because it allows us to query the whole chain to find validator deposits. We're getting really close to encouraging ETH2 enthusiasts to try our Teku and currently it's hard to setup besu in a way that will work - those performance improvements will fix that.
So, would it be reasonable to backport those optimisations to the 1.4 RC breanch and release an RC2 next week? Then Teku can just recommend people use Besu 1.4-RC2 which will hopefully also increase the amount testing it gets.
EdJoJob (Wed, 12 Feb 2020 23:00:18 GMT):
This was a DCO failure, since fixed
joshuafernandes (Thu, 13 Feb 2020 02:22:07 GMT):
@here: We're beginning the release process for hyperledger/besu -v 1.4.0-RC1
joshuafernandes (Thu, 13 Feb 2020 03:19:06 GMT):
@here: release process for hyperledger/besu is done - v1.4.0-rc1 is out
NicolasMassart (Thu, 13 Feb 2020 09:39:59 GMT):
I don't have any device advertising UPNP, that's the thing ;)
NicolasMassart (Thu, 13 Feb 2020 10:14:54 GMT):
Oups, ran NMAP to be sure, my box is advertising upnp service, even if the checkbox is unchecked in the UI... :weary:
NicolasMassart (Thu, 13 Feb 2020 10:15:41 GMT):
So I guess it's the device that doesn't comply to upnp. But this is probably not the cause of the shutdown instabilities.
shemnon (Thu, 13 Feb 2020 13:51:13 GMT):
Yea, post a bug for the first thread. Looks like we need to stop mining and wait for it to stop before shutting down the DB. I had mentioned it in a review but it was waived off.
NicolasMassart (Thu, 13 Feb 2020 13:51:43 GMT):
Will do, thanks.
NicolasMassart (Thu, 13 Feb 2020 16:21:20 GMT):
Each new external contributor we have on the doc repos fail to handle the DCO. I don't think any doc update or adding a banner about the DCO will change anything as it's obviously an issue with users not reading the guidelines first. Is there any way to prevent them to commit without the DCO in first place?
shemnon (Thu, 13 Feb 2020 16:27:40 GMT):
No, I don't think github allows hooks on PR publishing that would prevetnt a PR from being opened. Once the PR is opened all normal build checks come in and that's where the DCO checks are being applied.
faraggi (Thu, 13 Feb 2020 16:30:08 GMT):
What about a bot that comments on their PR with the error and instructions on how to fix it?
NicolasMassart (Thu, 13 Feb 2020 19:52:43 GMT):
The thing is that 100% of the contributors we had on the doc who had a DCO issue simply abandoned. So having a bot or myself telling them what to do to fix is useless. They are just lost.
shemnon (Thu, 13 Feb 2020 20:10:31 GMT):
The bad news is our hands are tied. The DCO mandate comes from the Linux Foundation.
shemnon (Thu, 13 Feb 2020 20:11:48 GMT):
The wiki does not have a DCO requirement.
NicolasMassart (Thu, 13 Feb 2020 20:33:44 GMT):
The doc is a repo, not a wiki. But I did not say to get rid of the DCO but find a way to make sure that contributors don't discover it only when CI runs on the PR.
rjones (Thu, 13 Feb 2020 20:36:39 GMT):
You could add a .git hook in your repo, right? an on-commit hook to force a `-s` or check the commit for a valid signature?
NicolasMassart (Thu, 13 Feb 2020 20:37:02 GMT):
Isn't the commit hook local only?
rjones (Thu, 13 Feb 2020 20:37:31 GMT):
sure, as part of your "byfn" readme you could say "add this"
rjones (Thu, 13 Feb 2020 20:38:17 GMT):
for gerrit, it was something like ```$ scp -p -P 29418
rjones (Thu, 13 Feb 2020 20:39:03 GMT):
perhaps part of your build process is to check for valid git.username and git.email and a valid commit on HEAD?
NicolasMassart (Thu, 13 Feb 2020 20:39:08 GMT):
At the scp line you already lost all our contributors. Well I think that the issue is they just don;t read the readme or any doc, otherwise the would know about the DCO. They are too used to the standard Github contribution scheme.
rjones (Thu, 13 Feb 2020 20:39:30 GMT):
gotcha.
NicolasMassart (Thu, 13 Feb 2020 20:52:47 GMT):
One issue we have also is that most contributors are using Github desktop our Github website directly to provide the changes, and they don't enable sign-off. So they would have to add it manually to the commit message.
One thing I wonder is, could we fix the DCO directly from the CI job? If no DCO added, use the Github user name and github no-reply email address which is username@users.noreply.github.com for the DCO?
rjones (Thu, 13 Feb 2020 21:05:12 GMT):
that isn't a valid DCO AFAIK
NicolasMassart (Thu, 13 Feb 2020 21:07:28 GMT):
Yes after writing this I told myself that the point of the DCO was to be added by the user, sorry. I'm trying to think in every direction to find something... tired to lose contributors because of something people say it will "minimize the barriers for contribution". In our case it's a blocker.
NicolasMassart (Thu, 13 Feb 2020 21:21:20 GMT):
The only thing contributors are forced to do is forking the repos first. Mayne we could use the fork event ( https://developer.github.com/v3/activity/events/types/#forkevent ) to trigger something.
The only action I see as something we could trigger is the creation of an issue in the new fork telling the user to be careful of the DCO, read guidelines and add thanks for contributing.
NicolasMassart (Thu, 13 Feb 2020 21:21:20 GMT):
The only thing contributors are forced to do is forking the repos first. Maybe we could use the fork event ( https://developer.github.com/v3/activity/events/types/#forkevent ) to trigger something.
The only action I see as something we could trigger is the creation of an issue in the new fork telling the user to be careful of the DCO, read guidelines and add thanks for contributing.
NicolasMassart (Thu, 13 Feb 2020 21:37:17 GMT):
Or could we say the doc doesn't need the DCO and only the Besu software repos?
NicolasMassart (Thu, 13 Feb 2020 21:37:51 GMT):
For more context on what the issue is: https://github.com/hyperledger/besu-docs/pull/166#issuecomment-585982354
NicolasMassart (Thu, 13 Feb 2020 21:39:23 GMT):
We had two this month. And the first one abandoned the PR. We had to close it as someone else inside our team made the same change before the external contributor.
rjones (Thu, 13 Feb 2020 23:49:00 GMT):
@dhuseby
hyperledger-bot (Fri, 14 Feb 2020 07:05:55 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/8298`
NicolasMassart (Fri, 14 Feb 2020 14:23:15 GMT):
I did not pay him to do that, I promise.
But then perhaps https://github.com/scottrigby/dco-gh-ui could help to prevent this on PR merge?
mbaxter (Fri, 14 Feb 2020 14:31:22 GMT):
The head of `master` branch for repository `hyperledger/besu` has been force updated to deal with a DCO issue, please follow the instructions in the following article to fix your local repositories: https://wiki.hyperledger.org/pages/viewpage.action?pageId=24781754#
rjones (Fri, 14 Feb 2020 18:14:07 GMT):
I appreciate that Besu is setting such a high bar for this.
ggarri (Sun, 16 Feb 2020 10:45:44 GMT):
Has joined the channel.
lucassaldanha (Mon, 17 Feb 2020 02:48:23 GMT):
We have a proposal for a new Besu maintainer. If you are a maintainer, you have 2 weeks to cast your vote!
https://lists.hyperledger.org/g/besu/message/47
faraggi (Mon, 17 Feb 2020 09:26:56 GMT):
I think Nico's fork notification could be a great way to avoid some DCO issues.
faraggi (Mon, 17 Feb 2020 09:26:56 GMT):
I think @NicolasMassart 's fork notification could be a great way to avoid some DCO issues.
shemnon (Tue, 18 Feb 2020 00:55:17 GMT):
APAC besu contributr call in ~ 5 min - https://consensys.zoom.us/j/524697595
atoulme (Tue, 18 Feb 2020 20:16:44 GMT):
hey folks "I'm not terribly excited about using an external library for what should be core functionality." <--- this is part of the review from @shemnon on https://github.com/hyperledger/besu/pull/383
atoulme (Tue, 18 Feb 2020 20:17:25 GMT):
OK, so is this core? Besu didn't have this capacity until now. Do you want to copy the code of Apache Tuweni into Besu for this? How is that going to be defined going forward?
arash009 (Wed, 19 Feb 2020 02:10:08 GMT):
Hi everyone. We have an RC2 being planned for release in ~24hrs time. There are a number of PRs that have gone into master since RC1 and a number that have been backported to release branch. Anything else outstanding that needs to come in? Please reach out if you have work in flight that we need to keep in mind.
lucassaldanha (Wed, 19 Feb 2020 02:13:55 GMT):
Hi Arash, we have finished the privacy work that needs to go into RC2
arash009 (Wed, 19 Feb 2020 02:19:27 GMT):
Have those changes been backported already?
lucassaldanha (Wed, 19 Feb 2020 02:19:44 GMT):
Of course! :)
arash009 (Wed, 19 Feb 2020 04:51:02 GMT):
Here are the other PRs that I think have since RC1 been merged to master. Which of these do we need to include in RC2 release:
https://github.com/hyperledger/besu/pull/367
https://github.com/hyperledger/besu/pull/392
https://github.com/hyperledger/besu/pull/401
https://github.com/hyperledger/besu/pull/400
https://github.com/hyperledger/besu/pull/407
https://github.com/hyperledger/besu/pull/409
hyperledger-bot (Wed, 19 Feb 2020 18:32:52 GMT):
publish for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/8539`
hyperledger-bot (Wed, 19 Feb 2020 19:07:14 GMT):
publish for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/8578`
hyperledger-bot (Wed, 19 Feb 2020 20:21:16 GMT):
publish for besu - release-1.4 : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/8587`
EdJoJob (Wed, 19 Feb 2020 21:12:19 GMT):
These jobs are timing out on the bintray upload. I am investigating now
rjones (Wed, 19 Feb 2020 21:15:04 GMT):
I can't even open bintray's website.
EdJoJob (Wed, 19 Feb 2020 21:23:30 GMT):
https://status.bintray.com/ implies that they are having UI issues, but not REST
hyperledger-bot (Wed, 19 Feb 2020 21:24:09 GMT):
publish for besu - release-1.4 : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/8588`
rjones (Wed, 19 Feb 2020 21:24:56 GMT):
@EdJoJob when I go here: https://bintray.com/beta/#/hyperledger-org/besu-repo?tab=packages I see an error which says there are unpublished files that will be deleted. Perhaps related?
rjones (Wed, 19 Feb 2020 21:25:23 GMT):
The updated messages says both: ```Investigating - We are experiencing some problems with bintray UI and REST API.
Our team is looking into it. ```
EdJoJob (Wed, 19 Feb 2020 21:25:24 GMT):
that will be related. The way we do the publish is upload and then mark as published
rjones (Wed, 19 Feb 2020 21:25:31 GMT):
gotcha.
EdJoJob (Wed, 19 Feb 2020 21:26:17 GMT):
I must have skimmed a little too quickly there
EdJoJob (Wed, 19 Feb 2020 23:18:23 GMT):
The release that was planned for today will be delayed until the bintray status is stable again
hyperledger-bot (Thu, 20 Feb 2020 00:31:12 GMT):
publish for besu - release-1.4 : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/8590`
EdJoJob (Thu, 20 Feb 2020 03:40:01 GMT):
Release 1.4.0-RC2 is happening now
NicolasMassart (Thu, 20 Feb 2020 13:17:08 GMT):
+1 for it to be a plugin
shemnon (Thu, 20 Feb 2020 16:26:30 GMT):
I was hpoing more contributors than Nicolas would have chimed in. In my mind the line is "would anyting other than an ethereum client ever use it." Bytes and units - yes others would use it. Crypo? Yep, definately used by others. But ethstats is something that only clients would ever use to report. So when weighing wether to add more jars to the dependencies that's when I ask "Should it be an external dependendcy or not"
shemnon (Thu, 20 Feb 2020 16:26:30 GMT):
I was hoping more contributors than Nicolas would have chimed in. In my mind the line is "would anyting other than an ethereum client ever use it." Bytes and units - yes others would use it. Crypo? Yep, definately used by others. But ethstats is something that only clients would ever use to report. So when weighing wether to add more jars to the dependencies that's when I ask "Should it be an external dependendcy or not"
shemnon (Thu, 20 Feb 2020 16:26:50 GMT):
That is for jars for the core client, which are mandatory for everyone running besu.
shemnon (Thu, 20 Feb 2020 16:28:05 GMT):
Enter the plugin system. Plugins are not mandatory and end users can strip them out, or not install them in the first place. And for this feature it may work well as a plugin becuase not all enterprises may want to post to ethstats.net.
shemnon (Thu, 20 Feb 2020 16:28:19 GMT):
Especially for their internal chains.
atoulme (Thu, 20 Feb 2020 17:46:44 GMT):
Well other clients such as rskj could use it. I also am building light clients with Tuweni that would use it.
shemnon (Thu, 20 Feb 2020 17:52:34 GMT):
what all is in that jar? Can it be done without logl?
shemnon (Thu, 20 Feb 2020 17:52:56 GMT):
Other softeware actually usingit supports the library case.
shemnon (Thu, 20 Feb 2020 17:52:56 GMT):
Other softeware actually using it supports the library case.
shemnon (Thu, 20 Feb 2020 17:53:14 GMT):
But I also feel the plugin route would be a good alternative too.
shemnon (Thu, 20 Feb 2020 17:54:57 GMT):
Does rootstock use ethstats? Or is it just the same protocol? Calling it ethstats and using it in rootstock seems like it needs a more generic name.
abdelhamidbakhta (Thu, 20 Feb 2020 18:18:06 GMT):
Has joined the channel.
abdelhamidbakhta (Thu, 20 Feb 2020 18:18:08 GMT):
New Maintainer candidate: Antoine Toulme - https://github.com/hyperledger/besu/pull/422
atoulme (Thu, 20 Feb 2020 20:14:31 GMT):
the naming discussion can happen in Apache Tuweni I guess.
atoulme (Thu, 20 Feb 2020 20:15:26 GMT):
logl is just a functional proxy to not tie ethstats to a particular logging impl. If it bothers you, we can also use lambdas to log. I think ethstats (right now) only logs very seldomly.
atoulme (Thu, 20 Feb 2020 20:16:12 GMT):
If I had to pick another stack than logl, I'd go with slf4j. Opinions in slf4j go the other way, but it's also a good proxy into any logging lib.
shemnon (Thu, 20 Feb 2020 20:29:35 GMT):
Right now the only logging library in any Besu code is Log4j2, for better or worse. Any change to that is an issue that IMHO should be addressed in isolation not in context of any specific contribution. Because java logging is complex it has long term impacts.
shemnon (Thu, 20 Feb 2020 20:30:27 GMT):
Another alternative is to use logl internally and provide only the jars for logl->log4j interop. We do this in several places where the log4j2 api jars are what we compile to and we look for the log4j2 impls underneath.
shemnon (Thu, 20 Feb 2020 20:31:14 GMT):
But if we re-examine logging one I think we should seriously consider is Google' fluent logger. It would solve a lot of problems we have accidentally passing in a throwable as a formatting argument. And provide richer configuraiton such as "log no more than once per minute"
shemnon (Thu, 20 Feb 2020 20:31:14 GMT):
But if we re-examine logging one I think we should seriously consider is Google's fluent logger. It would solve a lot of problems we have accidentally passing in a throwable as a formatting argument. And provide richer configuraiton such as "log no more than once per minute"
shemnon (Thu, 20 Feb 2020 20:31:24 GMT):
in line in the code for specific logs.
atoulme (Thu, 20 Feb 2020 20:34:37 GMT):
understood.
ajsutton (Thu, 20 Feb 2020 21:01:55 GMT):
Besu does have dependencies already using slf4j so that would be a preference for me over also introducing logl.
ajsutton (Thu, 20 Feb 2020 21:07:04 GMT):
I've had this thread open for a few days still pondering exactly where I sit on the question of what goes into besu vs Tuweni. I think it would be weird for example to have built an EVM in Tuweni and use that in Besu - just feels too core. That's exacerbated by the different foundations and governance involved as well.
It's not clear to me where ethstats sits on the "build vs buy" kind of spectrum. It's probably ok to "buy" but not completely clear cut for me.
I really like the idea of building EthStats as a plugin and it was my first thought when I saw the PR as well.
NicolasMassart (Mon, 24 Feb 2020 10:01:16 GMT):
I created a PR on the doc repos to propose Grant Noble as a doc maintainer, please have a look at the PR for more details on the process: https://github.com/hyperledger/besu-docs/pull/199
atoulme (Mon, 24 Feb 2020 18:34:13 GMT):
ok cool. I'll close the PR for now, with that decision.
timbeiko (Mon, 24 Feb 2020 22:10:56 GMT):
The Besu roadmap doc has been updated on the wiki to reflect our upcoming 1.4 release: https://wiki.hyperledger.org/display/BESU/Roadmap
terrencecooke (Tue, 25 Feb 2020 22:05:08 GMT):
Has joined the channel.
NicolasMassart (Wed, 26 Feb 2020 10:48:23 GMT):
We are 4 maintainers on this repos and we have now 2 validations and I'm the one who created the PR. The fourth, Steven, is not part of Pegasys anymore and I doubt that he will participate in this. Can we consider that 2+me gives us the 3 required maintainers without waiting for two weeks?
NicolasMassart (Wed, 26 Feb 2020 10:48:23 GMT):
We are 4 maintainers on this repos and we have now 2 validations and I'm the one who created the PR. The fourth, Steven, is not part of Pegasys anymore and I doubt that he will participate in this. Can we consider that 2+me gives us the absolute majority or do we just meet the 3 required maintainers and we have to wait for two weeks?
shemnon (Wed, 26 Feb 2020 16:00:29 GMT):
I am of the opinion that we should wait for inactivity before we remove steve as a maintainer. The "vote a maintainer out" clause was intended to deal with misconduct, not change in employment.
shemnon (Wed, 26 Feb 2020 16:01:11 GMT):
3/4 is an absolute majority.
shemnon (Wed, 26 Feb 2020 16:01:41 GMT):
You can and should vote on your own pr.
shemnon (Wed, 26 Feb 2020 16:02:37 GMT):
Since the mechanic is "approvals" and github won't let you approve your own PRs a "LGTM" or "I approve" in the comment stream should be sufficient
NicolasMassart (Wed, 26 Feb 2020 17:46:50 GMT):
Thanks I will merge then.
NicolasMassart (Wed, 26 Feb 2020 17:47:19 GMT):
https://github.com/hyperledger/besu-docs/pull/199 is now merged. Can anyone with admin rights on this repos add Grant as a maintainer please?
arash009 (Wed, 26 Feb 2020 23:02:55 GMT):
https://github.com/hyperledger/besu/pull/382 does this one need to be in for 1.4.0 release?
arash009 (Wed, 26 Feb 2020 23:04:06 GMT):
@shemnon @GregTheGreek
shemnon (Wed, 26 Feb 2020 23:05:04 GMT):
I pinged greg privately and @ them on github last week, no response.
shemnon (Wed, 26 Feb 2020 23:05:53 GMT):
Based on the tea-leaves of what I'm reading on the ETC-Coop's discord two weeks won't hurt, this will automatically show up in 1.4.1.
GregTheGreek (Wed, 26 Feb 2020 23:05:54 GMT):
hey sorry
GregTheGreek (Wed, 26 Feb 2020 23:06:04 GMT):
I just got the email
shemnon (Wed, 26 Feb 2020 23:06:11 GMT):
wait...
GregTheGreek (Wed, 26 Feb 2020 23:06:12 GMT):
woops sorry in the thread
GregTheGreek (Wed, 26 Feb 2020 23:06:21 GMT):
so
GregTheGreek (Wed, 26 Feb 2020 23:06:24 GMT):
we just had a call today
GregTheGreek (Wed, 26 Feb 2020 23:08:19 GMT):
ETC is reverting to vanilla istanbul
GregTheGreek (Wed, 26 Feb 2020 23:08:25 GMT):
So we'll be reverting it
shemnon (Wed, 26 Feb 2020 23:08:44 GMT):
:+1:
shemnon (Wed, 26 Feb 2020 23:09:00 GMT):
So Aztlan/Phoenix should be scrubbed from the codebase or will it make another return?
GregTheGreek (Wed, 26 Feb 2020 23:09:16 GMT):
yeah we'll scrub it
GregTheGreek (Wed, 26 Feb 2020 23:09:25 GMT):
either tonight or tomorrow
arash009 (Wed, 26 Feb 2020 23:11:19 GMT):
Cool, so we'll proceed with 1.4.0 release and the changes will be reflected in 1.4.1
joshuafernandes (Wed, 26 Feb 2020 23:36:23 GMT):
@all - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.4 until further notice
Benjamin (Thu, 27 Feb 2020 00:37:07 GMT):
Has joined the channel.
joshuafernandes (Thu, 27 Feb 2020 01:03:23 GMT):
@all - release process for hyperledger/besu, 1.4 is done
GregTheGreek (Thu, 27 Feb 2020 03:12:55 GMT):
:)
GregTheGreek (Thu, 27 Feb 2020 03:13:24 GMT):
thanks guys, really appreciate the professionalism. Its truly a pleasure to be contributing
NicolasMassart (Thu, 27 Feb 2020 10:29:26 GMT):
I don't have permission to view https://wiki.hyperledger.org/display/BESU/Contributing+Guidelines even when I'm logged in with my LFID. Is it expected?
faraggi (Thu, 27 Feb 2020 11:43:02 GMT):
That mega file doesn't exist anymore. It was replaced by the, more atomic, contributor wiki docs.
rjones (Thu, 27 Feb 2020 16:43:19 GMT):
@faraggi we have a redirect plugin installed - you could add a redirect to those pages to the new pages, if you like.
faraggi (Thu, 27 Feb 2020 16:45:43 GMT):
Thanks @rjones , I'll do that in the coming days
NicolasMassart (Thu, 27 Feb 2020 17:27:24 GMT):
It's still linked from https://wiki.hyperledger.org/display/BESU/Contributing+to+documentation
faraggi (Fri, 28 Feb 2020 20:19:12 GMT):
Hello all!
Github issues are now officially enabled for Besu. https://github.com/hyperledger/besu/issues
lucassaldanha (Sun, 01 Mar 2020 23:45:30 GMT):
Hi folks, my proposal to add Stefan Pingel as a maintainer has passed. According to our maintainer process, Stefan Pingel has become a Besu maintainer! :besu:
I'll merge the proposal PR in a moment :+1:
https://github.com/hyperledger/besu/pull/405
shemnon (Mon, 02 Mar 2020 01:50:31 GMT):
We also have another maintainer vote pending that closes thursday - https://github.com/hyperledger/besu/pull/422
shemnon (Mon, 02 Mar 2020 15:43:47 GMT):
Besu Contributor Call tomorrow - 3 Mar 1500 UTC - Right before the Hyperledger Global Forum keynote! - https://wiki.hyperledger.org/display/BESU/2018-03-03+Contributor+Call
shemnon (Mon, 02 Mar 2020 15:43:47 GMT):
Besu Contributor Call tomorrow - 3 Mar 1500 UTC - Right before the Hyperledger Global Forum keynote! - https://wiki.hyperledger.org/display/BESU/2020-03-03+Contributor+Call
jasonatran (Mon, 02 Mar 2020 17:40:09 GMT):
Has joined the channel.
shemnon (Tue, 03 Mar 2020 14:57:05 GMT):
Contributor call in 5 minutes: https://consensys.zoom.us/j/199741148
shemnon (Tue, 03 Mar 2020 15:11:44 GMT):
Because of EthCC and Hyperledger Global Fourm we didn't have quorum for a meeting.
shemnon (Tue, 03 Mar 2020 15:12:35 GMT):
I propose that we shift the meeting times by two weeks, so in two weeks we have an EMEA/AMER friendly time, and then an APAC.AMER friendly meeting. Making sure that each successful meeting alternates time zones.
shemnon (Tue, 03 Mar 2020 15:12:35 GMT):
I propose that we shift the meeting times by two weeks, so in two weeks we have an EMEA/AMER friendly time, and then an APAC.AMER friendly meeting four weeks from now instead of two. Making sure that each successful meeting alternates time zones.
shemnon (Tue, 03 Mar 2020 15:12:35 GMT):
I propose that we shift the meeting times by two weeks, so in two weeks we have an EMEA/AMER friendly time, and then an APAC/AMER friendly meeting four weeks from now instead of two. Making sure that each successful meeting alternates time zones.
GregTheGreek (Thu, 05 Mar 2020 15:02:01 GMT):
hey all - sorry for the craziness with ETC lately. #434 (https://github.com/hyperledger/besu/pull/434) is final and conforms to the latest spec
abdelhamidbakhta (Thu, 05 Mar 2020 15:53:00 GMT):
Hello, my proposal to add Antoine Toulme as a maintainer has passed. According to our maintainer process, Antoine Toulme has become a Besu maintainer! :besu:
abdelhamidbakhta (Thu, 05 Mar 2020 15:53:23 GMT):
The proposal PR has been merged - https://github.com/hyperledger/besu/pull/422
rjones (Thu, 05 Mar 2020 18:07:57 GMT):
@abdelhamidbakhta I've invited `atoulme` to the GitHub org with join `https://github.com/orgs/hyperledger/teams/besu-contributors/members`
shemnon (Thu, 05 Mar 2020 19:37:39 GMT):
I may not get to it until next week. Perhaps another maitainer can review it?
shemnon (Thu, 05 Mar 2020 19:37:39 GMT):
I may not get to it until next week. Perhaps another maintainer can review it?
shemnon (Fri, 06 Mar 2020 00:34:27 GMT):
Remind me monday if it doesn't get reviewed.
lucassaldanha (Fri, 06 Mar 2020 02:21:48 GMT):
I'm looking at it
GregTheGreek (Fri, 06 Mar 2020 02:59:25 GMT):
thanks :)
GregTheGreek (Fri, 06 Mar 2020 03:11:47 GMT):
In the `MainnetContractCreationProcessor` what is the `accountVersion`? Is that for EVM versioning?
GregTheGreek (Fri, 06 Mar 2020 03:14:37 GMT):
Also curious about `forceCommitAddresses` ? (happy to be pointed at an eip if its easier)
shemnon (Fri, 06 Mar 2020 06:32:59 GMT):
EIP-1702 support. Implemented it before it got pulled
shemnon (Fri, 06 Mar 2020 06:35:40 GMT):
Something to do with shanghai state pool clearing and the way parity did precompiles. For some reason they required some value or Parity deleted the precomple.
ajsutton (Fri, 06 Mar 2020 07:03:38 GMT):
Yeah not sure there is an EIP for that one. As Danno said it boiled down to a buggy implementation but either it was in both Parity and Geth or there was some other reason that the buggy chain wound up being considered canonical so now every client reproduces the bug for those key addresses in that particular hard fork.
shemnon (Tue, 10 Mar 2020 22:47:38 GMT):
PR review request - Code of Conduct update - https://github.com/hyperledger/besu/pull/446
All Maintainers should review this change before it goes in. This places us under only the Hyperledger Code of Conduct whereas we were under two before: Hyperledger's and the Contributor Covanant from the original Pantheon contribution.
rjones (Tue, 10 Mar 2020 22:52:43 GMT):
Is Besu moving from JIRA to GitHub issues, or using both going forward?
shemnon (Tue, 10 Mar 2020 23:41:20 GMT):
I think the desire is Github issues unless things go bad. So I wouldn't turn off JIRA quite yet.
rjones (Wed, 11 Mar 2020 00:13:31 GMT):
I was going to do a pass on the documentation, which contains pointers to the old JIRA, the new JIRA, and doesn'
rjones (Wed, 11 Mar 2020 00:13:31 GMT):
I was going to do a pass on the documentation, which contains pointers to the old JIRA, the new JIRA, and doesn't mention GitHub issues
shemnon (Wed, 11 Mar 2020 00:15:33 GMT):
@EdJoJob and @joshuafernandes would be the best to answer that.
madelinemurray (Wed, 11 Mar 2020 04:04:30 GMT):
@faraggi - you were updating for this weren't you?
arash009 (Wed, 11 Mar 2020 05:19:33 GMT):
We havent fully migrated all our Jira issues to GH, but thats the intention. Ideally we should update the documentaion to start pointing new issues to be created in GH whilst we do the migration.
faraggi (Wed, 11 Mar 2020 09:59:03 GMT):
Yes, I have an incoming PR with the github edits
as well as wiki edits for the same purpos
faraggi (Wed, 11 Mar 2020 09:59:03 GMT):
Yes, I have an incoming PR with the github edits
as well as wiki edits for the same purpose
faraggi (Wed, 11 Mar 2020 09:59:03 GMT):
Yes, I have an incoming PR with the repo edits
as well as wiki edits for the same purpose
faraggi (Wed, 11 Mar 2020 10:15:36 GMT):
https://github.com/hyperledger/besu/pull/448
rjones (Wed, 11 Mar 2020 15:02:03 GMT):
OK cool - I'll relax.
rjones (Wed, 11 Mar 2020 15:03:22 GMT):
This is part of an effort I was discussing with @grace.hartley - feel free to talk with her directly for details
timbeiko (Wed, 11 Mar 2020 19:47:29 GMT):
Besu 1.4.1 is going out tomorrow and there are a few PRs for which there is no changelog update:
* https://github.com/hyperledger/besu/pull/434 (cc: @mackcom)
* https://github.com/hyperledger/besu/pull/436 (cc: @anthonybuckle)
If you'd like the change documented in the release notes, please add a line to CHANGELOG.md by tomorrow afternoon EST. Thanks :smiley:
mackcom (Thu, 12 Mar 2020 15:08:43 GMT):
@timbeiko What do you need for the change log, and how do I add it?
shemnon (Thu, 12 Mar 2020 15:18:08 GMT):
@mackcom something like this - https://github.com/hyperledger/besu/pull/461
shemnon (Thu, 12 Mar 2020 15:19:12 GMT):
In the future it's easiest to do it as part of the PR. I myself am bad about it but it would be good if we all got in the habit of either (a) updating the changelog in the PR (b) saying that the PR is not changelog worthy or (c) pointing to another PR where the changelog entries are already taking care of what changed.
shemnon (Thu, 12 Mar 2020 15:20:10 GMT):
So I proposed last week we do the next contributor call on EMEA/AMER time. Was there any objection to or support of this proposal from any of the contributors?
mackcom (Thu, 12 Mar 2020 17:33:01 GMT):
Ok, I've put in PR#462 with the changelog update. thanks.
rjones (Fri, 13 Mar 2020 15:44:37 GMT):
Hey, folks. For first time contributors with DCO struggles, there is a way for you to fix it (usually). I was able to fix [this PR for cello](https://github.com/hyperledger/cello/pull/117) using the instructions in the error.
rjones (Fri, 13 Mar 2020 15:44:37 GMT):
Hey, folks. For first time contributors with DCO struggles, there is a way for you to fix it (usually). I was able to fix [this PR for cello](https://github.com/hyperledger/cello/pull/117) using [the instructions in the error](https://gist.github.com/ryjones/a203fdd910f13c5cd3e44da3346ae05b).
shemnon (Sat, 14 Mar 2020 22:38:30 GMT):
Next Tuesday is the next Besu contributor call, on EMEA/AMER friendly times. - https://wiki.hyperledger.org/display/BESU/2020-03-17+Contributor+Call
If you are in the USA check the times as the call is fixed against UTC, so the call may be an hour later.
timbeiko (Mon, 16 Mar 2020 15:25:56 GMT):
:tada: Hyperledger Besu 1.4.1 is out today :tada:
We’ve added more parity-style tracing (trace_transaction), the ability to add/remove people from privacy groups and other privacy, pruning, ETC and devp2p improvements
Get it at https://github.com/hyperledger/besu/releases/1.4.1 :besu:
shemnon (Tue, 17 Mar 2020 00:05:32 GMT):
Reminder, Besu Contributor call is in 15 hours from now, not 1 hour from now.
shemnon (Tue, 17 Mar 2020 14:08:52 GMT):
Contributor call at the top of the hour (~52 minutes) - https://consensys.zoom.us/j/199741148
srv.twry (Tue, 24 Mar 2020 14:34:42 GMT):
Has joined the channel.
benjamincburns (Tue, 24 Mar 2020 19:03:56 GMT):
Has joined the channel.
shemnon (Tue, 24 Mar 2020 23:51:43 GMT):
review request - fix checkLicense - https://github.com/hyperledger/besu/pull/568
hyperledger-bot (Wed, 25 Mar 2020 04:14:40 GMT):
referenceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/10538`
shemnon (Wed, 25 Mar 2020 04:58:16 GMT):
Looks to be a problem related to cloning the repo. Re-running before I go to bed.
EdJoJob (Wed, 25 Mar 2020 22:11:01 GMT):
We are currently releasing Besu 1.4.1
rjones (Thu, 26 Mar 2020 14:16:14 GMT):
BESU IS ACTIVE STATUS
madelinemurray (Fri, 27 Mar 2020 03:57:17 GMT):
🎉 Hyperledger Besu 1.4.2 is out 🎉
We’ve added more parity-style tracing (trace_block) and enhanced the syncing query and eth_syncing JSON-RPC API
Get it at https://github.com/hyperledger/besu/releases/1.4.2
atoulme (Fri, 27 Mar 2020 18:54:09 GMT):
hey folks, I am opening https://github.com/hyperledger/besu/pull/608/files for eth65 support.
atoulme (Fri, 27 Mar 2020 18:54:32 GMT):
It might need debugging/acceptance testing as this type of things tend to fall apart at the edge.
atoulme (Fri, 27 Mar 2020 18:54:40 GMT):
It's ready for review.
shemnon (Fri, 27 Mar 2020 19:25:46 GMT):
Nice. Have you given it a spin on mainnet? I understand 1/3 of geth nodes are already talking eth/65. IIRC transactions are still passed around even while syncing so we don't need a fully syned node to see it work.
atoulme (Sat, 28 Mar 2020 06:38:21 GMT):
No, not yet. I am building an acceptance test first.
atoulme (Sat, 28 Mar 2020 07:07:04 GMT):
alright, here is the important bit. An acceptance test you can see here: https://github.com/hyperledger/besu/pull/608/commits/defbed1d0abf2bf326b68178cac3aba53a2832a9
atoulme (Sat, 28 Mar 2020 07:07:17 GMT):
``` @Test
public void shouldGossipATransaction() {
final Account account = accounts.createAccount("account-one");
final Amount balance = Amount.ether(20);
TransferTransaction tx = accountTransactions.createTransfer(account, balance);
Hash txHash = archiveNode1.execute(tx);
cluster.verify(txPoolConditions.inTransactionPool(txHash));
}```
atoulme (Sat, 28 Mar 2020 07:07:43 GMT):
this is between 2 archive nodes, so neither can mint blocks. One receives a transaction, and it is gossiped to the other node.
madelinemurray (Mon, 30 Mar 2020 01:53:13 GMT):
📣 We've identified a critical issue for privacy users with private transactions created using Hyperledger Besu v1.3.4 or earlier. If you have a network with private transaction created using v1.3.4 or earlier, please read the following and take the appropriate steps:
https://wiki.hyperledger.org/display/BESU/Critical+Issue+for+Privacy+Users
As always, the PegaSys tam is happy to answer any questions or help with any issues you have. 📣
madelinemurray (Mon, 30 Mar 2020 01:53:13 GMT):
📣 We've identified a critical issue for privacy users with private transactions created using Hyperledger Besu v1.3.4 or earlier. If you have a network with private transaction created using v1.3.4 or earlier, please read the following and take the appropriate steps:
https://wiki.hyperledger.org/display/BESU/Critical+Issue+for+Privacy+Users
As always, the PegaSys team is happy to answer any questions or help with any issues you have. 📣
shemnon (Mon, 30 Mar 2020 05:14:35 GMT):
Besu Contributor Call tomorrow, 1 April APAC 30 Mar AMER - https://wiki.hyperledger.org/display/BESU/2020-04-01+Besu+Contributor+Call
shemnon (Mon, 30 Mar 2020 05:14:35 GMT):
Besu Contributor Call tomorrow, ~~1 April~~ 31 March APAC 30 Mar AMER - https://wiki.hyperledger.org/display/BESU/2020-03-31+Besu+Contributor+Call
shemnon (Mon, 30 Mar 2020 05:14:41 GMT):
Feel free to add agenda items
hyperledger-bot (Mon, 30 Mar 2020 17:20:45 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/sbs2001/besu/2`
atoulme (Mon, 30 Mar 2020 17:45:13 GMT):
@shemnon thanks for the review. I'll reply tonight after hours.
shemnon (Tue, 31 Mar 2020 00:26:15 GMT):
Contributor call in 30 minutes. My mistake on the header, I forgot march has 31 days.
shemnon (Tue, 31 Mar 2020 00:29:35 GMT):
https://consensys.zoom.us/j/524697595
hyperledger-bot (Tue, 31 Mar 2020 09:38:24 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/sbs2001/besu/3`
hyperledger-bot (Tue, 31 Mar 2020 09:40:03 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/sbs2001/besu/4`
hyperledger-bot (Tue, 31 Mar 2020 09:48:03 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/sbs2001/besu/5`
sbs2001 (Wed, 01 Apr 2020 12:17:06 GMT):
Has joined the channel.
sbs2001 (Wed, 01 Apr 2020 12:31:39 GMT):
Hello everyone, I am Shivam interested in contributing to Besu. Having said that, I picked up the issue https://github.com/hyperledger/besu/issues/490 to get a better understanding of the codebase, and I am stuck on the issue because I can't figure out
sbs2001 (Wed, 01 Apr 2020 12:31:39 GMT):
Hello everyone, I am Shivam interested in contributing to Besu. Having said that, I picked up the issue https://github.com/hyperledger/besu/issues/490 to get a better understanding of the codebase, and I am stuck on the issue because I can't figure out how to get a node by its id, and how to submit the hash rate once this node is obtained.
sbs2001 (Wed, 01 Apr 2020 12:32:47 GMT):
Please provide some pointers on how to achieve that.
faraggi (Thu, 02 Apr 2020 10:14:39 GMT):
Hi @sbs2001 ! welcome to the Besu rocketchat.
Have you taken a look at our wiki, which goes over how to contribute to the project?
https://wiki.hyperledger.org/display/BESU/Contributing
https://wiki.hyperledger.org/display/BESU/How+to+Contribute
Once you've done that, feel free to let us know which issu you'd prefer to work on so we can assign it to you.
sbs2001 (Thu, 02 Apr 2020 10:16:43 GMT):
I have gone through the guide :)
faraggi (Thu, 02 Apr 2020 10:17:55 GMT):
Awesome. Have you an idea of where you'd like to contribute?
sbs2001 (Thu, 02 Apr 2020 10:22:57 GMT):
https://github.com/hyperledger/besu/issues/490 (labelled as good first issue) is the specific issue I wanted to work on , APIs are something I have worked on past so I thought contributing to json-rpc client would be good starting point.
faraggi (Thu, 02 Apr 2020 11:16:17 GMT):
Concerning Issue#490, After looking at these docs, I believe you shouldn't get the miner's Id, but rather submit a random ID string.
https://github.com/ethereum/EIPs/blob/master/EIPS/eip-1474.md#eth_submithashrate
https://github.com/ethereum/wiki/wiki/JSON-RPC#eth_submithashrate
https://github.com/ethereum/go-ethereum/blob/a91b704b013aaabb3bfc796f5a3d8c940dfb7f05/consensus/ethash/api.go#L86
```
// SubmitHashrate can be used for remote miners to submit their hash rate.
// This enables the node to report the combined hash rate of all miners
// which submit work through this node.
//
// It accepts the miner hash rate and an identifier which must be unique
// between nodes.
```
NicolasMassart (Thu, 02 Apr 2020 12:36:53 GMT):
Hi all, I propose to add Karim @matkt as a Besu maintainer: https://github.com/hyperledger/besu/pull/640
NicolasMassart (Fri, 03 Apr 2020 14:43:42 GMT):
We reached the absolute majority (12 on 23). Who would like to merge and add Karim to the Github maintainers group?
shemnon (Fri, 03 Apr 2020 16:45:59 GMT):
I think we should hold the vote open at least three days to give everyone a business day to review.
hyperledger-bot (Fri, 03 Apr 2020 17:44:29 GMT):
unitTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/11636`
sumaids (Sat, 04 Apr 2020 10:45:23 GMT):
Has joined the channel.
sumaids (Sat, 04 Apr 2020 10:45:25 GMT):
#besu-contributors
jramps9 (Tue, 07 Apr 2020 15:52:15 GMT):
Has joined the channel.
jramps9 (Tue, 07 Apr 2020 15:52:16 GMT):
Hello Besu contributors! Friendly reminder to pls join us for the Marketing Committee-DevRel call tomorrow, 4/8 at 9am PT. This is a great opportunity to learn how you can get involved in many aspects of Hyperledger marketing including overall messaging, events, content, social media, PR etc. Feel free to add items to the agenda: https://wiki.hyperledger.org/x/_QjcAQ Hope to see you there :slight_smile:
kumar89 (Tue, 07 Apr 2020 15:57:35 GMT):
Has joined the channel.
rjones (Tue, 07 Apr 2020 18:37:21 GMT):
I got email from Circle CI that soon they will start expiring artifacts - does Besu rely on older artifacts stored on CCI infra?
rjones (Tue, 07 Apr 2020 18:39:10 GMT):
```The two changes you can expect to see are:
Build artifacts redirects: we are making non-HTML build artifacts on CircleCI redirect to short-lived pre-signed cloud storage URLs to improve the performance of our artifacts hosting. You should not notice any changes and no action is required.
Build artifacts storage policy: beginning on April 30th, we will start automatically deleting build artifacts that are older than 30 days for paid & free accounts. To download your older build artifacts before then, follow these instructions.
Note that this policy change will not impact other CircleCI output data such as caches, workspaces, and logs.```
shemnon (Tue, 07 Apr 2020 19:13:01 GMT):
I think we push the ones we care about onto artifactory/jfrog.
NicolasMassart (Wed, 08 Apr 2020 11:13:47 GMT):
Hi everyone, reminder that I will ask for merging this PR today EOD. So if you have any comment to make on it, please do. Thanks.
rjones (Wed, 08 Apr 2020 22:05:52 GMT):
New repo https://github.com/hyperledger/besu-native created per @benjamincburns
EdJoJob (Wed, 08 Apr 2020 22:06:17 GMT):
We are releasing besu 1.4.3
benjamincburns (Wed, 08 Apr 2020 22:21:33 GMT):
Thanks again, @rjones!
timbeiko (Thu, 09 Apr 2020 19:47:20 GMT):
:besu: We just released a new version of Besu, 1.4.3 :besu:
It now supports the eth/65 protocol and exposes re-orgs and transaction revert reasons via our Plugin API. You can get it here: https://github.com/hyperledger/besu/releases/tag/1.4.3
NicolasMassart (Fri, 10 Apr 2020 09:20:05 GMT):
Now that we pushed Besu 1.4.3 release, I think it's time to merge Karim's maintainer PR.
NicolasMassart (Fri, 10 Apr 2020 11:32:28 GMT):
I updated the PR with master and the unit tests are not passing: https://app.circleci.com/pipelines/github/hyperledger/besu/2551/workflows/87f10b76-af02-45b4-89bd-9f42dc3772ae/jobs/12481/tests
It there something I should do? I just modified a markdown file in my PR...
NicolasMassart (Fri, 10 Apr 2020 13:57:02 GMT):
PR https://github.com/hyperledger/besu/pull/640 is merged. Can an admin add Karim to the maintainers in Github please?
timbeiko (Sat, 11 Apr 2020 14:01:36 GMT):
I'm not sure if this is a calendar mistake on my part, but I have an invite for both the APAC/AMER and EMEA/AMER contributor call next week. Which one is happening? Both?
shemnon (Sat, 11 Apr 2020 16:53:50 GMT):
AMER. We did another shift with HLGF/EthCC. Perhaps we should discuss skips ont he next call, do we shift the schedule or just miss a month for one geography?
shemnon (Sat, 11 Apr 2020 16:54:01 GMT):
In the event we have to cancel meetings.
ajsutton (Sun, 12 Apr 2020 04:14:53 GMT):
Just missing a month and keeping the schedule consistent would be my vote. My calendar has wound up confused as well.
ajsutton (Sun, 12 Apr 2020 22:27:12 GMT):
We seem to have some intermittency in unit tests around EIP1559. https://app.circleci.com/pipelines/github/hyperledger/besu/2555/workflows/e7df6869-7e8b-4b92-ab78-0ae0cd8b552b/jobs/12502/
@abdelabdelhamidbakhta any chance there’s a static variable controlling 1559 somewhere that tests are toggling and interfering with each other?
ajsutton (Sun, 12 Apr 2020 22:55:54 GMT):
Yep, at least TransactionEIP1559Test is setting ExperimentalEIPs.eip1559Enabled and not restoring the original value afterwards.
ajsutton (Sun, 12 Apr 2020 23:14:01 GMT):
https://github.com/hyperledger/besu/pull/700 should fix it. If someone could review that and merge it should get our master branch build passing again.
shemnon (Mon, 13 Apr 2020 01:24:58 GMT):
approved and merged
shemnon (Mon, 13 Apr 2020 01:32:24 GMT):
Also, my attempt to make sure we only got hyperledger CI failures from the chatbot failed, so I rolled back the chatbot change. @rjones is there a deploy step needed?
timbeiko (Mon, 13 Apr 2020 13:41:52 GMT):
@shemnon I you mean EMEA/AMER, and not APAC/AMER, correct?
shemnon (Mon, 13 Apr 2020 15:17:16 GMT):
Uhh.. yea, it's always AMER I guess. EMEA/AMER.
shemnon (Mon, 13 Apr 2020 15:21:44 GMT):
Besu Contributor Call Tomorrow - https://wiki.hyperledger.org/display/BESU/2020-04-14+Besu+Contributor+Call - EMEA friendly times
rjones (Mon, 13 Apr 2020 15:43:02 GMT):
@shemnon heroku deploys on merge, so no.
shemnon (Mon, 13 Apr 2020 16:23:43 GMT):
review request - https://github.com/hyperledger/besu/pull/701 - this brings the execution time for reference tests down from 13:56 to 2:33 by disabling tests designed to consume a lot of resources (https://circleci.com/gh/hyperledger/besu/12515 vs https://circleci.com/gh/hyperledger/besu/12529)
benjamincburns (Mon, 13 Apr 2020 20:39:04 GMT):
another review request: https://github.com/hyperledger/besu-native/pull/2 - creates the linux build libsecp256k1 - native-only at this point. @shemnon will be adding additional steps to create a JNI wrapper and to package & distribute that.
EdJoJob (Mon, 13 Apr 2020 21:44:34 GMT):
@rjones could you please add `matkt` as a member of the hyperledger github org
benjamincburns (Mon, 13 Apr 2020 21:51:17 GMT):
@shemnon - as discussed OOB, here's a PR for the readme and the small fix to the `build.sh` script: https://github.com/hyperledger/besu-native/pull/3
shemnon (Tue, 14 Apr 2020 00:44:31 GMT):
@benjamincburns review request - first gradle bits for the native libs - https://github.com/hyperledger/besu-native/pull/4
faraggi (Tue, 14 Apr 2020 08:44:30 GMT):
:mega:
Besu Contributor Call Today - https://wiki.hyperledger.org/display/BESU/2020-04-14+Besu+Contributor+Call - EMEA friendly times
(T-7hs)
8am Tuesday (3 Mar) Los Angeles
11 am Tuesday (3 Mar) New York
3pm Tuesday (3 Mar) UTC
4 pm Tuesday (3 Mar) Paris/Berlin
1 am Wednesday (4 Mar) Brisbane
:mega:
NicolasMassart (Tue, 14 Apr 2020 15:48:45 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=G8xsXDXMdFY8oKTeN) Could we have a calendar in the Besu Confluence so we could sync with our own calendars and not miss a call and make sure we have the right time and date? I did not find one available in the templates but maybe it's just a matter of activating one? I don't know what's included with the Confluence license we have here.
NicolasMassart (Tue, 14 Apr 2020 15:49:03 GMT):
Could we have a calendar in the Besu Confluence so we could sync with our own calendars and not miss a call and make sure we have the right time and date? I did not find one available in the templates but maybe it's just a matter of activating one? I don't know what's included with the Confluence license we have here.
NicolasMassart (Tue, 14 Apr 2020 15:59:56 GMT):
something like https://confluence.atlassian.com/teamcal/create-add-and-edit-calendars-654213559.html
faraggi (Tue, 14 Apr 2020 16:00:26 GMT):
We actually have one, but it has some difficulties, @shemnon understands this better
shemnon (Tue, 14 Apr 2020 16:01:22 GMT):
There is this - https://wiki.hyperledger.org/display/HYP/Calendar+of+Public+Meetings
shemnon (Tue, 14 Apr 2020 16:02:09 GMT):
It took quite a bit of back end work with a new groups service to even get that working. Ry and the HLP IT group have some scars from that effort.
shemnon (Tue, 14 Apr 2020 16:02:56 GMT):
That being said, the APAC/EMEA toggle won't happen again. If we cancel then it's a month off for whoever gets impacted.
NicolasMassart (Tue, 14 Apr 2020 16:05:34 GMT):
Oh, it was not because of the cancelation, it's just that I did not add it to my own calendar. The full HLF one is a bit huge for my needs to be honest and as events don't seems to be visible as recurring meetings, it means I would have to add each one individually so there's as many chances that I miss them... But I guess this is the best we could have and that I may have to organise better on my side...
soyama-dclab (Wed, 15 Apr 2020 05:28:48 GMT):
Has joined the channel.
timbeiko (Wed, 15 Apr 2020 13:30:53 GMT):
Hey all,
While I assume most people are aware of this, it hasn't ever been announced here that we have the `Help Wanted` label on Github for issues we believe would be well suited for the community to take on, you can see the full list of issues here: https://github.com/hyperledger/besu/issues?q=is%3Aopen+is%3Aissue+label%3A%22help+wanted%22
There is also a `Good First Issues` label which, as the name indicates, highlights which of these `Help Wanted` issues may be best for people with less experience on the codebase. If you have any questions about a specific one, feel free to comment here or on the issue :smile:
benjamincburns (Wed, 15 Apr 2020 19:12:28 GMT):
PR Request: https://github.com/hyperledger/besu-native/pull/5 - likely one for @shemnon
benjamincburns (Wed, 15 Apr 2020 19:12:28 GMT):
Review Request: https://github.com/hyperledger/besu-native/pull/5 - likely one for @shemnon
benjamincburns (Wed, 15 Apr 2020 19:13:03 GMT):
note - I've validated the circleci config locally, but because it's a multiplatform build I wasn't able to test it locally
shemnon (Wed, 15 Apr 2020 19:14:23 GMT):
How do we get CircleCI turned on for a repo?
benjamincburns (Wed, 15 Apr 2020 19:14:39 GMT):
@shemnon I'll see if I can't run the circleci config on my fork - please hold off on merging until I get back to you
shemnon (Wed, 15 Apr 2020 19:14:52 GMT):
:+1:
benjamincburns (Wed, 15 Apr 2020 19:15:04 GMT):
also I think I Yeti can enable CircleCI for this repo - I'll get back to you about that too if that's not the case
benjamincburns (Wed, 15 Apr 2020 19:20:02 GMT):
@shemnon looks like there's some kind of error in the build script - I'll fix it and push whatever changes are necessary back to that PR
benjamincburns (Wed, 15 Apr 2020 19:20:33 GMT):
I'll also update the PR with a link to my test builds
benjamincburns (Wed, 15 Apr 2020 19:34:52 GMT):
looks like I don't have access to enable CircleCI for [the besu-native repo](https://github.com/hyperledger/besu-native), and I can't test the mac side of this build on my fork without a paid CircleCI account... I think we might need an assist from @rjones here
benjamincburns (Wed, 15 Apr 2020 19:35:20 GMT):
that said, I think I've identified the problem and things will work once CircleCI is enabled
benjamincburns (Wed, 15 Apr 2020 19:36:50 GMT):
oh, wait - I should've forked to the `PegaSysEng` github org - I think I can enable CircleCI and test there
benjamincburns (Wed, 15 Apr 2020 23:42:35 GMT):
Review request: https://github.com/hyperledger/besu-native/pull/6 -- same changes as earlier, but tested this time ;-)
benjamincburns (Wed, 15 Apr 2020 23:42:54 GMT):
@shemnon - not sure if you can approve that or not, since it contains a commit that you authored?
shemnon (Wed, 15 Apr 2020 23:48:07 GMT):
as long as we approve each others changes we are fine.
shemnon (Thu, 16 Apr 2020 00:55:14 GMT):
review request: need to change mac lib name: https://github.com/hyperledger/besu-native/pull/7
benjamincburns (Thu, 16 Apr 2020 01:06:13 GMT):
review request: https://github.com/hyperledger/besu-native/pull/8
rjones (Thu, 16 Apr 2020 15:18:46 GMT):
@benjamincburns is everything working OK with Circle CI, or do I need to do something else?
shemnon (Thu, 16 Apr 2020 15:21:00 GMT):
Looks like it worked. We got the code published to bintray as expected.
NicolasMassart (Thu, 16 Apr 2020 19:22:12 GMT):
@rjones do you have some news on @matkt maintainership please?
madelinemurray (Thu, 16 Apr 2020 23:15:21 GMT):
There's been an issue identified with the 1.4.3 release - I've updated the changelog to include this. Could someone review: https://github.com/hyperledger/besu/pull/734
atoulme (Fri, 17 Apr 2020 00:06:58 GMT):
reviewed and approved
atoulme (Fri, 17 Apr 2020 00:07:37 GMT):
that said the tidbit on compatibility is not as dire as it looks - it's just new nodes constantly throw exceptions when they communicate with older nodes. It's enough of a deterrent to call this incompatible.
atoulme (Fri, 17 Apr 2020 00:09:06 GMT):
On that note, this calls for an acceptance test that mimicks the mainnet. Say 100 tx per second, 3 besu nodes, and 100 peers connections and disconnections every minute or so might stress the system just like mainnet does.
madelinemurray (Fri, 17 Apr 2020 00:18:19 GMT):
Thanks @atoulme
madelinemurray (Fri, 17 Apr 2020 00:59:32 GMT):
Added a section to the Release Remediation Wiki page to include the situation where there's a significant issue with the release but no data corruption or security issue - https://wiki.hyperledger.org/display/BESU/Release+Remediation
hyperledger-bot (Fri, 17 Apr 2020 16:41:08 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/13366`
shemnon (Fri, 17 Apr 2020 17:14:01 GMT):
DCO failure. I force pushed 5e194d4013a42d56da43c540cf45d7cbea9256b2 off of master to add the needed signed-off-by line
atoulme (Sat, 18 Apr 2020 22:48:32 GMT):
Would anybody be available to review this PR? https://github.com/hyperledger/besu/pull/715
atoulme (Sat, 18 Apr 2020 22:48:54 GMT):
This stems from my discussion with @ajsutton earlier. The idea is to provide a good healthcheck for besu.
atoulme (Sat, 18 Apr 2020 22:49:06 GMT):
I modeled it against the typical linux service where pid is written to file.
atoulme (Sat, 18 Apr 2020 22:49:42 GMT):
Maybe more important to me, this PR adds splunk logging to Besu out of the box. https://github.com/hyperledger/besu/pull/725
atoulme (Sat, 18 Apr 2020 22:50:01 GMT):
Please note I have an other PR against go-ethereum that also adds splunk logging (it's been approved!)
shemnon (Sun, 19 Apr 2020 06:16:42 GMT):
I thought we had some people assigned to review those this week, I'll ask again in my monday standup. We've had a few fires to chase down the past couple of days.
atoulme (Sun, 19 Apr 2020 23:02:32 GMT):
Healthcheck just came in. One to go with splunk logging. Thanks for the reviews folks.
atoulme (Mon, 20 Apr 2020 16:23:05 GMT):
Anybody available to review https://github.com/hyperledger/besu/pull/725 ?
shemnon (Mon, 20 Apr 2020 17:03:52 GMT):
Please give us until the end of the business day. There's a lot going on at the moment.
shemnon (Mon, 20 Apr 2020 17:03:52 GMT):
Please give us until the end of the business day. There's a lot going on at the moment. We haven't had our standup yet.
shemnon (Mon, 20 Apr 2020 22:54:13 GMT):
Just now finishing my "must ship" tasks for this weeks release. I'll get a look at it tonight and make sure it runs w/o any of the vars set.
hyperledger-bot (Wed, 22 Apr 2020 00:40:19 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/13788`
hyperledger-bot (Wed, 22 Apr 2020 04:10:44 GMT):
assemble for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/13834`
shemnon (Wed, 22 Apr 2020 04:14:52 GMT):
This was a flakey test. Re-running it worked.
`org.hyperledger.besu.tests.acceptance.permissioning.NodeLocalAndOnChainPermissioningAcceptanceTest > testNodeCannotConnectWhenWhiteListedOnChainButNotLocal`
EdJoJob (Wed, 22 Apr 2020 04:17:01 GMT):
There was a DCO failure on master. Please rebase away if you have `772d48db553576776d09612bb3123f9bd24f35fc` in your history
EdJoJob (Wed, 22 Apr 2020 22:37:06 GMT):
Release 1.4.4 is in progress
EdJoJob (Wed, 22 Apr 2020 23:58:48 GMT):
The technical release of 1.4.4 is complete. Burn-in testing is underway
hyperledger-bot (Thu, 23 Apr 2020 18:08:55 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/14187`
shemnon (Thu, 23 Apr 2020 22:26:30 GMT):
flaky test. Re-running it fixed it.
brotchain (Fri, 24 Apr 2020 10:35:39 GMT):
Has joined the channel.
macfarla (Mon, 27 Apr 2020 00:15:28 GMT):
PicoCLI has the notion of `fallbackValue` https://picocli.info/#fallbackValue-annotation which we are not using in Besu. If an option is specified without a value on the command line (ie arity 0..1), it is assigned the `fallbackValue`. It's different to `defaultValue` (which is assigned if the option is not specified at all on the command line). Without using the ~fallbackValue~, there is some unexpected behaviour eg boolean attributes will be toggled if specified more than once on the command line. In Besu we currently get around this by custom code that throws a ParameterException if an option is specified more than once on the command line and/or in config files.
I propose
* adding `fallbackValue` to any CLI options that have arity of 0..1
* any boolean options that currently have arity of 1 can be changed to arity of 0..1 with a `fallbackValue`
The custom code that prevents options being specified multiple times could be removed, but this requires discussion since it is changing existing behaviour.
shemnon (Mon, 27 Apr 2020 02:15:52 GMT):
Where is this code that throws an exception if an option is specified more than once?
shemnon (Mon, 27 Apr 2020 02:20:43 GMT):
Besu contributor call Tomorrow - APAC/AMER friendly times (1am UTC) - https://wiki.hyperledger.org/display/BESU/2020-04-28+Contributor+call
shemnon (Mon, 27 Apr 2020 02:20:57 GMT):
Feel free to add your agenda items to the wiki page.
shemnon (Mon, 27 Apr 2020 02:21:14 GMT):
Where is this code that throws an exception if an option is specified more than once?
macfarla (Mon, 27 Apr 2020 05:44:29 GMT):
Apologies, it is not Besu code that does this. PicoCLI prevents the options being specified more than once on the command line. However, it is possible to specify an option on the command line and in the config file which is where confusion can arise.
shemnon (Mon, 27 Apr 2020 15:32:02 GMT):
Maybe a fix would be in order for the toml parser to do something like prevent boolean double flips, throw an error if an option is both on the CLI and in the config file, or at least a warning.
Fixing it in the toml defaults would fix it for future booleans whereas going to fallback only fixes it for curent booleans and we may make the same mistake in the future.
shemnon (Mon, 27 Apr 2020 15:32:34 GMT):
Can you present a concrete example of the problem as it currently exists?
shemnon (Mon, 27 Apr 2020 20:34:24 GMT):
zoom link for the contributor call - https://consensys.zoom.us/j/524697595 - (it wasn't on the wiki page when I first posted)
macfarla (Mon, 27 Apr 2020 20:46:06 GMT):
If I run `besu --config-file=config.toml --rpc-http-enabled`
where config.toml has `rpc-http-enabled=true`
besu starts with rpc disabled
shemnon (Mon, 27 Apr 2020 21:53:56 GMT):
Yesh... this is a PicoCLI usability surprise. We get the double flip because we do TOML as though they were defaults. But this also impacts us if we ever want boolean flags to default to true. We dodged this with Jake's CLI standards where all boolean flags are -enable only.
shemnon (Mon, 27 Apr 2020 21:54:46 GMT):
Separate design issue: if there is a CLI flag and a config file value, do we want to allow it to happen with CLI winning? Should we throw an error or log a warning if a flag is specified in both places?
shemnon (Mon, 27 Apr 2020 21:55:37 GMT):
We also probably want an error-prone check to deal with the boolean flag surprises. We can't be sure the next engineer writing this will know those issues.
ajsutton (Mon, 27 Apr 2020 22:26:23 GMT):
The priority is CLI arg, then system env var and then config file. So it can be useful to have a config file that sets `p2p-enabled=false` and then be able to override that with `--p2p-enabled` on the CLI.
Currently you can't do that and have to add the boolean explicitly (`--p2p-enabled=true`).
hyperledger-bot (Mon, 27 Apr 2020 23:53:38 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/14613`
usmansaleem (Tue, 28 Apr 2020 00:26:00 GMT):
Re-run the acceptance test fixed the issue. There are few intermittent failure acceptance tests.
shemnon (Tue, 28 Apr 2020 00:30:36 GMT):
The native libraries have a startup penalty. I'm contemplating turning them off for acceptance tests and relying on the unit tests for validaiton.
usmansaleem (Tue, 28 Apr 2020 00:31:11 GMT):
:+1:
madelinemurray (Tue, 28 Apr 2020 00:34:45 GMT):
📣 We just released a new version of Besu, 1.4.4 📣
1.4.4 includes logs and filters for private contracts, EIP-2315, Splunk logging integration, and optional native library encryption.
We've disabled eth/65 support by default and performed additional testing on 1.4.4 to ensure the issues with 1.4.3 are no longer occurring.
Changes and known issues are listed in the Changelong - https://github.com/hyperledger/besu/blob/master/CHANGELOG.md
hyperledger-bot (Tue, 28 Apr 2020 00:38:47 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/14632`
hyperledger-bot (Tue, 28 Apr 2020 01:02:14 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/14643`
stone-ch (Tue, 28 Apr 2020 01:15:43 GMT):
Has joined the channel.
EdJoJob (Tue, 28 Apr 2020 01:35:47 GMT):
There are two proposals currently under consideration for besu:
https://wiki.hyperledger.org/display/BESU/Proposal%3A+Feature+Flags
https://wiki.hyperledger.org/display/BESU/Proposal%3A+Create+a+Release+Candidate+for+every+release
Please comment on the wiki pages for feedback.
macfarla (Tue, 28 Apr 2020 20:00:43 GMT):
if we add a `fallbackValue` we can get the behaviour that Adrian describes.
K2LIS (Tue, 28 Apr 2020 20:46:54 GMT):
Has joined the channel.
shemnon (Tue, 28 Apr 2020 22:00:50 GMT):
One thing to watch out for is what does this do to our auto-help text. Will it add =
ajsutton (Tue, 28 Apr 2020 22:12:09 GMT):
Interestingly besu is currently inconsistent (`--p2p-enabled=
macfarla (Tue, 28 Apr 2020 23:31:17 GMT):
just checked. `fallbackValue` does not appear in the generated help text
macfarla (Tue, 28 Apr 2020 23:31:17 GMT):
ah ok but changing the arity does affect it
macfarla (Tue, 28 Apr 2020 23:31:17 GMT):
ah ok changing the arity does affect it
macfarla (Wed, 29 Apr 2020 00:47:52 GMT):
so if we add fallbackValue=true for all "enabled" booleans (so that they read semantically) and this adds `[=
usmansaleem (Wed, 29 Apr 2020 01:29:59 GMT):
Review request. https://github.com/hyperledger/besu/pull/808
usmansaleem (Wed, 29 Apr 2020 01:29:59 GMT):
Review request. https://github.com/hyperledger/besu/pull/808 (remove references of system property: besu.docker)
PaulHarris (Wed, 29 Apr 2020 02:18:01 GMT):
to me having the fallback value and arity 0..1 makes sense, then the enabled flags always fall back to true if just the flag is specified. hopefully the help generated isn't too bad, doesn't seem major.
hyperledger-bot (Wed, 29 Apr 2020 15:32:27 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/14858`
shemnon (Wed, 29 Apr 2020 16:48:33 GMT):
review request - library version updates - https://github.com/hyperledger/besu/pull/812
shemnon (Wed, 29 Apr 2020 16:49:45 GMT):
Again, a flakey test.
peter_somogyvari (Thu, 30 Apr 2020 19:35:16 GMT):
Has joined the channel.
peter_somogyvari (Thu, 30 Apr 2020 19:35:58 GMT):
Reposting here after realizing that there's a channel specifically for contributions:
> I dropped a PR just now. Happy to chat here about it as well not just the GH comments (in case anybody prefers one over the other).
> https://github.com/hyperledger/besu/pull/821
shemnon (Thu, 30 Apr 2020 20:07:41 GMT):
Review Request: Gradle 6.3 upgrade (supports Java 14) - https://github.com/hyperledger/besu/pull/822
shemnon (Thu, 30 Apr 2020 22:01:06 GMT):
Review Request: make loggers private, static, and final, enforced by errorprone. https://github.com/hyperledger/besu/pull/823
hyperledger-bot (Thu, 30 Apr 2020 23:08:41 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15106`
hyperledger-bot (Fri, 01 May 2020 00:05:22 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15115`
hyperledger-bot (Sat, 02 May 2020 15:02:54 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15195`
hyperledger-bot (Sat, 02 May 2020 16:43:31 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15197`
hyperledger-bot (Sun, 03 May 2020 01:57:27 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15233`
hyperledger-bot (Sun, 03 May 2020 23:55:44 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15246`
hyperledger-bot (Tue, 05 May 2020 00:49:10 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15386`
hyperledger-bot (Tue, 05 May 2020 06:51:03 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15426`
arash009 (Tue, 05 May 2020 07:19:48 GMT):
Hi all. Besu 1.4.5 is scheduled for release in less than 48hrs from now. If you have anything in flight it would be good to get it in otherwise it will be another 2 weeks for next release.
hyperledger-bot (Tue, 05 May 2020 13:37:09 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15482`
hyperledger-bot (Wed, 06 May 2020 07:01:11 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15540`
hyperledger-bot (Wed, 06 May 2020 16:57:58 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15619`
hyperledger-bot (Wed, 06 May 2020 17:12:24 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15623`
hyperledger-bot (Wed, 06 May 2020 21:06:32 GMT):
publish for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/15686`
joshuafernandes (Wed, 06 May 2020 23:50:00 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
madelinemurray (Thu, 07 May 2020 00:42:05 GMT):
Based on the string of test failures above, we're going to postpone this release while we look into what's happening.
madelinemurray (Thu, 07 May 2020 05:57:16 GMT):
Have raised an issue to track the test flakiness issue - https://github.com/hyperledger/besu/issues/864
madelinemurray (Thu, 07 May 2020 21:42:36 GMT):
There's more information in the issue above but based on investigation we're going to do the release planned for yesterday soon.
hyperledger-bot (Thu, 07 May 2020 22:58:38 GMT):
buildDocker for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16005`
hyperledger-bot (Thu, 07 May 2020 22:59:22 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16002`
joshuafernandes (Fri, 08 May 2020 00:20:42 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
peter_somogyvari (Fri, 08 May 2020 00:21:05 GMT):
Has left the channel.
joshuafernandes (Fri, 08 May 2020 01:22:16 GMT):
@here: is now complete
hyperledger-bot (Fri, 08 May 2020 02:53:46 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16091`
hyperledger-bot (Fri, 08 May 2020 03:02:31 GMT):
unitTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16099`
hyperledger-bot (Fri, 08 May 2020 03:15:48 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16104`
hyperledger-bot (Fri, 08 May 2020 03:25:59 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16107`
hyperledger-bot (Fri, 08 May 2020 03:35:13 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16108`
hyperledger-bot (Fri, 08 May 2020 03:46:39 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16109`
hyperledger-bot (Fri, 08 May 2020 04:00:42 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16112`
hyperledger-bot (Fri, 08 May 2020 13:15:31 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16160`
atoulme (Fri, 08 May 2020 19:58:31 GMT):
I am following the chat around discovery ; FWIW I implemented DNS-based discovery in Apache Tuweni. After its next release, I'd like to propose to add dns discovery to Besu using that module.
shemnon (Fri, 08 May 2020 21:10:08 GMT):
Are there hedges against DNS spoofing?
shemnon (Fri, 08 May 2020 21:10:15 GMT):
i.e. DNSSec support?
atoulme (Fri, 08 May 2020 22:06:09 GMT):
dnsjava support DNSSec, so I can code to piggyback on that.
atoulme (Fri, 08 May 2020 22:08:08 GMT):
can also use DoH - DNS over HTTP
atoulme (Fri, 08 May 2020 22:21:09 GMT):
bring in the wishlist @shemnon
shemnon (Fri, 08 May 2020 22:29:09 GMT):
Do any other ethereum clients do DNS discovery? i.e. do we have to interop with other clients or is this something we would be starting on our own?
atoulme (Fri, 08 May 2020 22:59:30 GMT):
geth does, yes
atoulme (Fri, 08 May 2020 22:59:51 GMT):
this is all documented in https://github.com/ethereum/EIPs/blob/master/EIPS/eip-1459.md
atoulme (Fri, 08 May 2020 23:00:43 GMT):
the list of nodes is there: https://github.com/ethereum/discv4-dns-lists
hyperledger-bot (Sat, 09 May 2020 00:42:33 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16267`
shemnon (Sat, 09 May 2020 00:53:26 GMT):
Open an issue on github so we can have a discussion there without build failures getting in the way ;)
hyperledger-bot (Sat, 09 May 2020 14:52:27 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16284`
bmg (Sat, 09 May 2020 15:09:20 GMT):
Hi all. I've found an edge case that the FlatTraceGenerator doesn't handle. I want to put together a test case to demonstrate it before raising the issue. I'm trying to figure out how the test blocks are generated for the tests so I can add an extra one to the test data export. Can someone point me in the right direction?
hyperledger-bot (Sat, 09 May 2020 15:13:51 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16289`
hyperledger-bot (Sat, 09 May 2020 16:12:37 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16293`
hyperledger-bot (Sat, 09 May 2020 16:12:40 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16294`
hyperledger-bot (Sat, 09 May 2020 16:15:04 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16295`
hyperledger-bot (Sat, 09 May 2020 16:17:20 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16296`
hyperledger-bot (Sat, 09 May 2020 16:25:17 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16303`
hyperledger-bot (Sat, 09 May 2020 16:26:52 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16305`
hyperledger-bot (Mon, 11 May 2020 01:35:28 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16363`
hyperledger-bot (Mon, 11 May 2020 01:45:32 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16367`
hyperledger-bot (Mon, 11 May 2020 15:43:52 GMT):
unitTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16402`
hyperledger-bot (Mon, 11 May 2020 18:53:43 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16423`
hyperledger-bot (Mon, 11 May 2020 18:53:54 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16422`
shemnon (Mon, 11 May 2020 19:07:06 GMT):
Besu contributor call in ~19 hours, EMEA/AMER friendly time - https://wiki.hyperledger.org/display/BESU/2020-05-12+Contributor+Call
hyperledger-bot (Mon, 11 May 2020 22:47:08 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16493`
hyperledger-bot (Mon, 11 May 2020 23:03:43 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16498`
lucassaldanha (Mon, 11 May 2020 23:16:00 GMT):
Hi folks, I have added a page on our wiki with a suggested Bug Triage process. For a while we have had the policy around how we assign a priority to a bug, but we have never started the process on doing it and communicating it properly. Please let me know what you think!
https://wiki.hyperledger.org/display/BESU/Bug+Triage+Process
shemnon (Tue, 12 May 2020 14:10:57 GMT):
Besu contributor call in > 1hour, EMEA/AMER friendly time - https://wiki.hyperledger.org/display/BESU/2020-05-12+Contributor+Call
hyperledger-bot (Tue, 12 May 2020 14:37:38 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16563`
shemnon (Tue, 12 May 2020 14:55:33 GMT):
https://consensys.zoom.us/j/199741148
hyperledger-bot (Tue, 12 May 2020 17:35:12 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16618`
hyperledger-bot (Tue, 12 May 2020 20:42:37 GMT):
assemble for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16644`
jramps9 (Wed, 13 May 2020 12:11:06 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT TODAY, 5/13! Take a look at the agenda and add items here: https://wiki.hyperledger.org/display/Marketing/2020-05-13+Meeting+notes
hyperledger-bot (Wed, 13 May 2020 15:23:13 GMT):
publish for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16735`
hyperledger-bot (Wed, 13 May 2020 19:22:31 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16782`
hyperledger-bot (Wed, 13 May 2020 19:37:35 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16790`
hyperledger-bot (Wed, 13 May 2020 19:38:43 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16791`
hyperledger-bot (Wed, 13 May 2020 19:51:04 GMT):
acceptanceTests for besu - release-1.4.5 : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16793`
hyperledger-bot (Wed, 13 May 2020 20:01:36 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16794`
arash009 (Wed, 13 May 2020 22:17:45 GMT):
Hi all. As we are still in the process of testing release 1.4.5-RC1, and the next release being due next week, I propose we cancel the actual 1.4.5 release, and move to 1.4.6 in the next release window which is in ~1 week. Any thoughts?
ajsutton (Wed, 13 May 2020 22:21:05 GMT):
I agree with scrapping this release, but since we haven't actually build a 1.4.5 I feel like the next release would still be 1.4.5 - just starting from a newer commit from master (and -RC2). Although, I don't think it really matters if we skip a version number.
arash009 (Wed, 13 May 2020 22:27:19 GMT):
Yeah what I'm unsure about is having an RC2 that is off of master, and not off of the RC1 branch, seems a bit at odds. If we going back to master I'd be tempted to skip the version to show the distinction
shemnon (Wed, 13 May 2020 23:14:01 GMT):
I think going RC 2 off of master is fine. I don't think RC2 always implies it is just deltas from RC1.
shemnon (Wed, 13 May 2020 23:14:33 GMT):
Plus then we will get questions from the field about "Where's 1.4.5? I need it for my collection"
benjamincburns (Wed, 13 May 2020 23:19:06 GMT):
my experience in this area is that the dev team will argue about release numbering far far more than users will care about it
benjamincburns (Wed, 13 May 2020 23:20:03 GMT):
and in this case since we're not breaking semver (does besu follow semver, anyway?) it seems really moot
rjones (Wed, 13 May 2020 23:55:46 GMT):
@benjamincburns all of our projects must follow semver. there was a regression in Fabric where there was a breaking change across a minor version, but AFAIK that's the only time that's happened
lucassaldanha (Thu, 14 May 2020 02:53:04 GMT):
Hey folks, I noticed that when we send a tx with the wrong chainId, Besu returns a generic "Invalid Params" (not super helpful). Is there a reason why we don't map the WRONG_CHAIN_ID tx validation result to a specific RPC error? Or did we simply forget to map this error? I'll push a fix for it but wanted to check if we have a specific reason for not doing this.
shemnon (Thu, 14 May 2020 05:10:35 GMT):
We're possibly swallowing the reason somewhere and throwing a generic failure. Because it's a signing error we may need to add some way to tease that detail out.
shemnon (Thu, 14 May 2020 05:10:42 GMT):
But worthey
hyperledger-bot (Thu, 14 May 2020 14:48:26 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16870`
hyperledger-bot (Thu, 14 May 2020 14:49:21 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16871`
hyperledger-bot (Thu, 14 May 2020 14:51:31 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16872`
hyperledger-bot (Thu, 14 May 2020 14:57:07 GMT):
acceptanceTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16877`
hyperledger-bot (Thu, 14 May 2020 14:57:09 GMT):
acceptanceTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16876`
hyperledger-bot (Thu, 14 May 2020 14:57:13 GMT):
acceptanceTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16875`
hyperledger-bot (Thu, 14 May 2020 14:57:16 GMT):
acceptanceTests for besu - master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16874`
hyperledger-bot (Thu, 14 May 2020 15:00:03 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16873`
hyperledger-bot (Thu, 14 May 2020 15:07:39 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16879`
hyperledger-bot (Thu, 14 May 2020 15:12:28 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16880`
hyperledger-bot (Thu, 14 May 2020 19:43:37 GMT):
integrationTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16929`
hyperledger-bot (Thu, 14 May 2020 19:43:37 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16932`
hyperledger-bot (Thu, 14 May 2020 19:43:37 GMT):
unitTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16931`
hyperledger-bot (Thu, 14 May 2020 20:29:01 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16938`
hyperledger-bot (Thu, 14 May 2020 20:39:23 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16941`
hyperledger-bot (Thu, 14 May 2020 20:39:24 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16940`
hyperledger-bot (Thu, 14 May 2020 20:39:26 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16942`
hyperledger-bot (Thu, 14 May 2020 20:39:30 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16944`
hyperledger-bot (Thu, 14 May 2020 20:39:33 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16946`
hyperledger-bot (Thu, 14 May 2020 20:39:37 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16948`
hyperledger-bot (Thu, 14 May 2020 20:39:41 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16950`
hyperledger-bot (Thu, 14 May 2020 20:39:51 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16952`
hyperledger-bot (Thu, 14 May 2020 20:39:53 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16954`
hyperledger-bot (Thu, 14 May 2020 20:57:50 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16956`
hyperledger-bot (Thu, 14 May 2020 21:24:10 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16961`
ajsutton (Thu, 14 May 2020 22:29:49 GMT):
@benjamincburns Is it possible to disable notifications from the flakiness-master branch to rocket chat? :)
hyperledger-bot (Thu, 14 May 2020 22:44:53 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16964`
benjamincburns (Thu, 14 May 2020 22:48:08 GMT):
I don't think I have access to configure that @ajsutton
benjamincburns (Thu, 14 May 2020 22:48:12 GMT):
I can check, though
benjamincburns (Thu, 14 May 2020 22:48:52 GMT):
yeah, unfortunately I don't have access for that
benjamincburns (Thu, 14 May 2020 22:49:01 GMT):
I'll take it up w/ the other yetis at our standup in 10
shemnon (Thu, 14 May 2020 22:53:16 GMT):
Script is here - https://github.com/hyperledger/hyperledger-rocket-chat-hubot/blob/master/scripts/circle-ci-notifications.coffee . I think Ry gave me commit access if someone more familir with coffescript can change the matcher to match the whole string
shemnon (Thu, 14 May 2020 22:53:45 GMT):
The issue I think is "flakiness-master" matches the "master" regex
rjones (Thu, 14 May 2020 22:53:47 GMT):
Yup. Heroku redeploys on merge, so you'll be fine
hyperledger-bot (Thu, 14 May 2020 22:55:07 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16966`
hyperledger-bot (Thu, 14 May 2020 23:06:26 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16968`
hyperledger-bot (Thu, 14 May 2020 23:11:15 GMT):
acceptanceTests for besu - flakiness-master : CANCELED. See more at `https://circleci.com/gh/hyperledger/besu/16970`
hyperledger-bot (Thu, 14 May 2020 23:23:16 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16971`
hyperledger-bot (Thu, 14 May 2020 23:36:15 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16973`
hyperledger-bot (Thu, 14 May 2020 23:47:55 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16975`
shemnon (Fri, 15 May 2020 00:05:59 GMT):
Can we get a formal 0.2.0 build of `besu-native` cut? It has the native library needed for the BLS-12 precompiles for the berlin hardfork.
hyperledger-bot (Fri, 15 May 2020 00:08:09 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16979`
hyperledger-bot (Fri, 15 May 2020 00:17:56 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16981`
hyperledger-bot (Fri, 15 May 2020 00:29:37 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16983`
hyperledger-bot (Fri, 15 May 2020 00:42:11 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16985`
hyperledger-bot (Fri, 15 May 2020 00:52:01 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16987`
hyperledger-bot (Fri, 15 May 2020 01:03:33 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16989`
hyperledger-bot (Fri, 15 May 2020 01:13:45 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16990`
EdJoJob (Fri, 15 May 2020 01:23:14 GMT):
I have https://github.com/hyperledger/hyperledger-rocket-chat-hubot/pull/38 to solve this
hyperledger-bot (Fri, 15 May 2020 01:23:23 GMT):
acceptanceTests for besu - flakiness-master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/16992`
shemnon (Fri, 15 May 2020 01:23:52 GMT):
On it's way...
shemnon (Fri, 15 May 2020 01:24:34 GMT):
Since I can't see the heroku console the only way to tell is if we get a master and fkaliness-master failures and we see what we expect to see.
usmansaleem (Fri, 15 May 2020 03:54:40 GMT):
Small PR review please. https://github.com/hyperledger/besu/pull/926 (reduce log noise by suppressing netty debug warnings)
hyperledger-bot (Fri, 15 May 2020 04:18:10 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/17033`
shemnon (Fri, 15 May 2020 04:24:06 GMT):
Looks like that did the trick.
hyperledger-bot (Fri, 15 May 2020 05:33:04 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/17047`
hyperledger-bot (Fri, 15 May 2020 13:28:09 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/17145`
hyperledger-bot (Fri, 15 May 2020 16:52:17 GMT):
acceptanceTests for besu - master : FAILED. See more at `https://circleci.com/gh/hyperledger/besu/17199`
hyperledger-bot (Fri, 15 May 2020 17:31:06 GMT):
acceptanceTests for besu - master : FAILED. See more: [https://circleci.com/gh/hyperledger/besu/17219](https://circleci.com/gh/hyperledger/besu/17219)
hyperledger-bot (Sat, 16 May 2020 05:09:20 GMT):
acceptanceTests for besu - master has recovered
hyperledger-bot (Mon, 18 May 2020 14:33:03 GMT):
acceptanceTests for besu - master : CANCELED. See more: [https://circleci.com/gh/hyperledger/besu/17360](https://circleci.com/gh/hyperledger/besu/17360)
hyperledger-bot (Tue, 19 May 2020 14:14:44 GMT):
assemble for besu - master : FAILED. See more: [https://circleci.com/gh/freighttrust/besu/2](https://circleci.com/gh/freighttrust/besu/2)
hyperledger-bot (Tue, 19 May 2020 14:30:09 GMT):
assemble for besu - master has recovered
benjamincburns (Tue, 19 May 2020 18:42:12 GMT):
this isn't our build... perhaps the webhook needs to be obfuscated and moved to a Circle environment var?
rjones (Tue, 19 May 2020 19:06:57 GMT):
I know an update to the regex here was tried once: https://github.com/hyperledger/hyperledger-rocket-chat-hubot/blob/master/scripts/circle-ci-notifications.coffee
benjamincburns (Tue, 19 May 2020 19:07:43 GMT):
ah, is hubot the recipient of the webhook? I thought it was RC getting it directly
benjamincburns (Tue, 19 May 2020 19:08:15 GMT):
doh - I suppose I should've read the URL more carefully :facepalm: https://github.com/hyperledger/besu/blob/master/.circleci/config.yml#L30
rjones (Tue, 19 May 2020 19:08:24 GMT):
https://github.com/hyperledger/hyperledger-rocket-chat-hubot/commit/81437050226911afd348ac6326f8bf85c53a1c11 was the attempt which was reverted
rjones (Tue, 19 May 2020 19:11:34 GMT):
there is an option in circleci to enable building forks - I just disabled that - should fix the problem (and stop us from spending money outside hyperledger)
benjamincburns (Tue, 19 May 2020 19:12:56 GMT):
let's see if that works. If not, I'll have a look at bringing back the regex in a way that it won't mask out our builds
benjamincburns (Tue, 19 May 2020 19:13:53 GMT):
we've set up additional build monitoring to alert the ConsenSys slack on failures to Besu and Orion, so we'll be able to check against that to be sure that things are still coming through
EdJoJob (Tue, 19 May 2020 22:51:24 GMT):
building forks is required to allow PR's to be built. Without that setting it is impossible for us to evaluate a pull-request which we require to be off a fork, considering the permissions in GitHub
EdJoJob (Tue, 19 May 2020 22:54:35 GMT):
(or at least that was the case when last I looked)
EdJoJob (Tue, 19 May 2020 22:55:14 GMT):
also this particular build was not being built under the Hyperledger CircleCI account, which is part of the reason it failed
hyperledger-bot (Tue, 19 May 2020 23:25:50 GMT):
acceptanceTests for besu - master : FAILED. See more: [https://circleci.com/gh/hyperledger/besu/17572](https://circleci.com/gh/hyperledger/besu/17572)
shemnon (Tue, 19 May 2020 23:56:44 GMT):
Review Request - rollback vertx.io version -https://github.com/hyperledger/besu/pull/951
rjones (Wed, 20 May 2020 00:03:20 GMT):
OK, I will undo that change.
rjones (Wed, 20 May 2020 00:04:30 GMT):
Done (undone)
madelinemurray (Wed, 20 May 2020 00:31:44 GMT):
In the process of testing 1.4.5-RC2, we're still experiencing issues with significant test flakiness (assuming everyone has noticed the alerts at this point) so are not planning to go ahead with the release for tomorrow. Alongside this we're also using the new proposed release process where we fastsync from scratch and monitor performance.
Once @shemnon's PR above is reviewed and merged, our proposed next steps are:
1. @joshuafernandes restarts our fastsync nodes with 1.4.5.-RC3
2. Rai rerun AT scripts to see if the flakiness has decreased
madelinemurray (Wed, 20 May 2020 00:32:21 GMT):
Based on the outcome of those, release next week.
madelinemurray (Wed, 20 May 2020 00:48:09 GMT):
@joshuafernandes - per the proposed release process we should probably also upgrade our already synced production nodes to 1.4.5-RC3 once we have it
madelinemurray (Wed, 20 May 2020 01:40:28 GMT):
@trent.mohay
trent.mohay (Wed, 20 May 2020 01:40:28 GMT):
Has joined the channel.
madelinemurray (Wed, 20 May 2020 01:55:06 GMT):
@rai
EdJoJob (Wed, 20 May 2020 02:18:44 GMT):
review-request for preventing false alerts from forks https://github.com/hyperledger/besu/pull/952
shemnon (Wed, 20 May 2020 06:19:12 GMT):
review request - improve import logging - https://github.com/hyperledger/besu/pull/957
deployer21 (Wed, 20 May 2020 18:13:00 GMT):
Has joined the channel.
diegoll (Wed, 20 May 2020 18:23:23 GMT):
Has joined the channel.
shemnon (Wed, 20 May 2020 18:47:32 GMT):
We're beginning the release process for hyperledger/besu-native, please don't merge anything to master until further notice
shemnon (Wed, 20 May 2020 19:07:38 GMT):
It's now safe to merge into master for hyperledger/besu-native again.
shemnon (Wed, 20 May 2020 19:29:58 GMT):
Re: vertx version downgrade. Some more focused internal testing shows that this may not be the main issue. 3.5.4 had 3/13 flakey tests, and 3.5.0 had 2/14. Given that there are known security advisories for 3.5.0 I vote we hold off on the downgrade and investigate further.
shemnon (Wed, 20 May 2020 19:29:58 GMT):
Re: vertx version downgrade. Some more focused internal testing shows that this may not be the main issue. 3.5.4 had 3/13 flakey runs, and 3.5.0 had 2/14. Given that there are known security advisories for 3.5.0 I vote we hold off on the downgrade and investigate further.
madelinemurray (Wed, 20 May 2020 19:35:56 GMT):
Thanks for the update @shemnon. So we should move ahead with restarting the fastsyncs compared to 1.4.2 without waiting for this merge by the sounds of it?
madelinemurray (Wed, 20 May 2020 19:36:48 GMT):
@timbeiko @EdJoJob @joshuafernandes @benjamincburns
madelinemurray (Wed, 20 May 2020 19:37:40 GMT):
Trying again to tag @shemnon
shemnon (Wed, 20 May 2020 19:38:03 GMT):
Yes, once we cut RC3 from current master.
rjones (Wed, 20 May 2020 20:50:46 GMT):
shemnon
joshuafernandes (Wed, 20 May 2020 21:13:56 GMT):
Sounds good :thumbsup: I'll tag 417e253 as the base for 1.4.5-RC3 and start the comparison off in a few minutes @shemnon @madelinemurray @EdJoJob @benjamincburns @timbeiko
madelinemurray (Wed, 20 May 2020 22:35:27 GMT):
Thanks @joshuafernandes
shemnon (Wed, 20 May 2020 23:01:26 GMT):
Fore review - BLS12-381 precompiles (EIP-2537) - https://github.com/hyperledger/besu/pull/964
shemnon (Wed, 20 May 2020 23:01:26 GMT):
For review - BLS12-381 precompiles (EIP-2537) - https://github.com/hyperledger/besu/pull/964
hyperledger-bot (Thu, 21 May 2020 01:10:23 GMT):
acceptanceTests for besu - release-1.4.5 : FAILED. See more: [https://circleci.com/gh/hyperledger/besu/17838](https://circleci.com/gh/hyperledger/besu/17838)
hyperledger-bot (Thu, 21 May 2020 02:41:55 GMT):
acceptanceTests for besu - release-1.4.5 has recovered
joshuafernandes (Thu, 21 May 2020 02:47:45 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.4.5 until further notice
hyperledger-bot (Thu, 21 May 2020 03:53:38 GMT):
acceptanceTests for besu - release-1.4.5 : FAILED. See more: [https://circleci.com/gh/hyperledger/besu/17956](https://circleci.com/gh/hyperledger/besu/17956)
hyperledger-bot (Thu, 21 May 2020 04:10:46 GMT):
acceptanceTests for besu - release-1.4.5 has recovered
hyperledger-bot (Thu, 21 May 2020 07:31:49 GMT):
acceptanceTests for besu - release-1.4.5 : FAILED. See more: [https://circleci.com/gh/hyperledger/besu/18007](https://circleci.com/gh/hyperledger/besu/18007)
joshuafernandes (Thu, 21 May 2020 11:01:36 GMT):
apologies, forgot to post back earlier - release 1.4.5 complete, feel free to merge
shemnon (Mon, 25 May 2020 01:10:04 GMT):
Besu Contributor call in just under 24 hours - https://wiki.hyperledger.org/display/BESU/2020-05-26+Contributor+Call - APAC/AMER times
shemnon (Mon, 25 May 2020 01:10:27 GMT):
Feel free to add agenda items directly to the agenda
atoulme (Tue, 26 May 2020 23:00:25 GMT):
added a couple items
atoulme (Tue, 26 May 2020 23:01:02 GMT):
oh looks like it was yesterday. Never mind.
benjamincburns (Tue, 26 May 2020 23:16:19 GMT):
seems that build failure notifications for builds on `master` aren't firing now - not sure why, but I'll have a look after standup in a bit
benjamincburns (Tue, 26 May 2020 23:47:07 GMT):
@rjones mind having a look at this when you get a sec? this is my best theory so far... https://github.com/hyperledger/hyperledger-rocket-chat-hubot/pull/40
benjamincburns (Tue, 26 May 2020 23:47:25 GMT):
also @EdJoJob if you're still around and maintaining things, please let me know what you think...
benjamincburns (Tue, 26 May 2020 23:57:23 GMT):
also we should probably investigate moving over to a more supported config for alerting webhooks. The `notify: webhooks: ...` that we're using now was actually an experimental feature in the CircleCI 1.0 config and it's no longer included in the v2.1 config that we're using now.
benjamincburns (Tue, 26 May 2020 23:57:23 GMT):
also we should probably investigate moving over to a more supported config for alerting webhooks. The `notify: webhooks: ...` that we're using now was actually an [experimental feature](https://circleci.com/docs/archive/configuration/#per-branch-notifications) in the CircleCI 1.0 config and it's no longer included in the v2.1 config that we're using now.
arash009 (Thu, 28 May 2020 05:30:50 GMT):
:besu: Besu 1.4.6-RC1 is being scheduled for tomorrow AM Australia time - Please try and have your PRs that are in flight in by then, and review anything that you've forgotten to put in the changelog!
joshuafernandes (Thu, 28 May 2020 22:45:48 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
joshuafernandes (Thu, 28 May 2020 23:17:44 GMT):
@here: It's now safe to merge into master for hyperledger/besu again
Yazanator (Fri, 29 May 2020 01:49:53 GMT):
Has joined the channel.
shemnon (Fri, 29 May 2020 01:50:48 GMT):
For keccak native, first the library needs to land here https://github.com/hyperledger/besu-native then get wired in via jna, probably like https://github.com/hyperledger/besu/blob/master/crypto/src/main/java/org/hyperledger/besu/crypto/SECP256K1.java
shemnon (Fri, 29 May 2020 01:55:02 GMT):
as for unit tests, https://github.com/hyperledger/besu/blob/master/ethereum/blockcreation/src/test/java/org/hyperledger/besu/ethereum/blockcreation/EthHashBlockCreatorTest.java tests the actual block creation
shemnon (Fri, 29 May 2020 01:56:23 GMT):
then https://github.com/hyperledger/besu/blob/master/ethereum/blockcreation/src/test/java/org/hyperledger/besu/ethereum/blockcreation/EthHashBlockCreatorTest.java with needed additions here- https://github.com/hyperledger/besu/blob/master/acceptance-tests/dsl/src/main/java/org/hyperledger/besu/tests/acceptance/dsl/node/configuration/BesuNodeConfigurationBuilder.java#L83 and here - https://github.com/hyperledger/besu/blob/master/acceptance-tests/dsl/src/main/java/org/hyperledger/besu/tests/acceptance/dsl/node/configuration/BesuNodeFactory.java#L75 for acceptance tests (likely a param for pow mining algorithm)
Yazanator (Fri, 29 May 2020 02:21:18 GMT):
cool that's awesome, this helps a lot
Yazanator (Fri, 29 May 2020 02:21:28 GMT):
i'll save the links
Yazanator (Fri, 29 May 2020 02:22:14 GMT):
thanks @shemnon
shemnon (Fri, 29 May 2020 18:25:31 GMT):
So 1.4.6-RC1 has a showstopper bug - https://github.com/hyperledger/besu/pull/1005 and I'm going to vote no-go.
shemnon (Fri, 29 May 2020 18:26:10 GMT):
Do we cherry pick the revert into what is RC1 or do we do a full scrub and re-build next wendsday? I'm up for either, just want to know what the rest of the maintainers would prefer.
shemnon (Fri, 29 May 2020 18:27:12 GMT):
(insert joke about cutting a release on friday, which we already did from an australian persoective)
shemnon (Fri, 29 May 2020 18:27:12 GMT):
(insert joke about cutting a release on friday, which we already did from an Australian perspective)
madelinemurray (Fri, 29 May 2020 20:52:59 GMT):
Which PR introduced the bug this PR fixes? How come our nightly boxes didn't show this issue? Was it only going to show when finding new peers rather than retaining existing peers?
rai (Fri, 29 May 2020 20:53:48 GMT):
Yeah, it's a finding new peers issue. It was @shemnon 's change to devp2p packet expirations
rai (Fri, 29 May 2020 20:54:15 GMT):
We actually found an existing bug though so it's actually good news!
shemnon (Fri, 29 May 2020 20:54:20 GMT):
There were _two_ bugs. Eth/64 and the devp2p packet expiration.
shemnon (Fri, 29 May 2020 20:54:25 GMT):
the expiration fix is inbound.
madelinemurray (Fri, 29 May 2020 20:54:27 GMT):
So if we cleared the peer table on our nightly upgrades then we could have found it sooner?
madelinemurray (Fri, 29 May 2020 20:54:35 GMT):
@rai - absolutely!
rai (Fri, 29 May 2020 20:55:04 GMT):
Yeah, I think we would've seen it in that case.
madelinemurray (Fri, 29 May 2020 20:55:34 GMT):
Is there an easy way to get a running node to drop all of it's peers?
shemnon (Fri, 29 May 2020 20:55:36 GMT):
For the eth/64 bug only geth and openethereum would reject us, and current versions only. Nethermind and older clients connected just fine. So Goerli has older clients that are compatible as do all the other netwokrs
shemnon (Fri, 29 May 2020 20:56:17 GMT):
Only Mordor was the stopper, because only newer clients have the network details as defaults.
shemnon (Fri, 29 May 2020 20:57:08 GMT):
Dropping all the peers won't help. Older clients will find you and try and connect to you. Some have _long_ memories about who they've seen.
madelinemurray (Fri, 29 May 2020 20:57:09 GMT):
My vote is for a full scrub and rebuild next Wednesday.
madelinemurray (Fri, 29 May 2020 20:58:07 GMT):
So it's really starting a new node from scratch we need to do more often if we're going to catch these types of bugs sooner?
shemnon (Fri, 29 May 2020 20:58:26 GMT):
basiclly, yes.
rai (Fri, 29 May 2020 20:58:43 GMT):
What about an admin rpc that will clear the peer table?
rai (Fri, 29 May 2020 20:58:43 GMT):
What about an admin rpc that will flush the peer table?
madelinemurray (Fri, 29 May 2020 20:58:55 GMT):
But that's a drag on the network if we're starting new fastsync nodes all the time isn't it?
madelinemurray (Fri, 29 May 2020 21:00:34 GMT):
An rpc that clears the peer table is probably easier in the long run than starting new nodes all the time isn't it if what we're trying to test for is 'can peer on mainnet' rather than 'can fastsync'
rai (Fri, 29 May 2020 21:01:44 GMT):
Yeah, it would definitely be a more targeted way to poke the bear
madelinemurray (Fri, 29 May 2020 21:27:09 GMT):
If we do full scrub and start next Wednesday is there a way to test that geth and openethereum clients won't reject us once the fixes are in? Or we won't know for sure until we rebuild and start fastsync again on Wednesday?
rai (Fri, 29 May 2020 21:27:36 GMT):
We could probably get that admin api in by next wednesday
shemnon (Fri, 29 May 2020 21:56:30 GMT):
a dev gorli chain should be sufficient. check the peer table and look for Geth running an eth/64 capability - https://besu.hyperledger.org/en/stable/Reference/API-Methods/#admin_peers
ajsutton (Fri, 29 May 2020 23:04:26 GMT):
Does besu actually store the peer table to disk? I didn't think it did. And even if we cleared our peer table, other nodes would still know about us and make inbound connections which is how I would have guessed we wound up finding peers.
ajsutton (Fri, 29 May 2020 23:05:35 GMT):
Right that's what shemnon said too - missed it in this wonderful RocketChat interface. :)
rai (Fri, 29 May 2020 23:29:15 GMT):
I don't know for sure but I just assumed that's what all p2p systems do so they don't always rely on bootnodes :P
rai (Fri, 29 May 2020 23:29:15 GMT):
I don't know for sure but I just assumed that's what all p2p systems did so they don't always rely on bootnodes :P
shemnon (Sat, 30 May 2020 02:11:12 GMT):
No, besu doesn't store the peer table to disk. But other nodes are up long enough that they will gossip an old nodes ghost for quite a while.
madelinemurray (Sat, 30 May 2020 03:48:43 GMT):
@shemnon - is dev-besu-ohio-mainnet-1.4.6-RC1-BIS the dev build with the fixes in it?
madelinemurray (Sat, 30 May 2020 03:48:51 GMT):
Or something else completely?
shemnon (Sat, 30 May 2020 03:49:00 GMT):
someone else's box.
madelinemurray (Sat, 30 May 2020 03:50:31 GMT):
Cool. The privacy nodes on Ropsten are behind now too - is that because they're on the RC version as well? Or is there another problem with them? dev-besu-ohio-privacy-reorg-0 and dev-besu-ohio-privacy-reorg-1
rjones (Sat, 30 May 2020 05:02:10 GMT):
https://lf-landscape.netlify.app/category=hyperledger&format=card-mode&grouping=category&selected=hyperledger-besu
shemnon (Sat, 30 May 2020 19:45:48 GMT):
Those ropsten nodes started a from-scratch fast sync. They finished chain download but are working on state download.
madelinemurray (Sun, 31 May 2020 19:43:20 GMT):
@shemnon - are https://github.com/hyperledger/besu/pull/1005 and https://github.com/hyperledger/besu/pull/1006 the two PRs needed to fix the 1.4.6-RC1 bugs? So if we started a fastsync with the latest master we'd be able to confirm we won't see the same issues as Friday?
shemnon (Sun, 31 May 2020 19:43:33 GMT):
Those are the two.
madelinemurray (Sun, 31 May 2020 19:43:42 GMT):
Great - thanks
shemnon (Sun, 31 May 2020 19:45:01 GMT):
1006 passed the EF's tests - https://hivetests.ethdevops.io/?suite=1590939003-029126e600861c71f531f2a85d61c160.json - down to 5 conformance bugs from 8 (which was up from 7 with the bug)
madelinemurray (Sun, 31 May 2020 19:55:07 GMT):
I'm not across what exactly those tests do. Do they tests test for the peering with other clients and versions? ie, if we pass those tests we'd expect to catch bugs like 1005 and 1006 before we cut an RC and start syncing?
shemnon (Sun, 31 May 2020 20:25:01 GMT):
They failures in the reference tests went up because we were rejecting all ping packets, including the two tests I was trying to fix. Now we are passing those two tests as well as all tests we passed before. So I consider this external validation the fixes worked as intended.
madelinemurray (Sun, 31 May 2020 20:42:59 GMT):
Thanks for the explanation. When/how do we run the the EF tests?
rjones (Mon, 01 Jun 2020 06:16:39 GMT):
Links to chat channels for logged-out users work now.
NicolasMassart (Mon, 01 Jun 2020 07:41:33 GMT):
I see latest fixes are in master which is 1.4.7 now. Are we going to port these fixes in 1.4.6 release branch or will 1.4.6 release be abandoned in favour of a 1.4.7?
shemnon (Mon, 01 Jun 2020 14:24:10 GMT):
the EF has a server at https://hivetests.ethdevops.io/ where they post results they run every 2 or so days. I've tried to build hive myself and the docs are awful. I'm working on the ethereum/tests gitter to see how it's done.
DannoFerrin (Mon, 01 Jun 2020 14:25:59 GMT):
Has joined the channel.
DannoFerrin (Mon, 01 Jun 2020 14:26:00 GMT):
Not only that, but you don't need an LFID to sign in!
benjamincburns (Mon, 01 Jun 2020 14:38:35 GMT):
We'll cherry pick the fixes over to the release_1.4.6 branch
bmg (Mon, 01 Jun 2020 15:25:41 GMT):
:woo: great, there's a couple of tracing fixes I'm interested in
shemnon (Mon, 01 Jun 2020 15:47:59 GMT):
User User_1 removed by shemnon.
rjones (Mon, 01 Jun 2020 15:57:15 GMT):
Interesting
benjamincburns (Mon, 01 Jun 2020 18:34:38 GMT):
Is there a way to configure notifications on the mobile app? For some reason it buzzes my phone for any message to this channel now 😣
benjamincburns (Mon, 01 Jun 2020 18:35:08 GMT):
I didn't see anything in the main settings pane 😕
bmg (Mon, 01 Jun 2020 18:35:57 GMT):
Filename
bmg (Mon, 01 Jun 2020 18:36:08 GMT):
Filename
bmg (Mon, 01 Jun 2020 18:36:33 GMT):
Just found it myself. Tab the name of the channel at the top of the screen
bmg (Mon, 01 Jun 2020 18:36:33 GMT):
Just found it myself. Tap the name of the channel at the top of the screen
bmg (Mon, 01 Jun 2020 18:37:03 GMT):
Then hit notifications at the bottom
benjamincburns (Mon, 01 Jun 2020 18:53:58 GMT):
thanks!
madelinemurray (Mon, 01 Jun 2020 20:21:12 GMT):
@bmg - which ones are those? I can't see any merged PRs in 1.4.7 that are related to tracing?
sumaids (Tue, 02 Jun 2020 09:12:23 GMT):
Has left the channel.
bmg (Tue, 02 Jun 2020 10:34:12 GMT):
@madelinemurray flat trace generator bug and debug operation tracer . I can look up the issue numbers a bit later
bmg (Tue, 02 Jun 2020 10:37:15 GMT):
#911 and #915 @madelinemurray
joshuafernandes (Tue, 02 Jun 2020 22:28:52 GMT):
@here - We're beginning the release process for hypperledger/besu, please don't merge anything to release-1.4.6 until further notice
joshuafernandes (Tue, 02 Jun 2020 23:16:04 GMT):
@here: - It's now safe to merge into release-1.4.6 for hyperledger/besu again.
kuronosec (Thu, 04 Jun 2020 11:03:41 GMT):
Has joined the channel.
shemnon (Thu, 04 Jun 2020 23:58:01 GMT):
Are build alerts properly firing on this channel? https://app.circleci.com/pipelines/github/hyperledger/besu/4312/workflows/fa228893-15fb-4614-9d93-ebdc8843368c/jobs/20128 was a DCO failure
rjones (Fri, 05 Jun 2020 00:28:21 GMT):
hmm I guess not.
NicolasMassart (Fri, 05 Jun 2020 15:46:33 GMT):
Not easy to review as it shows effect only once in master branch, but if you can have a look at this PR please https://github.com/hyperledger/besu-docs/pull/386
timbeiko (Fri, 05 Jun 2020 18:59:28 GMT):
:besu: *Release Information* :besu:
As most of you know, the 1.5 release of Besu is just around the corner, so I wanted to reiterate the release schedule we have planned:
* *June 17 (AMER)/ 18 (AUS): Besu 1.5 RC*
* *July 14 (AMER) /15 (AUS): Besu 1.5 Release*
We'll have a 4 week RC cycle to allow for ample testing, bug fixes and complete mainnet fast syncs.
timbeiko (Fri, 05 Jun 2020 18:59:28 GMT):
:besu: *Release Information* :besu:
As most of you know, the 1.5 release of Besu is just around the corner, so I wanted to reiterate the release schedule we have planned:
- *June 17 (AMER)/ 18 (AUS): Besu 1.5 RC*
- *July 14 (AMER) /15 (AUS): Besu 1.5 Release*
We'll have a 4 week RC cycle to allow for ample testing, bug fixes and complete mainnet fast syncs.
jagpreet (Mon, 08 Jun 2020 14:13:25 GMT):
Has joined the channel.
jagpreet (Mon, 08 Jun 2020 14:13:26 GMT):
Hi,
The peer discovery service uses UDP protocol as mentioned here https://github.com/PegaSysEng/besu-kubernetes/blob/master/helmfile/ibft2/charts/besu-node/templates/node-service.yaml
We use ambassador in BAF (blockchain automation framework) for proxy which doesn't support UDP protocol. Is there a support for TCP protocol for the same or any other alternatives?
sownak (Mon, 08 Jun 2020 14:33:36 GMT):
Has joined the channel.
shemnon (Mon, 08 Jun 2020 14:44:17 GMT):
You could use a static peers file, but that would only allow for a static topology. Dynamic peer discovery in Ethereum only uses UDP.
shemnon (Mon, 08 Jun 2020 16:56:26 GMT):
Besu contributor call tomorrow at 1500 UTC (EMEA/AMER friendly times) - https://wiki.hyperledger.org/display/BESU/2020-06-09+Besu+Contributor+Call+Agenda
shemnon (Mon, 08 Jun 2020 21:06:50 GMT):
feel free to add items to the agenda for tomorrows call.
jagpreet (Tue, 09 Jun 2020 05:19:59 GMT):
Okay
shemnon (Tue, 09 Jun 2020 14:52:17 GMT):
contributor call in 10 minutes - https://consensys.zoom.us/j/199741148
jramps9 (Tue, 09 Jun 2020 15:52:35 GMT):
Hi Besu contributors! Reminder the Marketing Committee-Dev Relations call is tomorrow, 6/10 at 9am PT. Hope to see you there! https://wiki.hyperledger.org/display/Marketing/2020-06-010+Meeting+notes
madelinemurray (Wed, 10 Jun 2020 21:09:37 GMT):
The 1.4.6 release was scheduled for today but the mainnet sync with the RC is still underway so I'm proposing we move the release to Monday to allow time for this to complete.
benjamincburns (Wed, 10 Jun 2020 21:25:14 GMT):
@madelinemurray I assume this will be Monday AU time? (that's fine for me - I work AU days)
madelinemurray (Wed, 10 Jun 2020 21:26:10 GMT):
@benjamincburns - good point. If you're happy to do Monday AU, then that probably makes sense. Not strongly attached to Monday AU or Monday US though.
joshuafernandes (Thu, 11 Jun 2020 04:44:22 GMT):
@here: The head of `master` branch for repository `hyperledger/besu` has been rebased to deal with a DCO issue, please follow the instructions here https://wiki.hyperledger.org/pages/viewpage.action?pageId=24781754# to fix your local repositories
GregTheGreek (Thu, 11 Jun 2020 14:03:02 GMT):
I came across this recently, https://www.mergefreeze.com/, might be usefull for the Release team. I know you usually just tell people not to merge, but this might work easier
benjamincburns (Thu, 11 Jun 2020 21:10:29 GMT):
Update on pre-release fast sync testing. We observed that the node under test did complete its sync, but the machine crashed not long after (about 1h). We don't have a conclusive reason for the crash yet, but some of the syslog messages make it look like a memory availability issue (udev was complaining that it couldn't create something due to a failure to allocate memory).
benjamincburns (Thu, 11 Jun 2020 21:10:29 GMT):
Update on pre-release fast sync testing. We observed that the node under test did complete its sync, but the machine crashed not long after (about 1h). We don't have a conclusive reason for the crash yet, but some of the syslog messages make it look like a memory availability issue (systemd-udevd was complaining that it couldn't fork a child process due a failure to allocate memory).
benjamincburns (Thu, 11 Jun 2020 21:10:29 GMT):
Update on pre-release ~fast~ full sync testing. We observed that the node under test did complete its sync, but the machine crashed not long after (about 1h). We don't have a conclusive reason for the crash yet, but some of the syslog messages make it look like a memory availability issue (systemd-udevd was complaining that it couldn't fork a child process due a failure to allocate memory).
benjamincburns (Thu, 11 Jun 2020 21:13:42 GMT):
Someone needs to make a call about what this crash means for our go/no-go decision on shipping in light of this. @timbeiko says that he thinks that he observed a node crashing not long after fast sync completed recently, which makes me think that there's enough to be concerned about this issue repeating for our users. That said, I don't think we really have enough information yet to even categorize it as a bug. That is, I have no idea how frequently this issue will occur in the real world.
If it's simply because the node we were running needed more memory, is this even a bug? The machine in question had 7.69GiB (thank Amazon for the weird numbers).
benjamincburns (Thu, 11 Jun 2020 21:15:37 GMT):
@madelinemurray what do you think we should do? My weak suggestion, based on the feeling of not having enough data, is to try running it again on a machine with more memory, with DEBUG logging enabled, and potentially with better system-level monitoring via something like telegraf.
ajsutton (Thu, 11 Jun 2020 21:21:49 GMT):
So, it finished fast sync and was processing the remaining blocks up to head and it ran out of memory. Right?
Do we have other nodes running the RC that were able to process those blocks?
lucassaldanha (Thu, 11 Jun 2020 21:23:12 GMT):
Any chance we can duplicate the db of this node and start two instances with the RC, 1 instance with the same 8GB and another with a bit more memory?
lucassaldanha (Thu, 11 Jun 2020 21:24:18 GMT):
@ajsutton idk if we have other instances running the RC. I think @timbeiko was running one (and apparently had issues as well).
shemnon (Thu, 11 Jun 2020 21:24:23 GMT):
It was more than 1 h. after 90 min it started falling behind and after 3h45m it ran out of system memory and the node became unresponsive.
shemnon (Thu, 11 Jun 2020 21:24:33 GMT):
It has reached full sync and was in sync for over an our.
shemnon (Thu, 11 Jun 2020 21:25:12 GMT):
My classic nodes have been fine, so I think it's an issue of state size.
shemnon (Thu, 11 Jun 2020 21:31:16 GMT):
This feels to me to be a memory issue with mainnet that should be solved by upping our minimum specs.
shemnon (Thu, 11 Jun 2020 21:31:16 GMT):
This feels to me to be a memory issue with mainnet that should be solved by upping our minimum specs for mainnet.
benjamincburns (Thu, 11 Jun 2020 21:43:41 GMT):
I believe it caught up with head before it ran out of memory.
benjamincburns (Thu, 11 Jun 2020 21:44:06 GMT):
I don't know that this will reproduce it, but it's worth a try, I suppose
benjamincburns (Thu, 11 Jun 2020 21:45:01 GMT):
are we targeting a particular memory spec? if so, I think it's a bad idea for this to prompt us to raise that target. Better to make the decision when we're not under pressure to release, as otherwise the target is meaningless.
joshuafernandes (Thu, 11 Jun 2020 21:45:19 GMT):
itd be 16G I think thats the next one for us
benjamincburns (Thu, 11 Jun 2020 21:45:29 GMT):
if we're not targeting a particular memory spec, then I have no objection to this
joshuafernandes (Thu, 11 Jun 2020 21:45:34 GMT):
or we tune the jvm with a host of 8G
joshuafernandes (Thu, 11 Jun 2020 21:46:32 GMT):
was there some new code added to sync's that increased mem footprint?
benjamincburns (Thu, 11 Jun 2020 21:46:48 GMT):
The obnoxious thing at the moment is that all of the system-level monitoring during the period which the node was malfunctioning was dropped, so we can really only infer what happened from clues in the besu log and syslog
benjamincburns (Thu, 11 Jun 2020 21:46:48 GMT):
The obnoxious thing at the moment is that all of the system-level monitoring during the period thatthe node was malfunctioning was dropped, so we can really only infer what happened from clues in the besu log and syslog
benjamincburns (Thu, 11 Jun 2020 21:46:48 GMT):
The obnoxious thing at the moment is that all of the system-level monitoring during the period that the node was malfunctioning was dropped, so we can really only infer what happened from clues in the besu log and syslog
shemnon (Thu, 11 Jun 2020 21:48:15 GMT):
We do load a new set of native libraries for BLS12 signatures
shemnon (Thu, 11 Jun 2020 21:50:44 GMT):
That's the only thing that strikes me as large scale memory usage, other than general mainnet load.
benjamincburns (Thu, 11 Jun 2020 22:30:01 GMT):
Just looking at the memory stats again, it did report a resident set of 7.33 GiB before it stopped reporting
joshuafernandes (Thu, 11 Jun 2020 22:33:56 GMT):
when should the OOM killer have fired?
joshuafernandes (Thu, 11 Jun 2020 22:41:52 GMT):
@lucassaldanha @ajsutton @shemnon anyone got any insights into this? @benjamincburns will be going offline soon and we'd like to set things up to get some metrics if possible
ajsutton (Thu, 11 Jun 2020 22:53:23 GMT):
Sorry, lots of meetings this morning. Personally I'm not too concerned that we need more than 8Gb for a MainNet sync - the world state for MainNet keeps growing and generally it's resource requirements keep growing too. I definitely wouldn't be holding the release just for that concern, though it probably is worth doing some work to better understand what's using the memory involved.
benjamincburns (Thu, 11 Jun 2020 22:55:19 GMT):
I'm mostly concerned about the stability aspect - I'm surprised that it managed to take down the whole machine...
joshuafernandes (Thu, 11 Jun 2020 22:55:30 GMT):
aye, lets hold till we get more data, @benjamincburns and I are looking at graphs trying to work out what happend atm, but arent clear as to what it was meant to be doing ie pruning kicked in/ full sync block issue / what exactly
benjamincburns (Thu, 11 Jun 2020 22:55:36 GMT):
normaly I would have expected the kernel to kill the process and systemd to restart it
joshuafernandes (Thu, 11 Jun 2020 22:55:58 GMT):
Yep, this has never happened before
benjamincburns (Thu, 11 Jun 2020 22:56:07 GMT):
Clipboard - June 11, 2020 3:55 PM
Juan-cc (Thu, 11 Jun 2020 22:57:53 GMT):
Has joined the channel.
ajsutton (Thu, 11 Jun 2020 22:58:59 GMT):
I've seen this happen before. Quite often see EC2 instances die like this when memory is pressed heavily. I get the impression the OOM killer is either effectively disabled or configured in a way that tends to make it kill the wrong stuff
benjamincburns (Thu, 11 Jun 2020 22:59:37 GMT):
I have the kernel logs though and it didn't even fire :-/
benjamincburns (Thu, 11 Jun 2020 22:59:52 GMT):
and I've definitely seen oomkiller working on EC2 instances in the past
benjamincburns (Thu, 11 Jun 2020 23:00:04 GMT):
though you're right - it often does kill the wrong process
ajsutton (Thu, 11 Jun 2020 23:00:40 GMT):
I've never had the kernel logs before so don't know what it was doing. But this is definitely not new behaviour to me - the question is more around whether the increase in memory requirements are an issue.
benjamincburns (Thu, 11 Jun 2020 23:00:45 GMT):
Personally I'd feel a lot better if we could determine the reason for the crash beyond just "it ate all the memory"
ajsutton (Thu, 11 Jun 2020 23:02:01 GMT):
Given we only started doing clean syncs on 8GB boxes recently, do we know it used to work reliably previously? If it has worked, in previous runs did it just squeak through using nearly all the memory?
joshuafernandes (Thu, 11 Jun 2020 23:02:27 GMT):
Come again, this has happened before?
ajsutton (Thu, 11 Jun 2020 23:03:42 GMT):
I've seen this kind of crash quite a while back around the time we were first implementing fast sync. If you used all the memory on the box it turned into a zombie. Seen it happen more recently with Teku as well (due to known memory usage growth during non-finalization)
benjamincburns (Thu, 11 Jun 2020 23:06:37 GMT):
is this on OpenJDK 11?
joshuafernandes (Thu, 11 Jun 2020 23:07:49 GMT):
should be
joshuafernandes (Thu, 11 Jun 2020 23:08:01 GMT):
11.0.2 from memory
joshuafernandes (Thu, 11 Jun 2020 23:09:26 GMT):
does physically enforcing max limits on mem prevent this?
joshuafernandes (Thu, 11 Jun 2020 23:09:26 GMT):
does physically enforcing java max limits on mem prevent this?
joshuafernandes (Thu, 11 Jun 2020 23:10:16 GMT):
ie if it cant use up 100% of resources at least the box is still alive for us to glean some info from
benjamincburns (Thu, 11 Jun 2020 23:14:23 GMT):
when running on bare metal without artificial memory limits imposed by something like cgroups, we should be seeing OutOfMemoryExceptions
ajsutton (Thu, 11 Jun 2020 23:14:24 GMT):
No. -Xmx only controls the on-heap memory, but RocksDB uses a bunch of off heap memory which isn't restricted by that. It makes it very difficult to correctly tune the memory settings for Besu with a big world state, relatively slow disk and limited memory (ie these EC2 instances)
benjamincburns (Thu, 11 Jun 2020 23:14:45 GMT):
yeah, but we have native code in the mix now
ajsutton (Thu, 11 Jun 2020 23:15:07 GMT):
You'll only get java OutOfMemoryError if the java heap runs out of memory. RocksDB's behaviour is different because it's native and using off heap.
benjamincburns (Thu, 11 Jun 2020 23:15:39 GMT):
yeah
ajsutton (Thu, 11 Jun 2020 23:16:18 GMT):
hmmm, the default grafana dashboard seems to be adding the two together which is very misleading.
benjamincburns (Thu, 11 Jun 2020 23:16:19 GMT):
I just wish the damn INFO level logs said something like "pruning database" or something right around the time things went pear shaped
benjamincburns (Thu, 11 Jun 2020 23:16:25 GMT):
but there's nothing meaningful in the logs
benjamincburns (Thu, 11 Jun 2020 23:16:38 GMT):
and there's nothing conclusive in any of our other monitoring
joshuafernandes (Thu, 11 Jun 2020 23:16:42 GMT):
so we need a proper memory profile perhaps? giving it more mem seems like it just buys extra time?
benjamincburns (Thu, 11 Jun 2020 23:17:06 GMT):
>hmmm, the default grafana dashboard seems to be adding the two together which is very misleading.
I've been looking to the resident set size - I think it's on the peering dashboard
benjamincburns (Thu, 11 Jun 2020 23:17:17 GMT):
that's system-wide though, not just besu
benjamincburns (Thu, 11 Jun 2020 23:17:56 GMT):
absent some information that tells us conclusively why this crashed, I think we have to run it again, otherwise why did we bother doing this test in the first place?
ajsutton (Thu, 11 Jun 2020 23:18:09 GMT):
My understanding, which could be outdated is that the biggest factor in memory usage is around world state and what transactions are doing. So potentially we just processed a series of transactions that loaded and/or changed a bunch of world state at once, causing more nodes to be held in memory in Besu and growing the size of RocksDB's memory because it couldn't compact fast enough.
ajsutton (Thu, 11 Jun 2020 23:18:30 GMT):
Well that's why I'm interested to know if we've seen this run well in the past and if so, how close to full memory was it.
benjamincburns (Thu, 11 Jun 2020 23:18:33 GMT):
it had finished processing old world state entries
ajsutton (Thu, 11 Jun 2020 23:18:43 GMT):
No, every transaction touches world state.
ajsutton (Thu, 11 Jun 2020 23:19:11 GMT):
If previous versions were really close to memory limits, then adding some more blocks on top to MainNet could easily be enough to push it over the limit.
benjamincburns (Thu, 11 Jun 2020 23:20:04 GMT):
>Well that's why I'm interested to know if we've seen this run well in the past and if so, how close to full memory was it.
Josh and I don't have any knowledge of this succeeding or failing in the past - we don't run full syncs very often
ajsutton (Thu, 11 Jun 2020 23:20:24 GMT):
When did we introduce this as a gate for releases? Is this the first one?
benjamincburns (Thu, 11 Jun 2020 23:20:32 GMT):
which is part of what has me erring toward running it again
ajsutton (Thu, 11 Jun 2020 23:20:51 GMT):
I'm questioning whether it's a sensible test to run.
madelinemurray (Thu, 11 Jun 2020 23:20:54 GMT):
We've done it for the last couple @ajsutton
ajsutton (Thu, 11 Jun 2020 23:21:03 GMT):
Do we have records showing what the memory usage was?
benjamincburns (Thu, 11 Jun 2020 23:21:04 GMT):
those were fast syncs, though - right?
madelinemurray (Thu, 11 Jun 2020 23:21:24 GMT):
All fast syncs move to full sync at the end - this was a fast sync
joshuafernandes (Thu, 11 Jun 2020 23:24:38 GMT):
hang on - i thought that was in the full sync state
shemnon (Thu, 11 Jun 2020 23:31:38 GMT):
It had gone fast sync and then had gone to full sync and had completed full sync.
shemnon (Thu, 11 Jun 2020 23:31:52 GMT):
Are we sure it's besu causing the remaining memory pressure?
joshuafernandes (Fri, 12 Jun 2020 00:32:09 GMT):
@here: We had a T3.large with Besu 1.4.6-RC2 drop all of a sudden - we believe this to be related to memory / slow disk and not really related to code directly. We are going to:
- Update our requirements to have a fast disk which doesnt put as much pressure on memory or slower disk with more memory
- spin up a few comparison boxes on i3.large & t3.xl and compare sync performance of 1.4.2 and 1.4.6 and get more data for this, consequently update our test box & process
- the release for 1.4.6 *will* go ahead on Monday AEST as planned
joshuafernandes (Fri, 12 Jun 2020 00:33:39 GMT):
I think thats the general gist, but please update anything if I've missed it?
madelinemurray (Fri, 12 Jun 2020 01:31:37 GMT):
I'll update the doc to indicate our minimum spec for mainnet is t3.xl
madelinemurray (Fri, 12 Jun 2020 02:55:03 GMT):
@joshuafernandes @benjamincburns - I'm updating the min specs now. Is 8GB for the JVM still right for mainnet?
benjamincburns (Fri, 12 Jun 2020 03:41:34 GMT):
The short answer is that right now we don't know. We're going to do some experiments to get a better idea. I expect that a recommendation of 8GB is a safe bet for now, though.
shemnon (Fri, 12 Jun 2020 03:58:41 GMT):
I think the correct answer is "let the JVM decide" and just specify a system memory minimum. Specifying -Xmx and -Xms turns off new logic that it uses to commuincate with the container to decide how much to let the heap grow. Our memory problems are more related to RocksDB than the JVM.
joshuafernandes (Mon, 15 Jun 2020 00:12:19 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.4.6 until further notice
joshuafernandes (Mon, 15 Jun 2020 02:16:27 GMT):
This is now complete
ajsutton (Mon, 15 Jun 2020 09:37:29 GMT):
The API for a `Gauge` in Besu's metrics doesn't support labels as we do for counters and timers. Originally this was a deliberate omission because it seemed better to just include it in the name. In practice however there's a few places where it winds up being quite clumsy - in particular the metrics around executor queue lengths. Any objections to adding label support to gauges? It would affect the plugin API but, while I haven't worked out what the API should be exactly, it should be natural to have it as an additional method so shouldn't have any backwards compatibility issues.
lucassaldanha (Mon, 15 Jun 2020 12:28:02 GMT):
No objections :+1:
shemnon (Mon, 15 Jun 2020 17:00:01 GMT):
The problem we will run into is that gauges are not idempotent. Once you add a labaled gauge you can only use that gauge going forward. This was my experience the last time I tried this.
shemnon (Mon, 15 Jun 2020 17:01:01 GMT):
There are many places in besu were we create a counter with the same name and Prometheous doesn't care. But if we have a gauge with two different callbacks Prom doesn't know how to handle it.
ajsutton (Mon, 15 Jun 2020 20:59:51 GMT):
From memory Prometheus won't let you create two counters with the same name either, but the Besu wrapper handles it so you get the same Counter instance back if you use the same name multiple times. That works great for counters because all the inc calls are meant to just add up.
For gauges having two with the same name doesn't make sense because they can only have one value at a time (and the API the besu wrapper exposes just provides a Supplier which is even less conducive to having two instances).
But with labels on gauges you could have the same name but different label values and that would be valid. The Besu wrapper will probably have to work a little magic to make it easy to use but I'm pretty sure it will be doable. It will still throw if you use the same name and label values though.
macfarla (Mon, 15 Jun 2020 22:57:05 GMT):
As many other products and platforms are doing, we are making the change in terminology in Besu from whitelist/blacklist to allowlist/denylist. Since this involves CLI options, we will first be adding additional CLI options as alternate names, and then will follow the process in Besu for breaking changes to remove the whitelist/blacklist CLI options. Work has started on this issue https://github.com/hyperledger/besu/issues/1076
mwagner (Tue, 16 Jun 2020 02:36:36 GMT):
Has joined the channel.
sumaids (Tue, 16 Jun 2020 02:36:53 GMT):
Has joined the channel.
joshuafernandes (Tue, 16 Jun 2020 02:38:27 GMT):
@here: please welcome @sumaids who is developing a K8S operator for Besu on Openshift as part of Hyperledger Internship projects 2020, more details https://wiki.hyperledger.org/pages/viewpage.action?pageId=31195317
joshuafernandes (Tue, 16 Jun 2020 02:38:55 GMT):
also @mwagner who is mentoring him from Red Hat
joshuafernandes (Wed, 17 Jun 2020 22:40:10 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
joshuafernandes (Wed, 17 Jun 2020 23:33:32 GMT):
@here: this is complete now
sumaids (Thu, 18 Jun 2020 06:56:33 GMT):
When I try to run run any example from besu-sample-networks, I get error
E: List directory /var/lib/apt/lists/partial is missing. - Acquire (13: Permission denied)
/bin/sh: 1: apk: not found
ERROR: Service 'bootnode' failed to build: The command '/bin/sh -c (apt-get update && apt-get install dos2unix || apk add --update dos2unix) && rm -rf /var/cache/apt/* && rm -rf /var/cache/apk/*' returned a non-zero code: 127
I am using MacOS, and docker, docker-compose installed with 6G memory alloted.
shemnon (Thu, 18 Jun 2020 23:06:52 GMT):
@joshuafernandes may be the best person to field this. However ConsensSys is observing Juneteenth so he may not be available to answer until the 22nd.
BillyLuedtke (Fri, 19 Jun 2020 03:52:59 GMT):
Has joined the channel.
jagpreet (Fri, 19 Jun 2020 04:41:33 GMT):
Has left the channel.
joshuafernandes (Fri, 19 Jun 2020 08:32:06 GMT):
I'll fix this when Im back in, it'd be the breaking change we introduced for user:group. In the interim, set the Besu version to 1.4.6 in the `env` or `.env` file at the root level
macfarla (Mon, 22 Jun 2020 03:35:24 GMT):
Besu contributor call tomorrow at APAC/AMER Friendly Time - 0100UTC (1100 Brisbane time) https://wiki.hyperledger.org/display/BESU/2020-06-23+Besu+Contributor+Call+Agenda
macfarla (Tue, 23 Jun 2020 01:29:39 GMT):
Besu contributor call notes https://wiki.hyperledger.org/display/BESU/2020-06-23+Contributor+Call+Notes
joshuafernandes (Tue, 23 Jun 2020 02:16:00 GMT):
https://github.com/PegaSysEng/besu-sample-networks/pull/84 fixes that issues and I'll merge this in today after its been reviewed
manumarchal (Wed, 24 Jun 2020 08:12:56 GMT):
Has joined the channel.
sownak (Wed, 24 Jun 2020 08:27:01 GMT):
Hello contributors, I had a query related to the production deployment of Besu. Having read https://besu.hyperledger.org/en/stable/HowTo/Deploy/Cloud/ and the sample Kubernetes project, is https://github.com/PegaSysEng/besu-kubernetes/tree/master/helm/consortium/besu the better architecture? In that case, is it expected that each participating organization run a set of validators and a set of nodes (with or without private TM)?
joshuafernandes (Wed, 24 Jun 2020 22:26:28 GMT):
It depends :) Some people that have everything in house ie Dapps, chain, other business apps etc (no external users) will find the non consortium variants easier to manage. If you have a core group admin'ing the validators and users come and go, then the consortium is the preferred approach. Couple of things to bear in mind that participating members do not _need_ to be validators - the can join as members - there is an upper bound on effective number of validators. Also when users join they can be their choice of setup (cloud, on prem VMs/instances/ etc or on K8S)
sownak (Thu, 25 Jun 2020 08:16:29 GMT):
Thanks. Are the validators members as well? i.e. do they participate in the blockchain?
joshuafernandes (Thu, 25 Jun 2020 08:26:58 GMT):
Yep, they're the most important part - think of them as founding members :) The validators create blocks so without them the chain wont grow. After the initial set are created you vote new validators in (members that will now be responsible for block creation)
joshuafernandes (Thu, 25 Jun 2020 08:27:46 GMT):
Essentially you have 3 loosely defined roles:
joshuafernandes (Thu, 25 Jun 2020 08:27:46 GMT):
Essentially you have 3 loosely defined roles:
- bootnodes - all nodes connect to them first to get a list of peers to connect to.
- validators - create blocks and the chain grows. Sometimes the initial validator pool are also bootnodes
- members - normal nodes which you'd pair with a privacy tx mgr
sownak (Thu, 25 Jun 2020 08:34:52 GMT):
Perfect, thanks a lot. Will the roles be different for another consensus? I guess validator is only for IBFT2
joshuafernandes (Fri, 26 Jun 2020 06:16:26 GMT):
This largely depends on the type of consensus algorithm used - POW have miners to create blocks eg ropsten network. POA algorithms like clique/ibft2/etc use validators eg (goerli network, private networks etc)
abdelhamidbakhta (Fri, 26 Jun 2020 15:53:25 GMT):
Hello, we are considering introducing lombok (https://projectlombok.org/) to reduce boilerplate code in Java classes. One of the main reason is to have some consistency across the codebase when using some design patterns. For instance, currently there are 3 different ways of implementing Builder pattern in Besu. We also believe introducing Lombok would improve readability and maintainability of the codebase. We have a showcase pull request (https://github.com/hyperledger/besu/pull/1148) to illustrate the potential benefits of using Lombok. This PR uses Value and Builder annotations from Lombok, the impacted class is a configuration object. This is a common case where we have repetitive code that could be drastically reduced. The impact for contributors would be to add a plugin in their favorite IDE. All instructions for IDE setup are here: https://projectlombok.org/setup/intellij. Please feel free to comment the pull request and/or discuss here if you have strong opinions. We plan to add a whitelist of Lombok annotations that are ok to use. We will enforce this rule with an error-prone check ultimately.
ajsutton (Fri, 26 Jun 2020 22:41:59 GMT):
I won't stand in the way of this but my preference is very strongly against it. The issue with tools like Lombok is that they reduce the lines of code by increasing the complexity. Contributors don't just need to configure their IDE, they also need to read up and fully understand what Lombok is doing so that they can read the code and understand the "magic" that these annotations do. I'm much more concerned with reducing complexity than with reducing lines of code so my preference would be to continue using plain Java - we can still fix any inconsistent patterns that are being used (and Lombok won't stop being introducing new inconsistencies).
atoulme (Fri, 26 Jun 2020 23:57:55 GMT):
I would tend to cautiously agree with Adrian.
I do not see how this helps the project - this code is already written. Are we planning on changing that part of the code?
It might be interesting to reduce complexity, maybe by removing builders altogether since they are not used as part of an API.
abdelhamidbakhta (Sat, 27 Jun 2020 05:29:32 GMT):
I strongly disagree @ajsutton because we want to enforce the use to only some of the annotations. Like Builder Getter Setter. They don't generate any magic code. Exactly same as if it was in vanilla Java. It cannot introduce any kind of complexity with this kind of annotations. Please give me a concrete example.
abdelhamidbakhta (Sat, 27 Jun 2020 05:29:37 GMT):
I strongly disagree @ajsutton because we want to enforce the use to only some of the annotations. Like Builder Getter Setter. They don't generate any magic code. Exactly same as if it was in vanilla Java. It cannot introduce any kind of complexity with this kind of annotations. Please give me a concrete example.
davemec (Sat, 27 Jun 2020 17:31:08 GMT):
Has joined the channel.
ajsutton (Sun, 28 Jun 2020 21:57:52 GMT):
So what I mean by "magic" is the generation that Lombok is doing. There is an annotation in the source file that "magically" expands to become a full builder. That introduces extra complexity for someone reading the source code because it's another layer of indirection - you can't just follow through the Java code in your IDE you have to also know or lookup exactly what the lombok annotations expand into. It's only a small amount of additional complexity and yes people will learn them but my experience is that complexity winds up being a problem because lots of little complexities build up.
Code is written much less often than it is read and writing code is much simpler than reading it (especially since the things Lombok can generate can generally also have source code generated by IDEs). As a result when I evaluate changes like this I inevitably land on the side of keeping code as simple to read as possible even if that makes writing it harder. I appreciate that you're making different trade offs and seeing the cost/benefits differently - it's ok for us to disagree on that. :)
I thought it was important to raise my concerns that we were reducing lines of code at the cost of extra complexity. But as I said I'm not attempting to block this despite personally not liking it. If others are generally in favour of this then I'll happily follow along.
trent.mohay (Sun, 28 Jun 2020 22:18:10 GMT):
It won't materially affect me if Lombok comes in - but I generally have concerns about including a code generation "engine-room" in the production code.
I suspect it won't solve our "3 different builder problem", but will instead add a 4th way of doing it. I also suspect that some developers won't use the Lombok solution so our builders will keep changing.TBH, my IDE does most of the auto-coding I need, and doesn't add extra layers to our actual code base.
We're not likely to go back and refactor code with the lombok solution, we've got more important features to produce - so it means we'll actually have a 4th builder pattern ...
trent.mohay (Sun, 28 Jun 2020 22:18:10 GMT):
It won't materially affect me if Lombok comes in - but I generally have concerns about including a code generation "engine-room" in the production code.
I suspect it won't solve our "3 different builder problem", but will instead add a 4th way of doing it. I also suspect that some developers won't use the Lombok solution so our builders will keep changing.
TBH, my IDE does most of the auto-coding I need, and doesn't add extra layers to our actual code base.
jframe (Mon, 29 Jun 2020 01:18:17 GMT):
I see little benefit in using Lombok. There is less code to write and look at but this comes with the downside of dealing with generated code during the build process and the additional setup to use the project. Whilst this extra code can be considered boilerplate it still important and I would rather our code was explicit rather than relying on code generation for this. It also seems like the wrong way to solve the code consistency issue for builders. That I think would be better solved by using errorprone or something similar rather than forcing using of Lombok.
abdelhamidbakhta (Mon, 29 Jun 2020 07:26:08 GMT):
Thank you for your feedback guys. I understand your concerns. Dealing with design pattern inconsistency was simply an example. I agree Lombok will not remove this for other inconsistencies. Dealing with generated code during build phase is transparent, do you see any issue @jframe ? @ajsutton I agree code is written less times than reading. And this is why i believe we should make reading easy as possible. Some of our classes are very big, and more than half of the code is for getters, setters, builders, constructors. They take place. I mean even mentally they take place in your brain. IMO it is better to have only code that brings business logic. You know directly what is important. And it was definitely important to raise your concerns indeed. I would agree if you were planning to use more complex annotations. However if you limit to trivial stuff like getters, setters and builders i don't think there is any kind of extra complexity. Would you be more willing to use something interface-based like https://immutables.github.io/ ? It provided safe defaults and does not require specific plugin.
abdelhamidbakhta (Mon, 29 Jun 2020 07:33:52 GMT):
I think https://immutables.github.io/ would actually be better if we consider using dependency injection.
abdelhamidbakhta (Mon, 29 Jun 2020 09:04:47 GMT):
And because immutables is restricted to generating simple, safe and consistent value objects we won't have to write error prone rules to whitelist only partial features like we would have with Lombok.
davemec (Mon, 29 Jun 2020 11:20:22 GMT):
I have used immutables in the past and they made the codebase a lot easier to both read and maintain. I think this is a better choice than Lombok.
abdelhamidbakhta (Mon, 29 Jun 2020 12:53:50 GMT):
Draft PR with Immutables: https://github.com/hyperledger/besu/pull/1166
timbeiko (Mon, 29 Jun 2020 13:51:37 GMT):
Did anyone change anything in terms of who we can assign on issues? A few weeks ago we were able to assign @davemec on tickets directly, but now it doesn't work anymore. Did it revert back to just maintainers?
shemnon (Mon, 29 Jun 2020 14:46:25 GMT):
So the goal wouldn't be to introduce a 4th builder pattern, but to replace the other 3 builder patterns we use into a single consistent pattern. It's much easier to read code if when you recognize a pattern it's always the same pattern.
shemnon (Mon, 29 Jun 2020 14:48:25 GMT):
Part of the technical debt we've incurred is slightly different implementation patterns in different areas. This is intended as a way to pay down that debt.
rjones (Mon, 29 Jun 2020 17:42:39 GMT):
@timbeiko I just sent an invite to @davemec
rjones (Mon, 29 Jun 2020 17:45:20 GMT):
@timbeiko it's possible. Not sure.
davemec (Mon, 29 Jun 2020 18:18:59 GMT):
Thanks
shemnon (Tue, 30 Jun 2020 17:10:51 GMT):
For today's release do we need to cherry pick fixes into the release branch or will we cut from master?
madelinemurray (Tue, 30 Jun 2020 20:39:32 GMT):
Cherry pick into the release branch. @joshuafernandes
benjamincburns (Tue, 30 Jun 2020 20:44:24 GMT):
Apparently we have two new known flaky tests. I've requested this in ConsenSys slack, but I'm going to request again here. If we know for sure that these tests are flaky and it's not due to a new bug, can we please make sure that they're ignored in the AT test run prior to the release? @madelinemurray @shemnon
Apparently the tests in question are `PeerDiscoveryTableRefreshTest.tableRefreshSingleNode` and `PeerDiscoveryTimestampsTest.lastSeenAndFirstDiscoveredTimestampsUpdatedOnMessage`
benjamincburns (Tue, 30 Jun 2020 20:45:18 GMT):
For reference, every time a test fails during a release it wastes about a half hour of Yeti's time, and IIRC, we need to see up to 3 separate green builds of Besu in order to complete the release. The time sink here is really significant for us.
Alternatively we can just cancel the release if we see a red build, but even in that case we need to roll things back.
timbeiko (Tue, 30 Jun 2020 20:51:48 GMT):
In the past we didn't really disable tests because it led to them being ignored. I get that it kind of wastes Yeti's time, which is an argument to push back releases to dev teams :sweat_smile:
For this specific case, I don't have a strong opinion. The tests have issues opened for each of them. I think if Yeti is releasing then it should be your call :+1:
timbeiko (Tue, 30 Jun 2020 20:53:25 GMT):
@davemec will put up a PR to disable those two flakey tests, and he's assigned to the issues associated with them @benjamincburns
timbeiko (Tue, 30 Jun 2020 20:55:06 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=ex82hkpjumQMzg3R2) Given this, we should probably combine everything we want to cherry-pick in into a single commit.
benjamincburns (Tue, 30 Jun 2020 20:55:31 GMT):
I assume that you're joking that they've been assigned to me... in case you're not, I'd recommend assigning them to someone who actually works in the besu codebase, otherwise they'll definitely get ignored.
benjamincburns (Tue, 30 Jun 2020 20:55:31 GMT):
I assume that you're joking that they're being assigned to me... in case you're not, I'd recommend assigning them to someone who actually works in the besu codebase, otherwise they'll definitely get ignored.
timbeiko (Tue, 30 Jun 2020 20:55:57 GMT):
@benjamincburns I meant they are assigned to David :-)
benjamincburns (Tue, 30 Jun 2020 20:56:55 GMT):
Unfortunately that's not how cherry-picking works. Think of a cherry-pick as a copy of another commit.
benjamincburns (Tue, 30 Jun 2020 20:57:09 GMT):
Ah my bad - I misread! :blush:
benjamincburns (Tue, 30 Jun 2020 20:58:34 GMT):
Yep, this is exactly why we're moving releases to the dev teams starting next time around. It's not that it "kind of" wastes yeti's time, it's that we've literally lost full weeks to this exact issue.
timbeiko (Tue, 30 Jun 2020 20:59:00 GMT):
I think this may be another flaky test we want to disable: https://github.com/hyperledger/besu/issues/1168
timbeiko (Tue, 30 Jun 2020 20:59:00 GMT):
I think this may be another flaky test we want to disable: https://github.com/hyperledger/besu/issues/1168 cc: @davemec
benjamincburns (Tue, 30 Jun 2020 20:59:36 GMT):
BTW, I
benjamincburns (Tue, 30 Jun 2020 20:59:36 GMT):
BTW, If we're not sure that these are bad tests and we think that they might be bugs, I'm more in favor of postponing the release than ignoring the tests. Obviously it's a bad policy if every intermittent test failure is automatically blamed on the test rather than the code under test. If we do that, we'll obviously release bugs to our users.
timbeiko (Tue, 30 Jun 2020 21:00:58 GMT):
Isn't it possible to bring in a subset of commits from master into the RC branch in one go?
ajsutton (Tue, 30 Jun 2020 21:01:55 GMT):
You can `git cherry-pick --no-commit` to combine multiple commits into one.
benjamincburns (Tue, 30 Jun 2020 21:02:11 GMT):
Technically yes, but it's a bad practice for a number of reasons. Commits should be thought of as semantic units of work. By smashing a bunch of unrelated commits together you wind up breaking that ideal, and making it harder to trace the changes back to their origin
ajsutton (Tue, 30 Jun 2020 21:02:15 GMT):
Whether we want to do that or not is a policy decision. :)
shemnon (Tue, 30 Jun 2020 21:03:08 GMT):
We can cherry pick in sequence and then do a rebase commit instead of a sqaush commit.
shemnon (Tue, 30 Jun 2020 21:03:25 GMT):
That was what my plan was.
shemnon (Tue, 30 Jun 2020 21:03:47 GMT):
(sorry, was playing drum lesson tech video chat support for the last hour)
benjamincburns (Tue, 30 Jun 2020 21:04:33 GMT):
you guys are impacted by this more than me, so this is the last thing I'll say about it... but I discourage combining unrelated commits. Apart from the above semantics & traceability issues, it also makes them more cumbersome to revert cleanly
shemnon (Tue, 30 Jun 2020 21:06:38 GMT):
What PRs do we want to cherry pick? My list
1124 - RevRel email
1125 - YOLO fix (?)
1128 - dev network fast sync
1146 - P2P amplicfication attack fix
1066 - Truffle trace fix
1154 - eth_getLogs cache fix
1149 - lacchain sync timeout fix
1175 - cache fix for 1146
shemnon (Tue, 30 Jun 2020 21:06:38 GMT):
What PRs do we want to cherry pick? My list
1124 - DevRel email
1125 - YOLO fix (?)
1128 - dev network fast sync
1146 - P2P amplicfication attack fix
1066 - Truffle trace fix
1154 - eth_getLogs cache fix
1149 - lacchain sync timeout fix
1175 - cache fix for 1146
shemnon (Tue, 30 Jun 2020 21:07:50 GMT):
I can do them all if there is agreement on the list.
shemnon (Tue, 30 Jun 2020 21:15:38 GMT):
Also 1116 - fix slow JWT validaiton
shemnon (Tue, 30 Jun 2020 21:20:32 GMT):
Here's the PR to rebase - https://github.com/hyperledger/besu/pull/1179
ajsutton (Tue, 30 Jun 2020 21:27:57 GMT):
Immutables is definitely much, much nicer than Lombok. It worked out of the box with the IDE and preserves the ability to "click through" code. The fact that we write an interface to define the object accurately reflects the fact that the implementation comes from elsewhere and Immutables actually generates a full Java source file for the implementation it generates. As a result you can find implementations of the interface and be taken to perfectly readable Java code defining it (it's even more lines of code that we'd write but I'm fine with that).
There's magic happening but the generated source file shows exactly what's been done and the complexity and need to understand Immutables impact writing code, not reading it - you have to know to write an interface with the right annotation and that the generated class will magically appear with an `Immutable` prefix in the name but since code is written much less than it's read even if you have to look up that construct every time you need it you don't get the same "death by a thousand cuts" you do if that complexity impacts reading the code.
Basically immutables resolves all the concerns I had with Lombok. From what I can see it looks really good.
shemnon (Tue, 30 Jun 2020 21:34:21 GMT):
You will like dagger then too.
shemnon (Tue, 30 Jun 2020 21:53:04 GMT):
Build for the cherry pick rollup is green - https://github.com/hyperledger/besu/pull/1179 - Can I get a review?
joshuafernandes (Tue, 30 Jun 2020 22:34:39 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.5.0 until further notice
joshuafernandes (Tue, 30 Jun 2020 23:22:15 GMT):
@here - This is now complete
trent.mohay (Wed, 01 Jul 2020 00:41:07 GMT):
My gut feel says Immutables won't change the amount of work we do - as our IDE does much of the heavy lift. But - if people feel like this will reduce overall tech debt and complexity, sounds good.
Do we need to consider if this affects PRs etc - i.e. will PRs be rejected if they don't use Immutables where possible? I.e. Is this part of our Coding standards?
davemec (Wed, 01 Jul 2020 00:50:46 GMT):
I would vote for this becoming a part of the coding standards if we move forward with using it. Otherwise the code can end up being hard to maintain if it is mixed in here and there.
jframe (Wed, 01 Jul 2020 01:46:14 GMT):
The immutables approach is much nicer with it not modifying existing classes and having the generated source code there. Wasn't too much trouble to get work with intellij though I did have to ask intellij to rebuild even after a gradle build so this might be an ongoing nuisance. So think it's going to be ok to use, not so sold on the benefits but the downsides that were there with lombok are gone for most part.
lehors (Wed, 01 Jul 2020 05:15:29 GMT):
Has left the channel.
abdelhamidbakhta (Wed, 01 Jul 2020 05:37:06 GMT):
Ok. Good to see that. Thanks for your patience and your feedback. We will move forward with Immutables then
macfarla (Thu, 02 Jul 2020 01:50:25 GMT):
PR to ignore test that is failing on unrelated PRs. It's being investigated but in the meantime it's impacting other PRs so as per discussion above propose ignoring for now https://github.com/hyperledger/besu/pull/1187
macfarla (Thu, 02 Jul 2020 06:49:38 GMT):
Release Philosophy added to wiki https://wiki.hyperledger.org/display/BESU/Release+Philosophy - additions/corrections welcome
sownak (Fri, 03 Jul 2020 16:15:46 GMT):
The blockchain-automation-framework team just published their approach for Besu on kubernetes. Have a look here https://blockchain-automation-framework.readthedocs.io/en/develop/architectureref/hyperledger-besu.html . Any comments are welcome.
shemnon (Mon, 06 Jul 2020 16:50:23 GMT):
Besu contributor call in ~22 hours (EMEA/AMER friendly times)
Add your agenda items to the agenda
shemnon (Mon, 06 Jul 2020 16:50:23 GMT):
Besu contributor call in ~22 hours (EMEA/AMER friendly times)
https://wiki.hyperledger.org/display/BESU/2020-07-07+Contributor+Call
Add your agenda items to the agenda
joshuafernandes (Wed, 08 Jul 2020 23:48:48 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.5.0 until further notice
joshuafernandes (Thu, 09 Jul 2020 00:35:15 GMT):
@here - It's now safe to merge into release-1.5.0 for hyperledger/besu again.
benjamincburns (Thu, 09 Jul 2020 00:38:20 GMT):
Just wanted to share some info here. Today I did some comparative performance testing to determine the impact of the JVM implementation on Besu performance. I tested OpenJDK 11, Amazon's Corretto, Azul Zulu, and GraalVM Community Edition. Of the ones tested, GraalVM had the best performance. We actually observed a 50 TPS throughput improvement as compared to OpenJDK.
I'm going to do some more testing tomorrow with the enterprise versions of GraalVM, but I wanted to drop a note here in the meantime.
benjamincburns (Thu, 09 Jul 2020 00:38:20 GMT):
Today I did some comparative performance testing to determine the impact of the JVM implementation on Besu performance. I tested OpenJDK 11, Amazon's Corretto, Azul Zulu, and GraalVM Community Edition. Of the ones tested, GraalVM had the best performance. We actually observed a 50 TPS throughput improvement as compared to OpenJDK.
I'm going to do some more testing tomorrow with the enterprise versions of GraalVM, but I wanted to drop a note here in the meantime.
benjamincburns (Thu, 09 Jul 2020 00:40:39 GMT):
All tests were run using Caliper on an IBFT 2.0 network with 4 validator nodes and 1 RPC node. The nodes were run in AWS, all in the same region, but not same AZ. Instance size was c5d.4xlarge.
lakshyakumar (Tue, 14 Jul 2020 08:46:43 GMT):
Has joined the channel.
lakshyakumar (Tue, 14 Jul 2020 08:46:44 GMT):
Hi Contributors,
lakshyakumar (Tue, 14 Jul 2020 08:58:40 GMT):
Hi Contributors,
while implementing the tls logic for Orion transaction Manager, faced an issue with certificate generation. I was following the documentation on the link https://docs.orion.pegasys.tech/en/latest/Tutorials/TLS/ and found an issue with the documentation.
so the nodeurl, clienturl and othernodes should be in a format
nodeurl = "https://
lucassaldanha (Tue, 14 Jul 2020 20:38:52 GMT):
Thanks! I'll take a look.
FYI we have an Orion discord channel for Orion specific topics. Here is the invite link: https://discord.gg/n8m22JK
joshuafernandes (Tue, 14 Jul 2020 22:05:41 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.5.0 until further notice
joshuafernandes (Tue, 14 Jul 2020 23:31:32 GMT):
this is now complete
sumaids (Thu, 16 Jul 2020 08:07:26 GMT):
Hi,
for extraData, we just do rlp( 32 0s + 40 chars of validators concatenated + 132 0s )
right?
I tried it in golang using ethereum-go, but it's not giving same answer as
`besu rlp encode`
sumaids (Thu, 16 Jul 2020 08:07:26 GMT):
sumaids (Thu, 16 Jul 2020 08:08:44 GMT):
I tried
`hexutil.Encode((rlp.EncodeToBytes(32 Os + 40 characters of addresses of nodes + 132 0s))
`
But answer is very different from
`besu rlp encode`
sumaids (Thu, 16 Jul 2020 08:08:44 GMT):
I tried
`hexutil.Encode((rlp.EncodeToBytes(32 Os + 40 characters of addresses of nodes + 132 0s))`
But answer is very different from
`besu rlp encode`
timbeiko (Thu, 16 Jul 2020 23:37:35 GMT):
:besu: :tada: *Besu 1.5 is Out* :tada: :besu:
Official announcement: https://www.hyperledger.org/blog/2020/07/16/announcing-hyperledger-besu-1-5-available-now
Download: https://github.com/hyperledger/besu/releases/tag/1.5.0
shemnon (Mon, 20 Jul 2020 15:56:39 GMT):
Contributor call in ~10 hours. Add your agenda items here - https://wiki.hyperledger.org/display/BESU/2020-07-21+Contributor+Call+Agenda
shemnon (Tue, 21 Jul 2020 00:56:20 GMT):
contributor call in 5 min
shemnon (Thu, 23 Jul 2020 20:13:48 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.5.1 until further notice
shemnon (Thu, 23 Jul 2020 20:16:45 GMT):
@here - We’re beginning the release process for hyperledger/besu, please don’t merge anything to master until further notice
shemnon (Thu, 23 Jul 2020 20:44:36 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
lakshyakumar (Mon, 27 Jul 2020 09:28:20 GMT):
Hi,
While following the besu documentation for deploying the smart contract over the besu network with IBFT consensus, I am getting the issues like method is not defined, I am following the link https://besu.hyperledger.org/en/stable/HowTo/Interact/Client-Libraries/web3js-eea/
Also while performing this request "curl -X POST --data '{"jsonrpc":"2.0","method":"eth_accounts","params":[],"id":53}'
shemnon (Mon, 27 Jul 2020 21:23:12 GMT):
Since IBFT networks don't (currently) have mining rewards you must use an `alloc` section in the genesis. Otherwise there's no way to put ether into the system, so the empty response to `eth_accounts` is expected.
shemnon (Mon, 27 Jul 2020 21:23:29 GMT):
what calls are you getting "method is not defined" from?
mwagner (Tue, 28 Jul 2020 02:09:46 GMT):
@joshuafernandes ping! you around ?
joshuafernandes (Tue, 28 Jul 2020 03:03:38 GMT):
Sorry @mwagner was a bit swaped with some dev work
lakshyakumar (Tue, 28 Jul 2020 03:53:08 GMT):
I was using the web3.eea.sendRawTransaction(contractOptions) function
lakshyakumar (Tue, 28 Jul 2020 03:54:19 GMT):
Then do you guys have a guide to generate the eth accounts to be kept in "alloc" section, for a production network?
rai (Tue, 28 Jul 2020 18:02:24 GMT):
Propostal to add @davemec as a Besu mainainer!
https://github.com/hyperledger/besu/pull/1267
shemnon (Tue, 28 Jul 2020 20:04:47 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-1.5.1 until further notice
shemnon (Tue, 28 Jul 2020 22:07:13 GMT):
here - It's now safe to merge into master for hyperledger/besu again.
shemnon (Tue, 28 Jul 2020 22:07:27 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
timbeiko (Wed, 29 Jul 2020 21:59:17 GMT):
:tada: :besu: *Besu 1.5.1 is now available* :besu: :tada:
Relase notes & download info: https://github.com/hyperledger/besu/releases/1.5.1
lakshyakumar (Mon, 03 Aug 2020 04:00:06 GMT):
Hi,
Can you point me out the guide to generate `eth_accounts` to be kept in `alloc` section of my production network `ibftConfigFile.json`, Thanks.
lakshyakumar (Mon, 03 Aug 2020 10:07:38 GMT):
can we use geth to generate the `eth_account`?
shemnon (Mon, 03 Aug 2020 14:52:14 GMT):
Contirbutor call in ~ 24 hours - EMEA times - https://wiki.hyperledger.org/display/BESU/2020-08-04+Contributor+Call+Agenda
shemnon (Mon, 03 Aug 2020 14:52:14 GMT):
Contributor call in ~ 24 hours - EMEA times - https://wiki.hyperledger.org/display/BESU/2020-08-04+Contributor+Call+Agenda
shemnon (Mon, 03 Aug 2020 17:24:46 GMT):
Besu itself does not support the eth_accounts API (it always returns empty), since it is associated with node controlled keys. It is a deliberate choice not to have besu act like a cryptocurrency wallet.
shemnon (Mon, 03 Aug 2020 17:25:13 GMT):
Consider looking at EthSigner (https://docs.ethsigner.pegasys.tech/en/latest/Reference/API-Methods/#eth_accounts) which is a service you put in front of besu and it supports the wallet methods.
shemnon (Tue, 04 Aug 2020 14:51:56 GMT):
EMEA/AMER contributor call in ~10 minutes - https://consensys.zoom.us/j/199741148
terrencecooke (Tue, 04 Aug 2020 16:45:50 GMT):
Hello Besu contributors, I am looking to make code contributions and I have a specific issue (#544) that I would like to be my first issue. How do I have this issue assigned to me?
shemnon (Tue, 04 Aug 2020 17:15:33 GMT):
I assigned you. Comments in the issue tend to be a good method, but can get lost in the shuffle.
terrencecooke (Tue, 04 Aug 2020 17:50:45 GMT):
Thanks Danno
timbeiko (Thu, 06 Aug 2020 18:15:47 GMT):
Here is a proposal to change when we do release candidates for Besu: https://wiki.hyperledger.org/pages/viewpage.action?pageId=38142364
It was discussed on the last contributor call and agreed upon by attendees. We are now sharing it to get feedback from a broader set of maintainers and contributors :slight_smile:
rjones (Fri, 07 Aug 2020 19:26:09 GMT):
Hi - I've been posting some meeting recordings to YouTube, and I noticed there are no Besu recordings. Am I looking in the wrong spot on the wiki? @Daniela_Barbosa
Daniela_Barbosa (Fri, 07 Aug 2020 19:26:09 GMT):
Has joined the channel.
shemnon (Fri, 07 Aug 2020 21:54:23 GMT):
They are in the meeting notes - https://wiki.hyperledger.org/display/BESU/Notes although I saw we've missed adding them as the roll in since June. Those have been updated.
rjones (Sat, 08 Aug 2020 16:30:11 GMT):
https://www.youtube.com/watch?v=ADhpra5c3rY
jramps9 (Tue, 11 Aug 2020 00:08:52 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT this Weds 8/12. Take a look at the agenda and add items here: https://wiki.hyperledger.org/display/Marketing/2020-08-12+Meeting+notes
aldoborrero (Tue, 11 Aug 2020 08:12:53 GMT):
Has joined the channel.
bmg (Tue, 11 Aug 2020 14:15:52 GMT):
Hi all. I'm trying to add some sub commands, and I need access to properties and other objects that are currently private in the BesuCommand object. I can do some dodgy reflection to work around it, but it's not ideal. Is there some other approach that I'm missing?
bmg (Tue, 11 Aug 2020 14:18:24 GMT):
Ideally it would be nice to have all the cli options defined in BesuCommand public for access in sub commands, and to modify the run() method in BesuCommand to perform all the initialisation but not call startSynchronization(). This would allow sub commands that re-use the parent run method to initialise everything
bmg (Tue, 11 Aug 2020 14:18:24 GMT):
Ideally it would be nice to have all the cli options defined in BesuCommand public for access in sub commands, and to modify the run() method in BesuCommand to perform all the initialisation but not call startSynchronization(). This would allow sub commands to re-use the parent run method or some other public init method to initialise everything
bmg (Tue, 11 Aug 2020 14:21:44 GMT):
Happy to create an issue for this (and implement it) if there isn't some existing approach that I've missed
bmg (Tue, 11 Aug 2020 14:43:51 GMT):
Making Blockchain, WorldState (read only) accessible via the plugin API would be nice also.... oh and being able to register additional json rpc commands.... :)
rai (Tue, 11 Aug 2020 15:10:15 GMT):
An issue describing it how you did here would be most welcome! We can discuss the merits of it in a more permanent way there.
aldoborrero (Tue, 11 Aug 2020 15:10:32 GMT):
Also, continuing on what Brian has said, is there also a possibility of having inside MainnetEvmRegistries class all the factory methods public? For example, in Besu Exflo, we are creating a new internal instance of the EVM to provide additional tracing semantics and we have to resort as well to use reflection: https://github.com/41north/besu-exflo/blob/54d6659f3d0bcdd1fc2fc95b20c06c7d319160fe/ingestion/base/src/main/kotlin/io/exflo/ingestion/tokens/EVMFactory.kt#L112
bmg (Tue, 11 Aug 2020 15:11:23 GMT):
@rai sounds good, I'll drop it in the repo later this evening
aldoborrero (Tue, 11 Aug 2020 15:11:41 GMT):
Better to discuss this in a proper issue as well
shemnon (Wed, 12 Aug 2020 15:46:37 GMT):
Definitely in a github issue so we can get the design right. My thinking is that this can be done with the fork numbers from the genesis file, IIRC those all are in the GenesisConfig but I don't think we have a plugin API for the genesis config.
bmg (Wed, 12 Aug 2020 15:50:59 GMT):
Yeah there is no plugin api for that, we ended up recreating a lot of things separately like protocol schedules and so on independently which is not ideal.
bmg (Wed, 12 Aug 2020 15:50:59 GMT):
Yeah there is no plugin api for that, we ended up recreating a lot of things separately like protocol schedules and so on which is not ideal
bmg (Wed, 12 Aug 2020 15:56:46 GMT):
Will add a few issues in the coming days feeding back in our experiences
shemnon (Wed, 12 Aug 2020 17:06:12 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 12 Aug 2020 17:39:10 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
terrencecooke (Thu, 13 Aug 2020 15:56:19 GMT):
Hello all! I am adding a new jsonrpc method to Besu. Is there a guide or convention used for determining a value for the "id" member of the Request object for each json-rpc method? I don't want to just assign one arbitrarily.
NicolasMassart (Thu, 13 Aug 2020 15:57:59 GMT):
Hi, we've had some issues on the Besu doc build on Read The Docs because of the code block tab plugin. It was deprecated and now it's not working anymore.
So I created a PR to fix this because it required to update all the templating system and the code blocks tab syntax... a lot of changes, I'm allready sorry for the reviewers...
Here is the PR to review: https://github.com/hyperledger/besu-docs/pull/489
Note that until this PR is merged, no other PR will be able to build, so it's a bit important.
NicolasMassart (Thu, 13 Aug 2020 17:29:19 GMT):
This PR is merged and build are now back to normal. I updated all the pending PR so you will be able to fix the few remaining issues (like md lint issues).
Also, new thing, all our PRs are now built by ReadTheDocs and the status is returned to the PR checks list.
It also means that you will be able to preview your PR changes directly in RTD by following the "detail" link in the checks list for all your new commits. No excuse not to check the rendering of your doc changes now ;)
Thanks for all your very useful contributions, I hope this will help making the contribution process even easier (I know testing locally is a bit hard sometimes)
shemnon (Thu, 13 Aug 2020 21:13:19 GMT):
It's whatever they passed in as part of the request - https://www.jsonrpc.org/specification#notification
timbeiko (Thu, 13 Aug 2020 22:31:07 GMT):
:besu: Besu 1.5.2 is now available :besu:
Release notes & download link: https://github.com/hyperledger/besu/releases/1.5.2
Highlights :stars:
* A new (experimental!) state backup and restore option
* Better ethstats support
:besu: Besu 1.5.2 is now available :besu:
Release notes & download link: https://github.com/hyperledger/besu/releases/1.5.2
Highlights :stars:
- A new (experimental!) state backup and restore option
- Better ethstats support
timbeiko (Thu, 13 Aug 2020 22:31:07 GMT):
:besu: Besu 1.5.2 is now available :besu:
Release notes & download link: https://github.com/hyperledger/besu/releases/1.5.2
Highlights :stars:
* A new (experimental!) state backup and restore option
* Better ethstats support
timbeiko (Thu, 13 Aug 2020 22:31:07 GMT):
:besu: Besu 1.5.2 is now available :besu:
Release notes & download link: https://github.com/hyperledger/besu/releases/1.5.2
Highlights :stars:
* A new (early access!) state backup and restore option
* Better ethstats support
NicolasMassart (Fri, 14 Aug 2020 10:16:56 GMT):
Hi everyone, here is another PR for doc repos admins to approve please (doc team is in vacation) https://github.com/hyperledger/besu-docs/pull/492
shemnon (Fri, 14 Aug 2020 14:57:05 GMT):
Done. How long are they on vacation?
terrencecooke (Fri, 14 Aug 2020 16:07:36 GMT):
Thanks for clarifying that shemnon. I wasn't sure.
deployer21 (Sun, 16 Aug 2020 16:59:01 GMT):
Has left the channel.
shemnon (Mon, 17 Aug 2020 16:09:23 GMT):
contributor call in just under 9 hours. Agenda - https://wiki.hyperledger.org/display/BESU/2020-08-18+Contributor+Call+Agenda
shemnon (Tue, 18 Aug 2020 00:51:20 GMT):
contributor call in ~10 minutes - https://consensys.zoom.us/j/524697595
shemnon (Tue, 18 Aug 2020 14:28:00 GMT):
What are the maintainers opinions on CalVer? - https://calver.org/ - Specifically moving Besu to YY.MINOR.MICRO. (Note that this would require a TSC blessing)
macfarla (Wed, 19 Aug 2020 04:17:35 GMT):
does this mean I won't have to run up mkdocs locally?
macfarla (Wed, 19 Aug 2020 04:23:30 GMT):
So instead of 1.5.2 and 1.6.0 we would have something like 20.5.2 and 20.6.0?
macfarla (Wed, 19 Aug 2020 04:25:29 GMT):
PR labels - we do use labels but not the ones that are described here https://wiki.hyperledger.org/display/BESU/Pull+Request+Labels - I propose removing this page and let the labels be self-documenting. Any objections?
shemnon (Wed, 19 Aug 2020 13:58:45 GMT):
I was thinking more like MM.Q, so 20.1, 20.2, 20.3, 20.4, and MM.Q.MICRO for bi-weekly - 20.1.1, 20.1.2, etc.
shemnon (Wed, 19 Aug 2020 13:58:45 GMT):
I was thinking more like YY.Q, so 20.1, 20.2, 20.3, 20.4, and YY.Q.MICRO for bi-weekly - 20.1.1, 20.1.2, etc.
ajsutton (Thu, 20 Aug 2020 02:40:46 GMT):
Makes sense to me.
lakshyakumar (Thu, 20 Aug 2020 05:22:57 GMT):
Hi,
I am trying to enable the tls on besu node in my production network, following the process defined here https://besu.hyperledger.org/en/stable/HowTo/Configure/Configure-TLS/
In the prerequisite section you mention about Besu’s password-protected PKCS #12 keystore, but I am not able to findout the guide which gives some indecation about the content and structure of the keystore. could you please point me to some reference where the structure of the content of the keystore required is described.
Thanks you
madelinemurray (Thu, 20 Aug 2020 07:03:45 GMT):
hi @lakshyakumar - creating a PKCS #12 keystore isn't specific to Besu which is why it's not covered in the Besu documentation. Searching should turn up plenty of guides/examples but let us know if you have any issues getting it set up.
rai (Fri, 21 Aug 2020 15:18:16 GMT):
sounds good to me
cjhare (Mon, 24 Aug 2020 05:29:21 GMT):
Sounds cool :+1:
Is there a particular motivation for suggesting the change?
usmansaleem (Tue, 25 Aug 2020 02:06:14 GMT):
PR review request (minor changes to acceptance tests arguments to capture jvm error logs in case of test failures). https://github.com/hyperledger/besu/pull/1327
macfarla (Tue, 25 Aug 2020 03:30:14 GMT):
Deleted!
macfarla (Tue, 25 Aug 2020 03:30:14 GMT):
PR labels wiki page: Deleted!
shemnon (Tue, 25 Aug 2020 05:17:20 GMT):
review reqeust for EvmTool, needed for EvmLab intragrations (Ethereum Foundation Fuzz Testing)
shemnon (Tue, 25 Aug 2020 05:17:20 GMT):
review reqeust for EvmTool, needed for EvmLab intragrations (Ethereum Foundation Fuzz Testing) - https://github.com/hyperledger/besu/pull/1320
ravinayag (Wed, 26 Aug 2020 12:25:06 GMT):
Has joined the channel.
shemnon (Wed, 26 Aug 2020 17:02:51 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 26 Aug 2020 17:51:40 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
NicolasMassart (Thu, 27 Aug 2020 07:43:00 GMT):
Yes
NicolasMassart (Thu, 27 Aug 2020 07:46:16 GMT):
Screenshot 2020-08-27 at 09.44.10.png
NicolasMassart (Thu, 27 Aug 2020 07:46:29 GMT):
https://github.com/hyperledger/besu-docs/pull/498
NicolasMassart (Thu, 27 Aug 2020 07:54:17 GMT):
We are all back ;)
timbeiko (Thu, 27 Aug 2020 20:12:17 GMT):
:partying_face: :besu: *Besu 1.5.3 is now available* :besu: :partying_face:
Release notes & download link: https://github.com/hyperledger/besu/releases/1.5.3
*Highlights* :star2:
- A long-standing known issue using permissioning with Kubernetes is now fixed :tada:
- Early access DNS support :calling:
- Updates to Onchain Privacy Group Management :man_detective_light_skin_tone:
NicolasMassart (Fri, 28 Aug 2020 07:25:55 GMT):
I updated contribution doc for this https://wiki.hyperledger.org/display/BESU/MkDocs+And+Markdown+Guide#MkDocsAndMarkdownGuide-PreviewwithReadTheDocs
shemnon (Mon, 31 Aug 2020 16:23:03 GMT):
Besu Contributor Call in ~23 hours (EMEA/AMER times) - https://wiki.hyperledger.org/display/BESU/2020-09-01+Contributor+Call
Yazanator (Tue, 01 Sep 2020 15:20:43 GMT):
https://github.com/ethereumclassic/ECIPs/pulls
Yazanator (Tue, 01 Sep 2020 15:21:29 GMT):
that's our ecip repo with open PRs that are being considered for ETC and in turn would be contributed to Besu
Yazanator (Tue, 01 Sep 2020 15:21:49 GMT):
this is the invite link to ETC Discord for anyone interested in asking questions https://discord.gg/dwxb6nf
timbeiko (Tue, 01 Sep 2020 16:06:49 GMT):
There was a request for an AuRa consensus implementation posted in the Github issues a few days ago: https://github.com/hyperledger/besu/issues/1338
While the PegaSys team doesn't have the bandwidth right now to lead an implementation, it's something that's been on our TODO list for a while now :sweat_smile: If anyone wants to take this on, we'd be happy to help with a codebase overview and other support required by implementers to get it done :muscle: !
NicolasMassart (Fri, 04 Sep 2020 09:37:04 GMT):
I also updated the PR template to provide a link template and a new checlist item https://github.com/hyperledger/besu-docs/pull/507
NicolasMassart (Fri, 04 Sep 2020 09:37:31 GMT):
I also updated the PR template to provide a link template and a new checlist item https://github.com/hyperledger/besu-docs/pull/507
lakshyakumar (Mon, 07 Sep 2020 05:45:18 GMT):
Hi Contributors,
On our production besu network when modifying the `extradata` field in `genesis` file from the generated one to the rlp encoded array of validator nodes, the network stoped deploying the smartcontract on the nodes. The faliure in the deployment script caused by the timeout after the "waiting for the transction to be mined" message from the node. The network was working fine before editing the genesis file.
can you suggest some possible causes?
attaching the genesis file after editing.
lakshyakumar (Mon, 07 Sep 2020 05:46:59 GMT):
Hi Contributors,
On our production besu network when modifying the extradata field in genesis file from the generated one to the rlp encoded array of validator nodes, the network stoped deploying the smartcontract on the nodes. The faliure in the deployment script caused by the timeout after the "waiting for the transction to be mined" message from the node. The network was working fine before editing the genesis file.
can you suggest any possible causes?
lakshyakumar (Mon, 07 Sep 2020 05:46:59 GMT):
Hi Contributors,
On our production besu network when modifying the `extradata` field in `genesis` file from the generated one to the rlp encoded array of validator nodes, the network stoped deploying the smartcontract on the nodes. The faliure in the deployment script caused by the timeout after the "waiting for the transction to be mined" message from the node. The network was working fine before editing the genesis file.
can you suggest any possible causes?
lakshyakumar (Mon, 07 Sep 2020 05:46:59 GMT):
Hi Contributors,
On our production besu network when modifying the `extraData` field in `genesis` file from the generated one to the rlp encoded array of validator nodes, the network stoped deploying the smartcontract on the nodes. The faliure in the deployment script caused by the timeout after the "waiting for the transction to be mined" message from the node. The network was working fine before editing the genesis file.
can you suggest any possible causes?
lakshyakumar (Mon, 07 Sep 2020 05:49:56 GMT):
{
"config": {
"chainId": 2018,
"muirglacierblock": 0,
"contractSizeLimit": 2147483647,
"ibft2": {
"blockperiodseconds": 2,
"epochlength": 30000,
"requesttimeoutseconds": 10
}
},
"nonce": "0x0",
"timestamp": "0x58ee40ba",
"gasLimit": "0x1fffffffffffff",
"difficulty": "0x1",
"mixHash": "0x63746963616c2062797a616e74696e65206661756c7420746f6c6572616e6365",
"coinbase": "0x0000000000000000000000000000000000000000",
"extraData": "0xf87ea00000000000000000000000000000000000000000000000000000000000000000f854949cf4a3dc32444966d896599c1cdee6d30ab518db9467d7ca4e1ef022650f17f0c7d780597e6ff657aa94555462c53f31f7d301565c831df5d3669fb56d9d947619c6c0bf475e2cbeed4df81cec109e1da27ba3808400000000c0"
}
here is the genesis file I tried to make, I have used the `"constantinoplefixblock": 0`,
in one of the tries but still failing to get the transaction mined.
shemnon (Tue, 08 Sep 2020 15:16:49 GMT):
So when you changed the genesis, is this prior to mining any blocks? And all validators have the same contracts?
shemnon (Tue, 08 Sep 2020 15:18:24 GMT):
Is this a long lived chain? If you want to add and remove validators you need to use the ibft series of JSON-RPC calls and the other validators need to vote the nodes in/out.
shemnon (Tue, 08 Sep 2020 15:19:11 GMT):
But to deploy a contract you just need to have a peer on the network that talks to one of the validators, and they can introduce the contract to the chain. Try setting `--bootnode=` with all of the enodes of the validator nodes.
shemnon (Tue, 08 Sep 2020 15:21:13 GMT):
or adding them in `static-nodes.json` file for the non-validating peers.
macfarla (Tue, 08 Sep 2020 20:47:36 GMT):
Proposal for modifying the process to becoming a besu-docs maintainer https://wiki.hyperledger.org/display/BESU/Proposal+-+Besu+maintainers+become+docs+maintainers - feedback welcome!
lakshyakumar (Wed, 09 Sep 2020 03:19:54 GMT):
we are creating the `extraData` field with validator information in genesis prior to setup the network.
I have used the static-nodes.json to provide the enode of all the peers, and able to see in the logs that the member peer is connected to all the validator peers. any other possible cause?
lakshyakumar (Wed, 09 Sep 2020 03:19:54 GMT):
we are creating the `extraData` field with validator information in genesis prior to setup the network.
we have used the static-nodes.json to provide the enode of all the peers, and able to see in the logs that the member peer is connected to all the validator peers. any other possible cause?
shemnon (Wed, 09 Sep 2020 05:43:13 GMT):
You can try cranking up the logging level on the node you are sending the deployment transaction to (`--logging=DEBUG` or `--logging=TRACE`). See if it hits the transaction pool, then see if it is marked as invalid for any reason. Could be non-zero gas price and zero account balance, could be nodes are not propagating zero gas transactions, could be something wrong with the contract and it fails execution prior to mining.
rai (Wed, 09 Sep 2020 17:48:54 GMT):
welp
timbeiko (Thu, 10 Sep 2020 23:18:21 GMT):
:besu: :partying_face: *Besu 1.5.4 is out!* :partying_face: :besu:
Release notes & download link: https://github.com/hyperledger/besu/releases/1.5.4
Highlights :eyes:
- A new privacy JSON RPC API, priv_debugGetStateRoot :woman_detective_light_skin_tone:
- Better re-org logging :chains:
- A new debug API, debug_batchSendRawTransaction to send multiple txns with a single call :100:
jramps9 (Fri, 11 Sep 2020 17:42:36 GMT):
Nice! Hyperledger marketing will add to the dev newsletter @timbeiko going out today :)
jramps9 (Mon, 14 Sep 2020 17:36:15 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT on 9/16 this week. Take a look at the agenda and add items here: https://wiki.hyperledger.org/display/Marketing/2020-09-16+Meeting+notes
jorgeRodriguez (Tue, 15 Sep 2020 02:09:33 GMT):
Has joined the channel.
andrewyen (Tue, 15 Sep 2020 06:08:52 GMT):
Has joined the channel.
andrewyen (Tue, 15 Sep 2020 06:08:52 GMT):
Hi everyone!
I would like contribute a tool for Besu(not contribute to Besu Blockchain itself), and where can I summit our projtect?
andrewyen (Tue, 15 Sep 2020 06:09:06 GMT):
Thank you!
shemnon (Thu, 17 Sep 2020 05:05:58 GMT):
Do you have a pubic repo for this tool we can look at? We haven't started a contrib repo yet, nor accepted any other repos in yet, but that's mostly because we haven't had any standalone ccontributions yet.
Librechain (Thu, 17 Sep 2020 09:42:53 GMT):
Has joined the channel.
Librechain (Thu, 17 Sep 2020 09:42:54 GMT):
Hey eveyrone.
Librechain (Thu, 17 Sep 2020 09:45:34 GMT):
I have a fairly substantive undertaking and was looking to hire some Hyperledger Besu developers to work alongside me as we build out this protocol.
shemnon (Thu, 17 Sep 2020 13:53:32 GMT):
#jobs may be a better channel for that discussion.
rjones (Thu, 17 Sep 2020 15:27:30 GMT):
Hi, if you have anything you'd like to see added to the /dev/weekly newsletter, please comment in the next two hours: https://wiki.hyperledger.org/pages/viewpage.action?pageId=39618911
andrewyen (Fri, 18 Sep 2020 08:56:38 GMT):
@shemnon here! https://github.com/bsostech/go-besu
Please have a look :)
shemnon (Fri, 18 Sep 2020 14:41:27 GMT):
replied privately.
andrewyen (Mon, 21 Sep 2020 06:22:37 GMT):
Thanks @shemnon
shemnon (Wed, 23 Sep 2020 17:05:48 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 23 Sep 2020 17:38:16 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
timbeiko (Thu, 24 Sep 2020 20:36:01 GMT):
:besu: :mega: *Besu 1.5.5 is out!* :mega: :besu:
*Release notes & download link: https://github.com/hyperledger/besu/releases/1.5.5*
:star2: Highlights :star2:
- The new version of the web3js-eea library (v0.10) supports the onchain privacy group management changes made in Besu v1.5.3
- Added debug_getBadBlocks JSON-RPC API to analyze and detect consensus flaws.
- Fixed a long-standing issue with getting logs. This will be helpful for people wanting to validate Eth2 deposit contracts off Goerli with Besu.
jramps9 (Fri, 25 Sep 2020 00:02:08 GMT):
nice! @timbeiko will include in the dev weekly newsletter tomorrow
shemnon (Fri, 25 Sep 2020 21:37:53 GMT):
Proposal to change Besu to the CalVer versioning system - https://wiki.hyperledger.org/display/BESU/Proposal+-+CalVer+for+Besu+Releases
shemnon (Fri, 25 Sep 2020 21:38:59 GMT):
We will discuss this at the next two contributor calls (next Tuesday and the 12/13th of October). If you support it or have concerns please add them to the wiki page.
NicolasMassart (Mon, 28 Sep 2020 15:59:49 GMT):
How can we fix the typo in calendar invite? Not a life threat but still would be nice to have "Contrubutor" change to "Contributor" :wink:
rjones (Mon, 28 Sep 2020 16:02:38 GMT):
Done :)
shemnon (Tue, 29 Sep 2020 03:11:31 GMT):
Besu Contributor Call in ~ 12 hours - https://wiki.hyperledger.org/display/BESU/2020-09-29+Besu+Contributor+Call
shemnon (Tue, 29 Sep 2020 22:29:02 GMT):
review request - https://github.com/hyperledger/besu-native/pull/21 - This updates matter-labs' EIP2537 library to include their EIP196/197 support. Once merged and released this will improve performance in Besu.
shemnon (Tue, 29 Sep 2020 22:31:39 GMT):
This is the change in Besu that will follow once we have a new besu-native release - https://github.com/shemnon/besu/commit/a90611b1dbfc3d83bf6e45913ebd550fb74c7d7f
dgt1nsty (Fri, 02 Oct 2020 14:30:42 GMT):
Has joined the channel.
bmg (Wed, 07 Oct 2020 17:55:08 GMT):
Any of you guys been playing around with turbo geth? Seeing some noise about 5 day sync times for archive nodes. Curious to hear any thoughts you guys might have on some of the approaches they're using with respect to the storage layout and approach to the full sync
schroedingerscode (Wed, 07 Oct 2020 18:19:06 GMT):
Has joined the channel.
shemnon (Thu, 08 Oct 2020 03:21:40 GMT):
there's a number of things we need to re-architect to get similar mainnet performance. The first is my "Bonsai Tries" project where we make trie leaves O(1) readable.
shemnon (Thu, 08 Oct 2020 03:22:09 GMT):
The "Silkwork" project is also another thing we are keeping an eye on, standard component APIs between microservices would be a boon to all ethereum like blockchains.
aldoborrero (Thu, 08 Oct 2020 07:28:46 GMT):
https://github.com/AlexeyAkhunov/papers/blob/master/Turbo-Geth-Silkworm.pdf
aldoborrero (Thu, 08 Oct 2020 07:30:15 GMT):
About the Silkwork I haven't found anything, never heard of it before. Can you provide more context @shemnon ? Only I found that PDF from TurboGeth
shemnon (Thu, 08 Oct 2020 15:01:55 GMT):
https://github.com/torquem-ch/silkworm - Apache 2 c++ rewrite.
aldoborrero (Thu, 08 Oct 2020 16:09:26 GMT):
Thank you!
shemnon (Thu, 08 Oct 2020 16:33:28 GMT):
here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Thu, 08 Oct 2020 16:33:43 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
atoulme (Thu, 08 Oct 2020 16:44:00 GMT):
@rai completely forgot - week is crazy here - I have an eth65 sniffer that can spew out transactions at besu automatically.
atoulme (Thu, 08 Oct 2020 16:44:05 GMT):
Wanna try it out sometime?
atoulme (Thu, 08 Oct 2020 16:44:46 GMT):
if so let me know. We can do it live so I can pass on how I do it
atoulme (Thu, 08 Oct 2020 16:45:07 GMT):
I had one issue with genesis hash not being correct. I might need to fix the way I read the dev genesis file.
rai (Thu, 08 Oct 2020 16:47:44 GMT):
That'd be great! rop a meeting on my cal (ratan.sur@consenys.net) for anytime before 12pm Eastern time.
rai (Thu, 08 Oct 2020 16:47:44 GMT):
That'd be great! Drop a meeting on my cal (ratan.sur@consenys.net) for anytime before 12pm Eastern time when things are less crazy.
rai (Thu, 08 Oct 2020 16:47:44 GMT):
@atoulme That'd be great! Drop a meeting on my cal (ratan.sur@consenys.net) for anytime before 12pm Eastern time when things are less crazy.
atoulme (Thu, 08 Oct 2020 16:48:22 GMT):
buddy 12pm EST is my 9am.
rai (Thu, 08 Oct 2020 16:48:30 GMT):
oh, you're not in france?
atoulme (Thu, 08 Oct 2020 16:48:30 GMT):
you really don't want to see me before 9
rai (Thu, 08 Oct 2020 16:48:33 GMT):
haha
atoulme (Thu, 08 Oct 2020 16:49:00 GMT):
Now I have been experimenting with different flavors of smoke. It's awesome.
rai (Thu, 08 Oct 2020 16:49:18 GMT):
So then anytime after 3pm eastern
atoulme (Thu, 08 Oct 2020 16:49:32 GMT):
I feel like you're trying to negotiate
atoulme (Thu, 08 Oct 2020 16:49:58 GMT):
ok. I'll get my life in order and get 30 minutes with you
shemnon (Thu, 08 Oct 2020 17:13:05 GMT):
@here - It’s now safe to merge into master for hyperledger/besu again.
atoulme (Thu, 08 Oct 2020 17:21:35 GMT):
@shemnon am toying with the idea of having separate executables for besu vs besu-tracing. Would it be possible to enable tracing via an env var instead? That would probably be a bit less disruptive. Otherwise I am not sure where we reconcile this to a switch.
atoulme (Thu, 08 Oct 2020 17:21:54 GMT):
And of course tracing disabled by default - no need to turn that on unless the user explicitly asks for it.
shemnon (Thu, 08 Oct 2020 17:21:58 GMT):
Are you willing to do the build file mungling in the gradle script?
atoulme (Thu, 08 Oct 2020 17:22:17 GMT):
@shemnon you know I am. It doesn't even hurt anymore.
shemnon (Thu, 08 Oct 2020 17:22:37 GMT):
Perhaps we should have a run script template we keep in source control.
shemnon (Thu, 08 Oct 2020 17:22:49 GMT):
not sure gradle supports taht.
shemnon (Thu, 08 Oct 2020 17:22:49 GMT):
not sure gradle supports that.
shemnon (Thu, 08 Oct 2020 17:23:03 GMT):
but an env var switch is fine.
atoulme (Thu, 08 Oct 2020 17:23:39 GMT):
that would be cool if gradle was a bit more reliable, indeed. Right now we depend a bit on their application plugin to do the right thing if we upgrade.
atoulme (Thu, 08 Oct 2020 17:23:53 GMT):
We already apply regexp to mangle the start script. This is not different.
shemnon (Thu, 08 Oct 2020 17:27:32 GMT):
Looks like we can template it - https://docs.gradle.org/current/dsl/org.gradle.jvm.application.tasks.CreateStartScripts.html
atoulme (Thu, 08 Oct 2020 17:29:10 GMT):
thx will look
shemnon (Thu, 08 Oct 2020 19:14:49 GMT):
We have an RC2 in two weeks. I'm eyeing this tracing fix and the ETC ECIP1099 changes for being cherry picked into 20.10.0-RC2
shemnon (Thu, 08 Oct 2020 19:25:34 GMT):
To clarify what our proposed shift to CalVer version numbering this is what furure releases would look like for the next two quarterly releases if we are able to keep the 2 week cadence: https://wiki.hyperledger.org/display/BESU/Future+Release+Dates
macfarla (Thu, 08 Oct 2020 23:09:36 GMT):
:pencil: Call for enthusiastic volunteers to become Besu docs maintainers! Comment on this thread if you're keen.
As per this proposal https://wiki.hyperledger.org/display/BESU/Proposal+-+Besu+maintainers+become+docs+maintainers
macfarla (Thu, 08 Oct 2020 23:49:36 GMT):
I hereby volunteer
pinges (Fri, 09 Oct 2020 00:09:54 GMT):
I do volunteer as well
bmg (Fri, 09 Oct 2020 12:29:04 GMT):
@shemnon do you have any links for the bonsai trie stuff, I'd like to learn more
shemnon (Fri, 09 Oct 2020 15:13:29 GMT):
me too
shemnon (Fri, 09 Oct 2020 15:14:29 GMT):
overview doc - still implementing (hopefully by christmas but stuff always gets in the way) https://wiki.hyperledger.org/display/BESU/Bonsai+Tries+Design+Overview
atoulme (Mon, 12 Oct 2020 20:27:56 GMT):
@shemnon the tracing PR is better now, should be ready to review
atoulme (Mon, 12 Oct 2020 20:29:07 GMT):
@rai I have prepared for you a test in Tuweni, that you can enable and run to connect to a Besu node. It then sends pending transactions every 100ms for 10minutes.
atoulme (Mon, 12 Oct 2020 20:29:17 GMT):
https://github.com/apache/incubator-tuweni/blob/master/devp2p-eth/src/integrationTest/kotlin/org/apache/tuweni/devp2p/eth/SendPendingTransactionsTest.kt
shemnon (Mon, 12 Oct 2020 23:21:37 GMT):
Besu Contributor call 1 hour 40 miuntes - https://wiki.hyperledger.org/display/BESU/2020-10-13+Hyperledger+Besu+Contributor+Call
timbeiko (Tue, 13 Oct 2020 14:14:35 GMT):
:wave:
jramps9 (Wed, 14 Oct 2020 14:07:49 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT TODAY. Take a look at the agenda and add items if you'd like here: https://wiki.hyperledger.org/display/Marketing/2020-10-14+Meeting+notes
timbeiko (Wed, 14 Oct 2020 14:42:34 GMT):
:besu: :warning: *Besu 20.10.0-RC1 is out* :warning: :besu:
_This is the first release candidate towards our 20.10.0 so there may still be some rough edges. Stability issues have been observed on mainnet and are being worked on. If you want to try the latest and greatest, this has it, but please wait until 20.10.0 to update production nodes._
:star2: Highlights :star2:
- Support added for the YOLOv2 ephemeral testnet
- Added a few debugging APIs: debug_standardTraceBlockToFile and debug_standardTraceBadBlockToFile
- Added --start-block and --end-block to the blocks import command
- Added support for multi-tenancy when using the early access feature of onchain privacy group management
- Fixed a longstanding eth/65 issue
- … and last but not least, as you’ve probably noticed, Besu is moving to CalVer! Here is a short explanation, and expect a proper blog post soon: https://wiki.hyperledger.org/display/BESU/Proposal+-+CalVer+for+Besu+Releases
shemnon (Wed, 21 Oct 2020 16:49:35 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-20.10.0 until further notice
shemnon (Wed, 21 Oct 2020 17:22:21 GMT):
@here - It's now safe to merge into release-20.10.0 for hyperledger/besu again.
timbeiko (Wed, 21 Oct 2020 18:41:52 GMT):
:besu: :relieved: Besu 20.10.0-RC2 is out :relieved: :besu:
:warning: This is our second release candidate towards our 20.10.0 release. It should be more stable than RC1, but still isn’t intended for production use. :warning:
:star2: Highlights :star2:
- Reverted our eth/65 fix and having it enabled by default as we keep investigating :male-detective:
- Support for the latest ETC fork, Thanos :etc:
- Added the Open Telemetry Java agent to report traces to a remote backend :java:
- Added EvmTool binary to the distribution, which is a CLI that can execute EVM bytecode and execute ethereum state tests :hammer_and_wrench:
You can get the release here: https://github.com/hyperledger/besu/releases/tag/20.10.0-RC2
timbeiko (Wed, 21 Oct 2020 18:41:52 GMT):
:besu: :relieved: Besu 20.10.0-RC2 is out :relieved: :besu:
:warning: This is our second release candidate towards our 20.10.0 release. It should be more stable than RC1, but still isn’t intended for production use. :warning:
:star2: Highlights :star2:
- Reverted our eth/65 fix and having it enabled by default as we keep investigating :man_detective_medium_light_skin_tone:
- Support for the latest ETC fork, Thanos
- Added the Open Telemetry Java agent to report traces to a remote backend
- Added EvmTool binary to the distribution, which is a CLI that can execute EVM bytecode and execute ethereum state tests :hammer_and_wrench:
You can get the release here: https://github.com/hyperledger/besu/releases/tag/20.10.0-RC2
atoulme (Mon, 26 Oct 2020 17:06:41 GMT):
@shemnon I assigned a PR to you re metrics. Lmk if you want to go over it with me.
atoulme (Mon, 26 Oct 2020 17:07:47 GMT):
Would be grateful for a review on https://github.com/hyperledger/besu/pull/1493 (from anyone)
shemnon (Tue, 27 Oct 2020 13:45:40 GMT):
In just over an hour (1:15 to be precise) we will have Besu Office Hours (FKA the contributor call). After business we will have an open discussion on Ethereum Hard Forks. - https://wiki.hyperledger.org/display/BESU/2020-10-27+Office+Hours
jramps9 (Thu, 29 Oct 2020 13:53:25 GMT):
Hello! we're low on project news and updates for the dev/weekly newsletter going out tomorrow 10/30. If there's anything you'd like to suggest, please do so in the comments here: https://wiki.hyperledger.org/pages/viewpage.action?pageId=41584474
shemnon (Thu, 29 Oct 2020 15:05:05 GMT):
I don't think we have anything this week. Thanks for checking in.
crypto_beep (Wed, 04 Nov 2020 17:08:46 GMT):
Has joined the channel.
mateo_ventures (Thu, 05 Nov 2020 17:06:31 GMT):
Has joined the channel.
jramps9 (Mon, 09 Nov 2020 18:13:04 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT on 11/11 this week. Take a look at the agenda and add items if you'd like here: https://wiki.hyperledger.org/display/Marketing/2020-11-11+Meeting+notes
shemnon (Mon, 09 Nov 2020 18:46:52 GMT):
We will not be holding the Besu Contributor Call this week. Our next call will be in two weeks on EMEA/AMER times
atoulme (Wed, 11 Nov 2020 18:56:33 GMT):
hey folks, I have opentelemetry metrics working well now here. I would be interested in peppering Besu at strategic places with tracing as well.
Tracing allows you to track how much time is spent in specific places. I see how anything touching merkle trie or the JSON-RPC interface would make sense. Where else would it make sense to add tracing? Would the blockchain classes also make sense like validating blocks?
shemnon (Wed, 11 Nov 2020 19:48:40 GMT):
org.hyperledger.besu.ethereum.mainnet.AbstractBlockProcessor#processBlock
atoulme (Wed, 11 Nov 2020 19:48:52 GMT):
nice
abdelhamidbakhta (Thu, 12 Nov 2020 08:10:44 GMT):
Maybe tracing the processing of transactions could be interesting
abdelhamidbakhta (Thu, 12 Nov 2020 08:10:56 GMT):
https://github.com/hyperledger/besu/blob/9062140452fb054f2487a9096395fee7132a6a0b/ethereum/core/src/main/java/org/hyperledger/besu/ethereum/mainnet/MainnetTransactionProcessor.java#L237
abdelhamidbakhta (Thu, 12 Nov 2020 08:13:25 GMT):
Do you plan to implement a custom OperationTracer and use middleware pattern to inject open tracing if enabled ?
atoulme (Thu, 12 Nov 2020 23:41:16 GMT):
I have opened https://github.com/hyperledger/besu/pull/1557, you can take a look. The code of Otel itself is well optimized to stay out of the way, so it won't affect performance :latin_cross:
atoulme (Thu, 12 Nov 2020 23:41:16 GMT):
I have opened https://github.com/hyperledger/besu/pull/1557, you can take a look. The code of Otel itself is well optimized to stay out of the way, so it won't affect performance :fingers_crossed:
atoulme (Tue, 17 Nov 2020 01:32:07 GMT):
@abdelhamidbakhta shall I volunteer you for review?
atoulme (Tue, 17 Nov 2020 21:03:49 GMT):
qui ne dit mot consent - I'll add you to reviewers
shemnon (Tue, 17 Nov 2020 22:57:03 GMT):
We're queueing up a release for tomorrow, so we are a bit full. I'll try and get the review tomorrow evening.
shemnon (Tue, 17 Nov 2020 22:58:01 GMT):
Have you done any load testing? I'm particulraly concerned about the wrapping of the trie node reads and DB reads. We once had prometheus metrics there and the performance impact was very noticeable, and it got worse when metrics were turned on.
shemnon (Tue, 17 Nov 2020 22:58:52 GMT):
anything that is expected to be sub-millisecond I think are bad candidates for tracing.
atoulme (Tue, 17 Nov 2020 23:01:03 GMT):
I have soaked this code with a mainnet sync
shemnon (Tue, 17 Nov 2020 23:11:50 GMT):
how long did it take?
atoulme (Tue, 17 Nov 2020 23:48:25 GMT):
It ran out of disk after 200Gb, roughly 48h in. It didn't seem to have much problems. It was also monitored by our ethlogger tool which kept asking for every block and transaction along the way.
atoulme (Tue, 17 Nov 2020 23:48:51 GMT):
Do you have specs for how you run nodes for this type of testing? Type of instance, how much disk?
shemnon (Wed, 18 Nov 2020 05:16:30 GMT):
300Gib (growing like mainnet grows), 2.5 days.
shemnon (Wed, 18 Nov 2020 05:16:43 GMT):
Those numbers make me concerned it was slower.
shemnon (Wed, 18 Nov 2020 05:17:08 GMT):
A Goerli full sync is a good measure. 8 hours/80 Gib on a mac.
atoulme (Wed, 18 Nov 2020 05:30:47 GMT):
I don’t have the detail of when it filled the disk. Note also that having the json rpc calls probably ate some of the cpu
atoulme (Wed, 18 Nov 2020 05:31:02 GMT):
So I can rerun if you like
shemnon (Wed, 18 Nov 2020 17:35:27 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 18 Nov 2020 18:28:48 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
timbeiko (Thu, 19 Nov 2020 20:58:21 GMT):
:besu: *Besu 20.10.1 is out* :besu:
:star2: Highlights :star2:
- Two new JSON RPC endpoints to get miner rewards data, `eth_getMinerDataByBlockHash` & `eth_getMinerDataByBlockNumber`
- A new JSON RPC endpoint to change the gas limit, `miner_changeTargetGasLimit`
- The transaction pool now prioritizes transactions by gas prices
- EVMTool is now part of the binary distribution
- We now provide a new metric to track performance during block import: MegaGas/Seconds
- We have support for OpenTelemetry metrics
- We added support for qip714block config parameter in genesis file, paving the way towards permissioning interoperability between Besu and GoQuorum.
You can get the release here: https://github.com/hyperledger/besu/releases/tag/20.10.1
jramps9 (Fri, 20 Nov 2020 13:44:58 GMT):
nice! we will add to the "turkey edition" of dev/weekly today @timbeiko :-)
shemnon (Tue, 24 Nov 2020 02:44:58 GMT):
Besu Office Hours in ~12 hours - https://wiki.hyperledger.org/display/BESU/2020-11-24+Office+Hours
shemnon (Tue, 24 Nov 2020 15:03:16 GMT):
office hours for next 30 minutes - https://consensys.zoom.us/j/199741148
atoulme (Wed, 25 Nov 2020 22:06:53 GMT):
This is a bug fix for the latest release for a bug specifically on Windows I introduced: https://github.com/hyperledger/besu/pull/1608
It comes with a new CircleCI build that also checks the script on Windows, so this doesn't happen again.
atoulme (Mon, 30 Nov 2020 19:36:59 GMT):
https://github.com/hyperledger/besu/pull/1635 is up for review
shemnon (Wed, 02 Dec 2020 17:31:43 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 02 Dec 2020 18:44:15 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
timbeiko (Thu, 03 Dec 2020 22:20:27 GMT):
:besu: Besu 20.10.2 is out :besu:
:star2: Highlights :star2:
- Added support for batched requests in WebSockets. #1583
- Added protocols section to admin_peers to provide info about peer health. #1582
- Added CLI option --goquorum-compatibility-enabled to enable GoQuorum compatibility mode. #1598. (This mode is incompatible with Mainnet)
- Fixed a longstanding issue with eth/65. It should now work on mainnet, and will likely be on by default starting from 20.10.3 :tada:
jramps9 (Mon, 07 Dec 2020 18:00:02 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT on 12/9 this week. Take a look at the agenda and add items if you'd like here: https://wiki.hyperledger.org/display/Marketing/2020-12-09+Meeting+notes
jagpreet (Thu, 10 Dec 2020 03:48:29 GMT):
Has joined the channel.
jagpreet (Thu, 10 Dec 2020 03:48:30 GMT):
Hi team.
When I am trying to define
`--rpc-http-api=EEA,WEB3,ETH,NET,PRIV,PERSONAL`
It is failing because `PERSONAL` is not found.
Is it not present in HL Besu?
shemnon (Thu, 10 Dec 2020 03:51:11 GMT):
Not in besu. It was a deliberate architectural decision not to put wallet functionality in the core client. If you need a wallet solution ethsigner (https://docs.ethsigner.consensys.net/en/stable/HowTo/Get-Started/Install-Binaries/) provides a JSON-RPC proxy that serves the PERSONAL rpcs and forwards the rest off to the configured node.
jagpreet (Thu, 10 Dec 2020 03:53:16 GMT):
Thanks for a quick reply. I will look into the ethsigner binary.
shemnon (Thu, 10 Dec 2020 23:41:22 GMT):
Do we need https://github.com/hyperledger/besu-permissioning-smart-contracts or should we retire the repo?
rjones (Fri, 11 Dec 2020 00:24:44 GMT):
let me know if you want it archived - since it was never used, deleting it is also an option.
shemnon (Fri, 11 Dec 2020 01:38:08 GMT):
review request - common repo fixes for besu-native
https://github.com/hyperledger/besu-native/pull/25
gen_el (Mon, 14 Dec 2020 09:11:31 GMT):
Has joined the channel.
sauveergoel (Wed, 16 Dec 2020 14:58:40 GMT):
Has joined the channel.
sauveergoel (Wed, 16 Dec 2020 14:58:41 GMT):
Hi, I had question regarding the version support for besu, are we actively developing 1.4.4 and 1.5.5 both or is it that only the support is given for 1.4.4 and active development is on 1.5.5, or is it something else?
shemnon (Wed, 16 Dec 2020 16:30:12 GMT):
Active development is only on the most recent releases, which is 20.10.x at the moment.
shemnon (Wed, 16 Dec 2020 16:30:29 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 16 Dec 2020 16:31:11 GMT):
If critical bug fixes are neede in 1.4.x and 1.5.x we will do them, but the chains should all be compatible with future releases.
shemnon (Wed, 16 Dec 2020 16:31:11 GMT):
If critical bug fixes are needed in 1.4.x and 1.5.x we will do them, but the chains should all be compatible with future releases.
shemnon (Wed, 16 Dec 2020 18:26:17 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
sauveergoel (Thu, 17 Dec 2020 08:19:44 GMT):
Thank you for your response 🙂
timbeiko (Thu, 17 Dec 2020 20:46:45 GMT):
:christmas_tree: :besu: *Besu 20.10.3 is out!* :besu::christmas_tree:
:star2: *Highlights* :star2:
- eth/65 is now enabled by default :tada: !
- We’ve fixed some edge cases with our tracing APIs not aligning with Parity’s
- Block Validation errors are now are at the INFO log level
- Added memory as an option to --key-value-storage. This ephemeral storage is intended for sync testing and debugging.
You can get it here: https://github.com/hyperledger/besu/releases/tag/20.10.3
mackcom (Mon, 21 Dec 2020 15:32:37 GMT):
Hi @timbeiko, nice to meet you, just picking up the thread from the telegram chat. Have a nice holiday!
rjones (Mon, 21 Dec 2020 16:17:01 GMT):
Howdy. [A new automerge feature is in beta at GitHub](https://docs.github.com/en/free-pro-team@latest/github/collaborating-with-issues-and-pull-requests/automatically-merging-a-pull-request); feel free to enable it per-repo if you desire.
madelinemurray (Wed, 30 Dec 2020 03:01:59 GMT):
@atoulme - documenting this (old) PR of yours - https://github.com/hyperledger/besu/pull/715 and have a couple of questions.
1. Is the `--pid-path` CLI option is only relevant when you're using the Docker image shipped with Besu?
2. If you use the Docker image and don't specify the `--pid-path` option, the pid file is written to a tmp file?
3. If you do specify `--pid-path`, the pid file is written to that file?
madelinemurray (Wed, 30 Dec 2020 04:08:51 GMT):
@atoulme - and another one - https://github.com/hyperledger/besu/pull/1492
I'm adding the `--metrics-protocol` CLI option but do we also need equivalent docs on using open telemetry metrics as we have for Prometheus? https://besu.hyperledger.org/en/stable/HowTo/Monitor/Metrics/
atoulme (Sun, 03 Jan 2021 01:38:49 GMT):
hey Madeline, just seeing this
atoulme (Sun, 03 Jan 2021 01:38:49 GMT):
hey Madeline, just seeing this, I'll reply in threads.
atoulme (Sun, 03 Jan 2021 01:39:56 GMT):
1. No, you could use a linux service for example and use that to know the PID of Besu.
atoulme (Sun, 03 Jan 2021 01:41:12 GMT):
2. If you look here, I made the PID path a default part of the besu config for Docker: https://github.com/hyperledger/besu/pull/715/files#diff-f34da55ca08f1a30591d8b0b3e885bcc678537b2a9a4aadea4f190806b374ddcR18 - and it's used in the HEALTHCHECK instruction a few lines below.
atoulme (Sun, 03 Jan 2021 01:41:19 GMT):
3. yes?
atoulme (Sun, 03 Jan 2021 01:44:02 GMT):
sounds like we would yes. That said, we have a few more changes on the way for that. My latest PR here adds some docs: https://github.com/hyperledger/besu/pull/1557/files under docs/tracing
shemnon (Tue, 05 Jan 2021 02:04:33 GMT):
Auto-merge doesn't re-merge, you still have to do that yourself. Still, it's an improvement for when I forget to come back to merge.
NicolasMassart (Mon, 11 Jan 2021 11:18:17 GMT):
Hi, could anyone have a look at this doc site change please?
https://github.com/hyperledger/besu-docs/pull/576
I ask here as it implies putting back Ethical ads back and I don't want anyone to be surprised by that.
Displaying ads was initially required by RTD.org as it's a free OSS hosting plateform and a way to fund them, but the custom template we used was not able to do it. Now that I found a way to add it back, we may be able to show our support to the great work RTD community does.
Thanks.
jramps9 (Mon, 11 Jan 2021 15:26:47 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT on 1/13 this week. Take a look at the agenda and add items if you'd like here: https://wiki.hyperledger.org/display/Marketing/2021-01-13+Meeting+notes
davemec (Wed, 13 Jan 2021 17:18:36 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 13 Jan 2021 17:19:24 GMT):
@here - We’re beginning the release process for hyperledger/besu, please don’t merge anything to master until further notice
shemnon (Wed, 13 Jan 2021 18:03:47 GMT):
@here - It’s now safe to merge into master for hyperledger/besu again.
shemnon (Thu, 14 Jan 2021 21:49:05 GMT):
review request - https://github.com/hyperledger/besu/pull/1766 - version updates
timbeiko (Thu, 14 Jan 2021 23:19:52 GMT):
:besu: *Besu 20.10.4 is out* :besu:
_Note: this is the last bi-weekly release before our quarterly RC._
:star2: Highlights :star2:
- Implemented EIP-778: Ethereum Node Records (ENR) #1680
- Implemented EIP-868: Node Discovery v4 ENR Extension #1721
- Added revert reason to eth_estimateGas RPC call. #1730
You can get it here: https://github.com/hyperledger/besu/releases/tag/20.10.4
atoulme (Fri, 15 Jan 2021 02:48:20 GMT):
done, approved
shemnon (Mon, 18 Jan 2021 19:09:12 GMT):
In ~ 20 hours we will have Hyperledger Office Hours (EMEA/AMER friendly times). In addition to standard agenda items Antoine Toulme will spend some time discussing the new OpenTelemetry additions to Besu. https://wiki.hyperledger.org/display/BESU/2021-01-19+Office+Hours
shemnon (Tue, 19 Jan 2021 15:00:52 GMT):
Besu Office Hours now - https://consensys.zoom.us/j/199741148
diegoll (Tue, 19 Jan 2021 16:37:11 GMT):
Hello @shemnon, would this recording be available?
shemnon (Tue, 19 Jan 2021 16:37:40 GMT):
Yes, once I get the email from zoom I'll link it in the agenda
perusworld (Wed, 20 Jan 2021 02:28:23 GMT):
Has joined the channel.
NicolasMassart (Wed, 20 Jan 2021 17:31:05 GMT):
Hi, can someone look at this? It's a simple one.
If you don't like the idea, please discuss it here.
NicolasMassart (Wed, 20 Jan 2021 17:31:29 GMT):
Hi, can someone look at this? It's a simple one.
If you don't like the idea, please discuss it here.
rjones (Thu, 21 Jan 2021 15:36:20 GMT):
Hi. I propose archiving [this repo](https://github.com/hyperledger/besu-permissioning-smart-contracts) as it has been unused since I created it.
rjones (Thu, 21 Jan 2021 16:53:48 GMT):
Hi - Bintray, as a product, is being sunset, and Besu is one of the projects using it. I've [filed a bug](https://github.com/hyperledger/besu-native/issues/26).
shemnon (Thu, 21 Jan 2021 17:23:58 GMT):
Can you post a similar bug in hyperledger/besu?
rjones (Thu, 21 Jan 2021 17:28:50 GMT):
Sure!
rjones (Thu, 21 Jan 2021 17:35:23 GMT):
https://github.com/hyperledger/besu/issues/1826
shemnon (Thu, 21 Jan 2021 19:01:21 GMT):
thanks
DinDjarin (Fri, 22 Jan 2021 04:23:54 GMT):
Has joined the channel.
shemnon (Fri, 22 Jan 2021 14:13:49 GMT):
Call recording is posted - https://wiki.hyperledger.org/display/BESU/2021-01-19+Office+Hours+Notes
atoulme (Fri, 22 Jan 2021 17:17:16 GMT):
thanks, pushed to twitter
shemnon (Mon, 25 Jan 2021 18:19:54 GMT):
Oracle moved the location of their GraalVM docker containers, which broke CI. To ensure your code still build when submitted make sure you are merged with master that uses the new docker locations: https://github.com/hyperledger/besu/commit/cfd9eb8ea8b9b7ae9c398c788e6f3ad92f821f54
rjones (Mon, 25 Jan 2021 18:26:44 GMT):
Hi, the following repos still use `master` as the default branch: `besu, besu-docs, besu-native, homebrew-besu`. `main` [is the current guidance for naming](https://github.com/github/renaming).
shemnon (Wed, 27 Jan 2021 16:04:20 GMT):
@here - We’re beginning the release process for hyperledger/besu, please don’t merge anything to master until further notice
shemnon (Wed, 27 Jan 2021 18:12:26 GMT):
@here - It’s now safe to merge into master for hyperledger/besu again.
shemnon (Fri, 05 Feb 2021 00:31:19 GMT):
Review Request - https://github.com/hyperledger/besu-native/pull/27/files - Bintray exodus
diegoll (Fri, 05 Feb 2021 02:17:26 GMT):
Hi @shemnon, wouldn't be a barrier for some users to force them to have a github account to build the distribution? Certainly it will not be a problem for someone cloning the repository, but what about someone just downloading the source tarball for building it? I don't know if this is the right place to ask this or if you prefer a comment in the PR
diegoll (Fri, 05 Feb 2021 02:20:29 GMT):
Hi @shemnon, once bintray shuts down, is the idea to get the `org.hyperledger.besu:besu-native-altbn128` artifact from github packages?
atoulme (Fri, 05 Feb 2021 02:22:29 GMT):
Folks, I am opening up https://github.com/hyperledger/besu/pull/1882 to start working on introducing new forms of PoW - namely keccak256 for ETC.
atoulme (Fri, 05 Feb 2021 02:23:15 GMT):
This PR is just a lot of refactoring and introducing more variables, and must not have any functional impact. I mean to push this in first to get an easier time introducing a different PoW later.
shemnon (Fri, 05 Feb 2021 04:23:02 GMT):
So... watch this space. There's an issue with github packages and authentication. It may move to artifactory.
atoulme (Fri, 05 Feb 2021 04:38:01 GMT):
Not seeing tests kick in... what am I missing?
shemnon (Fri, 05 Feb 2021 04:53:12 GMT):
Hmm... triggered for me - https://app.circleci.com/pipelines/github/hyperledger/besu https://app.circleci.com/pipelines/github/hyperledger/besu?branch=pull%2F1884 - could you try a force push with a slightly different commit message?
atoulme (Fri, 05 Feb 2021 05:11:23 GMT):
Good now
atoulme (Fri, 05 Feb 2021 06:47:18 GMT):
Thank you.
diegoll (Fri, 05 Feb 2021 12:01:48 GMT):
yes, that would have been my next question, if that wouldn't force everybody to have a github account. Certainly, it will not be a problem for somebody cloning the repo, but maybe for whom want to build it from the source tarball
rjones (Fri, 05 Feb 2021 14:01:43 GMT):
@shemnon a `git commit --amend -s` with no text change should update the timestamps, yielding a new hash. ```% git log -1 --pretty=oneline
885b4cd993da9a1c8d970b78d07c0d19b3201d68 (HEAD -> main) Update repos
% git commit --amend -s
[main 8eafdbe] Update repos
Date: Tue Jan 26 08:05:05 2021 -0800
11 files changed, 102 insertions(+), 12 deletions(-)
% git log -1 --pretty=oneline
8eafdbe9b73a5eb572cf6e17553d634ecb568674 (HEAD -> main) Update repos
% git diff 885b4cd993da9a1c8d970b78d07c0d19b3201d68 8eafdbe9b73a5eb572cf6e17553d634ecb568674
%
```
jramps9 (Mon, 08 Feb 2021 15:15:04 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT on 2/10 this week. Take a look at the agenda and add items if you'd like here: https://wiki.hyperledger.org/display/Marketing/2021-02-10+Meeting+notes
shemnon (Thu, 11 Feb 2021 16:43:16 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-21.1.0 until further notice
shemnon (Thu, 11 Feb 2021 17:36:25 GMT):
@here - It's now safe to merge into release-21.1.0 for hyperledger/besu again.
jframe (Fri, 12 Feb 2021 03:22:08 GMT):
I've just added a new git submodule for some QBFT reference tests so these tests can live in a seperate repo. When you next update to latest master you will need to do a `git submodule update` to get the new submodule so Besu builds again.
Helen_Garneau (Fri, 12 Feb 2021 13:56:50 GMT):
Has joined the channel.
NicolasMassart (Sat, 13 Feb 2021 12:24:33 GMT):
Could anyone review the PR to update Postman doc links please? https://github.com/hyperledger/besu-docs/pull/603
shemnon (Tue, 16 Feb 2021 01:21:03 GMT):
review request: https://github.com/hyperledger/besu/pull/1917
rjones (Wed, 17 Feb 2021 19:31:44 GMT):
Hi, folks. I'm updating the `maintainers` list and I don't have a valid email for: ```$ grep -Ri grant.noble hyperledger/*
hyperledger/besu-docs/MAINTAINERS.md:| Grant Noble | grantnoble | grantnoble |
```
rjones (Wed, 17 Feb 2021 19:31:44 GMT):
Hi, folks. I'm updating the `maintainers` list and I don't have a valid email for: ```$ grep -Ri grant.noble hyperledger/*
hyperledger/besu-docs/MAINTAINERS.md:| Grant Noble | grantnoble | grantnoble |
$ grep -Ri cjhare hyperledger/*
hyperledger/besu/MAINTAINERS.md:| Chris Hare | CjHare | cjhare |
```
shemnon (Wed, 17 Feb 2021 20:40:06 GMT):
They are not at ConsenSys anymore so I can't bug them on slack.
rjones (Wed, 17 Feb 2021 21:04:14 GMT):
OK. If they aren't active any more, perhaps they become maintainers emeritus?
shemnon (Wed, 17 Feb 2021 21:20:27 GMT):
Yea, we haven't run a cleanup query, I'll be sure to look that over for the quarterly report. There are several with no commits in the past quarter that should be migrated.
NicolasMassart (Mon, 22 Feb 2021 09:49:09 GMT):
Hi, it looks like Circle CI is not running on doc repos PRs since a few days. Did anyone do some changes?
For instance this PR https://github.com/hyperledger/besu-docs/pull/615 is still waiting for CI to be started. There's no workflow started in Circle CI.
However nightly workflow seems to work.
rjones (Mon, 22 Feb 2021 15:51:49 GMT):
It looks like PR 603 is the last time it ran
rjones (Mon, 22 Feb 2021 16:53:50 GMT):
@NicolasMassart I don't see anywhere that stuff has changed. I triggered a re-delivery of the message, and it didn't build
NicolasMassart (Mon, 22 Feb 2021 20:06:25 GMT):
Strange... Let me have a closer look tomorrow.
NicolasMassart (Tue, 23 Feb 2021 07:32:00 GMT):
It seems that it unlocked itself... I won't investigate until it fails again, it's good for now ;)
rjones (Tue, 23 Feb 2021 16:54:38 GMT):
@NicolasMassart I promise I did nothing but re-trigger the event send
NicolasMassart (Tue, 23 Feb 2021 16:56:53 GMT):
It indeed seems that sometimes Circle CI has issues and can't run on the webhook trigger. If retriggering is enough, I'm fine with that.
rjones (Tue, 23 Feb 2021 16:58:31 GMT):
do you have permissions to do that directly?
NicolasMassart (Tue, 23 Feb 2021 17:23:26 GMT):
yes I do on the doc repos.
shemnon (Wed, 24 Feb 2021 17:33:54 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-21.1.0 until further notice
shemnon (Wed, 24 Feb 2021 19:08:01 GMT):
@here - Release is done, It's now safe to merge into hyperledger/besu again.
shemnon (Thu, 25 Feb 2021 17:24:06 GMT):
review request: fix hadolint error. https://github.com/hyperledger/besu/pull/1954
atoulme (Fri, 26 Feb 2021 02:05:19 GMT):
Review request: change graphql handling of missing account: https://github.com/hyperledger/besu/pull/1951
atoulme (Fri, 26 Feb 2021 02:14:34 GMT):
Any chance we can close older pull requests? This one for example: https://github.com/hyperledger/besu/pull/200
atoulme (Fri, 26 Feb 2021 17:27:36 GMT):
Hey folks, besides Danno, anyone with time and interest to review https://github.com/hyperledger/besu/pull/1882 ?Your feedback would be very appreciated.
atoulme (Mon, 01 Mar 2021 00:02:35 GMT):
Review request: Fix how ethproxy handles the initial message https://github.com/hyperledger/besu/pull/1959
atoulme (Tue, 02 Mar 2021 08:33:19 GMT):
Review request: Fix createdContract in graphql https://github.com/hyperledger/besu/pull/1965
atoulme (Tue, 02 Mar 2021 15:40:53 GMT):
@shemnon did you mean to approve with this :white_check_mark: ?
shemnon (Tue, 02 Mar 2021 17:07:46 GMT):
I thought I had approved it. Must have been a different PR. But I went in and approved it.
atoulme (Tue, 02 Mar 2021 17:26:15 GMT):
I have been sending a bunch of PRs. I appreciate you looking through them.
atoulme (Tue, 02 Mar 2021 17:26:51 GMT):
This is another small one: Review request: Do not return errors when no account is present, return a zero balance instead: https://github.com/hyperledger/besu/pull/1951
shemnon (Tue, 02 Mar 2021 20:10:40 GMT):
abdelhamidbakhta
shemnon (Tue, 02 Mar 2021 20:11:01 GMT):
atoulme
shemnon (Tue, 02 Mar 2021 20:12:06 GMT):
cjhare
shemnon (Tue, 02 Mar 2021 20:12:39 GMT):
davemec
shemnon (Tue, 02 Mar 2021 20:12:48 GMT):
mackcom
shemnon (Tue, 02 Mar 2021 20:13:15 GMT):
iikirilov
shemnon (Tue, 02 Mar 2021 20:13:24 GMT):
jframe
shemnon (Tue, 02 Mar 2021 20:14:03 GMT):
joshuafernandes
shemnon (Tue, 02 Mar 2021 20:14:10 GMT):
lucassaldanha
shemnon (Tue, 02 Mar 2021 20:14:36 GMT):
macfarla
shemnon (Tue, 02 Mar 2021 20:14:44 GMT):
madelinemurray
shemnon (Tue, 02 Mar 2021 20:15:27 GMT):
mbaxter
shemnon (Tue, 02 Mar 2021 20:15:56 GMT):
NicolasMassart
shemnon (Tue, 02 Mar 2021 20:16:19 GMT):
pinges
shemnon (Tue, 02 Mar 2021 20:16:57 GMT):
trent.mohay
shemnon (Tue, 02 Mar 2021 20:17:10 GMT):
rai
shemnon (Tue, 02 Mar 2021 20:17:43 GMT):
timbeiko
shemnon (Tue, 02 Mar 2021 20:17:51 GMT):
usmansaleem
shemnon (Tue, 02 Mar 2021 20:19:20 GMT):
I've set all maintainers to moderators of #besu and #besu-contributors. If I missed you ping me. If you don't want privs to do stuff like at-here (the main motivation) let me know and I can roll it back.
atoulme (Tue, 02 Mar 2021 20:28:31 GMT):
I also approved this PR if you're interested in merging: https://github.com/hyperledger/besu/pull/1669\
shemnon (Fri, 05 Mar 2021 05:16:56 GMT):
I should ask @ryjones for a github "merge" emoji, but they are considering a move to matrix off of rocket chat, so I'll wait until then.
atoulme (Fri, 05 Mar 2021 05:17:18 GMT):
yeah
atoulme (Fri, 05 Mar 2021 05:17:32 GMT):
thanks for the reviews
shemnon (Fri, 05 Mar 2021 15:22:29 GMT):
We will need to do an out of band release to get a change like this in for Ethereum mainnet - https://github.com/hyperledger/besu/compare/master...shemnon:cutOff
shemnon (Fri, 05 Mar 2021 15:25:16 GMT):
My proposal:
(a) restore the release-21.1.0 branch at the last release point
(b) wait for reference tests to be released that include something like this https://github.com/ethereum/tests/pull/793
(c) apply just that change against 21.1.0
(d) release Monday as 21.1.1
(e) tweak main branch to be 21.1.2-SNAPSHOT
shemnon (Fri, 05 Mar 2021 15:27:45 GMT):
(d) could be to release just as soon as tests pass.
shemnon (Fri, 05 Mar 2021 15:46:43 GMT):
@here Can we get other maintainers opinion on my proposed 21.1.1 out-of-band release :point_up:
shemnon (Fri, 05 Mar 2021 15:47:10 GMT):
Basically mainnet just dropped 2314 from mainnet.
shemnon (Fri, 05 Mar 2021 15:47:10 GMT):
Basically mainnet just dropped EIP-2314 from mainnet and we need to ship it before ropsten forks.
rai (Fri, 05 Mar 2021 15:49:35 GMT):
sounds good to me
rai (Fri, 05 Mar 2021 15:50:29 GMT):
I'll make the PR to remove 2315 from master and we can cherry-pick it over
shemnon (Fri, 05 Mar 2021 15:52:21 GMT):
Updating the reference tests will block the release however, so that needs to finish first. Because it's Dmitry in Russia it may not happen until much later today since it's somewhere around midnight there.
rai (Fri, 05 Mar 2021 15:53:12 GMT):
For sure, just gonna have the PR so when the reference tests do get updated, we can be ready to pull that into the same PR
timbeiko (Fri, 05 Mar 2021 16:16:22 GMT):
We should also update the 21.1.0 release notes, right?
rai (Fri, 05 Mar 2021 16:25:07 GMT):
yes, I'll include that
atoulme (Fri, 05 Mar 2021 17:06:52 GMT):
this is fine by me
atoulme (Fri, 05 Mar 2021 17:08:33 GMT):
looking for eip-2315 context
atoulme (Fri, 05 Mar 2021 17:08:33 GMT):
looking for eip-2314 context
shemnon (Fri, 05 Mar 2021 17:14:18 GMT):
https://twitter.com/TimBeiko/status/1367848166125494272
atoulme (Fri, 05 Mar 2021 17:14:35 GMT):
oh so it's 2315
atoulme (Fri, 05 Mar 2021 17:14:39 GMT):
I was confused
shemnon (Fri, 05 Mar 2021 17:14:41 GMT):
:popcorn: https://www.youtube.com/watch?v=xWfR-WxjmYg
atoulme (Fri, 05 Mar 2021 17:14:55 GMT):
huh is there a birthday celebrated during that call?
shemnon (Fri, 05 Mar 2021 17:14:57 GMT):
Oh, I typed it wrong. I shouldn't work before a nice hot shower.
atoulme (Fri, 05 Mar 2021 17:15:17 GMT):
I relate
timbeiko (Fri, 05 Mar 2021 17:15:47 GMT):
Is there a way to give `@sajz` write access for the Issues in Besu without her being a maintainer?
timbeiko (Fri, 05 Mar 2021 17:16:07 GMT):
She's the new Besu PM and it will make it much easier for her to do her job if she can at least open/close/move issues on Besu.
timbeiko (Fri, 05 Mar 2021 17:16:22 GMT):
cc: @rjones
atoulme (Fri, 05 Mar 2021 17:16:27 GMT):
BTW and apropos of nothing, I think it might be great going forward to have discussions in the besu team a bit more in the open on this type of things. I didn't see the discussion on discord but saw the comment on ACD. I'd like to get more involved and open to more involvement from others.
atoulme (Fri, 05 Mar 2021 17:16:54 GMT):
@timbeiko we can vote her in as a maintainer
atoulme (Fri, 05 Mar 2021 17:17:20 GMT):
Who is she? Can we meet?
timbeiko (Fri, 05 Mar 2021 17:17:45 GMT):
@atoulme got it. She doesn't have "5 significant contributions", and by virtue of her role probably never will. I (and other Besu PMs) got "lucky" in that we had our contributions in when the docs repo was still bundled, so they counted as "Besu contributions".
shemnon (Fri, 05 Mar 2021 17:18:06 GMT):
I would count changelog wrangling as significant contributions.
timbeiko (Fri, 05 Mar 2021 17:18:09 GMT):
She is taking over my role as a product manager for Besu at ConsenSys, I suspect she'll be at the next Besu contributor call, yes :-)
atoulme (Fri, 05 Mar 2021 17:18:33 GMT):
then she can participate - showing up on calls, helping folks on the besu channel, help with release notes, file issues
atoulme (Fri, 05 Mar 2021 17:18:36 GMT):
and we will vote her in
atoulme (Fri, 05 Mar 2021 17:18:43 GMT):
cause we definitely need the help
timbeiko (Fri, 05 Mar 2021 17:18:56 GMT):
Ok got it :+1: I think that works. Especially if we can count the changelog contributions, which she'll do as part of her role :+1:
atoulme (Fri, 05 Mar 2021 17:19:08 GMT):
awesome
atoulme (Fri, 05 Mar 2021 17:19:32 GMT):
If she makes contribs to the besu-docs repo it helps too
atoulme (Fri, 05 Mar 2021 17:20:24 GMT):
In the mean time sorry - all I have is her handle `sajz` - who is she? Where is she at?
shemnon (Fri, 05 Mar 2021 17:20:27 GMT):
Screen Shot 2021-03-05 at 10.19.23 AM.png
shemnon (Fri, 05 Mar 2021 17:20:29 GMT):
There's a Triage role that allows issue management
timbeiko (Fri, 05 Mar 2021 17:20:52 GMT):
Just sent her the invite :-)
timbeiko (Fri, 05 Mar 2021 17:21:35 GMT):
@shemnon what's the link to that panel?
shemnon (Fri, 05 Mar 2021 17:22:01 GMT):
Settings - Manage Access. I think you need the admin role to see it.
shemnon (Fri, 05 Mar 2021 17:22:34 GMT):
Hence the screenshot
timbeiko (Fri, 05 Mar 2021 17:23:01 GMT):
Got it - don't have that access. Maybe a good topic for the next call. A triage role would also be useful for new engineers joining. I remember David having issues moving his tickets when he started. It seems like a quick fix to remove a lot of friction.
atoulme (Fri, 05 Mar 2021 17:23:25 GMT):
at the Apache foundation there was an equivalent discussion
atoulme (Fri, 05 Mar 2021 17:24:10 GMT):
imo and based off those discussions, it's not just code that counts, showing up counts too. The contribs have to be significant in the sense that they help the project, and that can mean holding calls, helping folks, writing tutorials, docs, filing issues, etc.
rjones (Fri, 05 Mar 2021 17:32:32 GMT):
@timbeiko @grace.hartley I think we have like four parallel conversations here :)
rjones (Fri, 05 Mar 2021 17:36:59 GMT):
take a look at insights, we count stuff like wiki edits and the like.
grace.hartley (Fri, 05 Mar 2021 17:47:49 GMT):
haha yes! thanks for everyone's help here
shemnon (Fri, 05 Mar 2021 20:57:55 GMT):
Changed my mind on going off of 21.1.0 release. Since Monday is only 2 days earlier I think we should just do an early 21.1.1 release
shemnon (Fri, 05 Mar 2021 20:57:58 GMT):
From main.
atoulme (Fri, 05 Mar 2021 22:23:23 GMT):
if you're the one volunteering the release, that's fine. Out of interest, we do have release instructions documented right?
atoulme (Fri, 05 Mar 2021 22:55:56 GMT):
looks like this page is of interest: https://wiki.hyperledger.org/display/BESU/Release+Rotations I don't see release instructions, is there a "release" task in gradle or something?
shemnon (Sat, 06 Mar 2021 01:02:47 GMT):
Theres an checklist we had in a consensys repo to verify all the steps, I'll have to rely on rai or abdel to post a wiki page outlineing all the steps. Basically it's checking in a pr with the reelase version, letting the build run, and then add changelog entries and hash references to a few other repos (docs, homebrew). ConsenSys will also update some canaries they run for their benefit but that doesn't need to be part of the public repo.
shemnon (Sat, 06 Mar 2021 01:03:16 GMT):
the gradle build detects when its a non-snapshot build and does the extra steps around the binary, which IIRC are very few.
atoulme (Sat, 06 Mar 2021 04:02:55 GMT):
yeah would be good to have that out in the open as a wiki page, including the SOP steps such as sending a here to the room about the release starting.
shemnon (Sat, 06 Mar 2021 14:33:31 GMT):
My thought is to do "release parties" over video the weeks we don't do Office Hours. Someone drives the automation and while it's building we have a non-business office hours.
atoulme (Sun, 07 Mar 2021 03:19:51 GMT):
Sure. I’m just a bit busy out there so hard to attend to stuff
shemnon (Mon, 08 Mar 2021 02:09:00 GMT):
review request Remove EIP-2315 from Berlin: https://github.com/hyperledger/besu/pull/1986
rai (Mon, 08 Mar 2021 16:13:20 GMT):
@here - we're beginning the release process for besu, please don't merge anything
mwagner (Mon, 08 Mar 2021 17:21:35 GMT):
Has left the channel.
rai (Mon, 08 Mar 2021 17:23:39 GMT):
@here - the release is done, it's safe to merge to master again
rhart92 (Mon, 08 Mar 2021 23:47:28 GMT):
Has joined the channel.
shemnon (Tue, 09 Mar 2021 15:02:14 GMT):
matkt
shemnon (Wed, 10 Mar 2021 06:39:24 GMT):
review request - Bintray is dead to me - https://github.com/hyperledger/besu/pull/1968
Helen_Garneau (Wed, 10 Mar 2021 12:59:20 GMT):
Hello Besu contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT today. Take a look at the agenda and add items if you'd like here: https://wiki.hyperledger.org/x/Nqx6Ag
Helen_Garneau (Wed, 10 Mar 2021 14:44:51 GMT):
(Please note new zoom info!)
grace.hartley (Wed, 10 Mar 2021 16:36:10 GMT):
User User_2 added by grace.hartley.
grace.hartley (Wed, 10 Mar 2021 16:38:56 GMT):
@SajidaZ
SajidaZ (Wed, 10 Mar 2021 16:49:49 GMT):
Hey @atoulme I just joined RocketChat. I wasn't aware of the time of the Besu Contributors call (which is in about 10minutes) and I have a scheduling conflict. I might be able to connect for the first few minutes to do a quick intro (if appropriate) before switching to my other meeting.
SajidaZ (Wed, 10 Mar 2021 16:49:49 GMT):
Hey @atoulme I just joined RocketChat. I wasn't aware of the time of the Besu Contributors call (which is in about 10minutes) and I have a scheduling conflict. I might be able to connect for the first few minutes to do a quick intro (if appropriate) before switching to my other meeting, or just attend the next one.
NicolasMassart (Wed, 10 Mar 2021 16:58:57 GMT):
Hi team, could you review https://github.com/hyperledger/besu-docs/pull/622 please?
atoulme (Wed, 10 Mar 2021 17:04:37 GMT):
Nice to meet you. We have a contributors call right now?
shemnon (Thu, 11 Mar 2021 05:25:59 GMT):
review request, Day 2 - Bintray is dead to me - https://github.com/hyperledger/besu/pull/1968
atoulme (Thu, 11 Mar 2021 08:20:08 GMT):
review request: Forbid unprotected transactions by default - https://github.com/hyperledger/besu/pull/1980
usmansaleem (Sat, 13 Mar 2021 00:59:39 GMT):
review request: Update EthSigner to latest version to be used in acceptance tests. https://github.com/hyperledger/besu/pull/1987
shemnon (Sat, 13 Mar 2021 14:38:15 GMT):
https://github.com/hyperledger/besu/pull/2015 fixes a consensus issue on Ropsten. Release party on Monday?
atoulme (Sat, 13 Mar 2021 19:41:51 GMT):
looks like a few other fixes are required?
atoulme (Sat, 13 Mar 2021 19:41:58 GMT):
not sure what the situation is
shemnon (Sat, 13 Mar 2021 20:10:56 GMT):
Just that one to bring the nodes back into consensus. The only must have line is removing it from the protocol spec. The rest were to disassociate BLS12 from the berlin fork, but keep it in as it looks likely for London.
atoulme (Sun, 14 Mar 2021 02:02:03 GMT):
what about this one? https://github.com/hyperledger/besu/pull/2014
shemnon (Sun, 14 Mar 2021 02:26:10 GMT):
Wasn't aware of it. That would be in the release too.
shemnon (Sun, 14 Mar 2021 17:04:47 GMT):
Review request: add Hyperledger standard RepoLinter - https://github.com/hyperledger/besu-native/pull/34
Since it's a PR from outside the repo GitHub won't execute it until merged. This is a security consideration on GitHub's part. Ignore PR #35 in the mean time.
atoulme (Sun, 14 Mar 2021 17:19:01 GMT):
Approved
atoulme (Sun, 14 Mar 2021 17:20:27 GMT):
Review request: add Astor testnet https://github.com/hyperledger/besu/pull/2017
shemnon (Sun, 14 Mar 2021 18:29:14 GMT):
Didn't take. I had a speling error - https://github.com/hyperledger/besu-native/pull/35
atoulme (Sun, 14 Mar 2021 18:46:47 GMT):
approved
rai (Mon, 15 Mar 2021 13:54:31 GMT):
I think a release today makes sense 😅
rai (Mon, 15 Mar 2021 14:00:56 GMT):
We don't have it our release process that we do anything special for the version number in an out of band release, correct? So it'd just be 21.1.2 this time?
matkt (Mon, 15 Mar 2021 15:08:49 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Mon, 15 Mar 2021 15:35:46 GMT):
Only if we are releasing a bugfix only version. Were this a mainnet consensus issue it would be worth doing a cherry-pick only release.
shemnon (Mon, 15 Mar 2021 15:36:20 GMT):
but I'm 100% ok with a release from main, with all accumulated change
atoulme (Mon, 15 Mar 2021 16:02:41 GMT):
Yes please
rai (Mon, 15 Mar 2021 19:51:50 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
lucassaldanha (Mon, 15 Mar 2021 22:20:28 GMT):
I noticed that our `besu --version` command now is printing some Log4j msg before printing the version. I believe this was unintentional. Am I the only one with this behaviour?
```
$ besu --version
2021-03-16 11:18:14,249 main INFO Registered Log4j as the java.util.logging.LogManager.
besu/v21.1.2/osx-x86_64/oracle_openjdk-java-11
```
shemnon (Tue, 16 Mar 2021 00:12:37 GMT):
https://github.com/hyperledger/besu/pull/1678
shemnon (Tue, 16 Mar 2021 00:13:11 GMT):
My thinking is you have a customized Log4J config that it being pulled in via ENV vars somehow.
lucassaldanha (Tue, 16 Mar 2021 01:05:57 GMT):
I have a custom log4j config. But I wasn't getting this message on previous versions.
lucassaldanha (Tue, 16 Mar 2021 01:06:26 GMT):
As long as this is something we are aware it guess it is ok :)
shemnon (Tue, 16 Mar 2021 01:59:10 GMT):
in your custom log4j `s/status="INFO"/status="WARN"` - only log4j configurations of WARN or higher will go to console.
shemnon (Tue, 16 Mar 2021 01:59:10 GMT):
in your custom log4j `s/status="INFO"/status="WARN"/g` - only log4j configurations of WARN or higher will go to console.
macfarla (Tue, 16 Mar 2021 04:02:42 GMT):
Was the last office hours call recorded?
macfarla (Tue, 16 Mar 2021 04:05:06 GMT):
Also we have two pages on the release schedule - maybe it makes sense to combine them?
- this one https://wiki.hyperledger.org/display/BESU/Future+Release+Dates
- and this one https://wiki.hyperledger.org/display/BESU/Release+Rotations
macfarla (Tue, 16 Mar 2021 04:05:51 GMT):
ah this is still in the future, I can wait :)
NicolasMassart (Tue, 16 Mar 2021 13:53:17 GMT):
it's in one hour
timbeiko (Tue, 16 Mar 2021 14:49:26 GMT):
:de::rotating_light: Besu Berlin Bug :rotating_light::de:
We found an issue related to the Berlin upgrade in Besu 21.1.0 and 21.1.1. If you are using Besu, please download 21.1.2 and upgrade your node as soon as possible. The Goerli upgrade is scheduled for less than 24h from now.
https://github.com/hyperledger/besu/releases/tag/21.1.2
timbeiko (Tue, 16 Mar 2021 14:49:26 GMT):
:de::rotating_light: Besu Berlin Bug :rotating_light::de:
We found an issue related to the Berlin upgrade in Besu 21.1.0 and 21.1.1. If you are using Besu, please download 21.1.2 and upgrade your node as soon as possible. The Goerli upgrade is scheduled for less than 24h from now.
https://github.com/hyperledger/besu/releases/tag/21.1.2
timbeiko (Tue, 16 Mar 2021 14:49:26 GMT):
:de::rotating_light: Besu Berlin Bug :rotating_light::de:
We found an issue related to the Berlin upgrade in Besu 21.1.0 and 21.1.1. If you are using Besu, please download 21.1.2 and upgrade your node as soon as possible. The Goerli upgrade is scheduled for less than 24h from now.
https://github.com/hyperledger/besu/releases/tag/21.1.2
shemnon (Tue, 16 Mar 2021 16:12:41 GMT):
shemnon (Tue, 16 Mar 2021 16:16:25 GMT):
Office Hours Question: Do we want to change meeting times/days?
Since we are going to have to update calendar invites with new meeting codes this is a good time to ask the current maintainers if they would prefer different times and or days for the office hours call.
I think we will still need at least two meetings as we have 3 major time zone areas so support (AMER, EMEA, AU/NZ) but if we want to make dramatic changes now is the time.
atoulme (Tue, 16 Mar 2021 16:48:19 GMT):
It's a bit confusing - sorry I could not join today, had a conflict. I know you mean to have office hours to help folks with Besu - is that working right now?
atoulme (Tue, 16 Mar 2021 16:48:51 GMT):
Can we engage in a good way where we have meetings to discuss specific things?
shemnon (Tue, 16 Mar 2021 16:55:52 GMT):
I am a bit concerned about meeting fatigue, so we've been trying to keep the maintainer business and open call on the same call, called "office hours". Rather than a meeting a week alternating between maintainers and community.
shemnon (Tue, 16 Mar 2021 16:56:19 GMT):
Should we do a separate community call? I was thinking that it followed demand rather than preceeded it.
atoulme (Tue, 16 Mar 2021 18:04:55 GMT):
so, meetings are hard. The Apache folks tend to default to email for that reason. I also have seen a good stride of improvement in the Apache Kafka community where they use the wiki for design discussions.
atoulme (Tue, 16 Mar 2021 18:07:34 GMT):
That doesn't really answer what you're asking here. I think doing meetings is hard and somewhat wasteful, but they can be good if there is an agenda, participation and moderation. Do we *need* meetings right now?
atoulme (Tue, 16 Mar 2021 18:09:11 GMT):
maybe we could do a review of issues and backlog and create a roadmap for the project, that'd be great.
shemnon (Tue, 16 Mar 2021 18:30:08 GMT):
the TSC likes to have each project have at least some in-video session. And I think it's better for "making it real" to have the video sessions.
shemnon (Tue, 16 Mar 2021 18:30:19 GMT):
As for email, nobody in ethereum uses email.
shemnon (Tue, 16 Mar 2021 18:31:08 GMT):
My concern of not having an interactive meeting is that we will split into multiple small groups and randomly add things without any co-ordination. And that also gives a greater gravitation pull to only working with people at your own company.
shemnon (Tue, 16 Mar 2021 18:31:31 GMT):
Hence the request to have a roadmap discussion at the next two calls.
shemnon (Tue, 16 Mar 2021 18:32:06 GMT):
It's also a lot easier to do support type tasks interatively on a zoom call with a screen share than playing "guess the wrong config item" via chat.
shemnon (Tue, 16 Mar 2021 18:32:21 GMT):
But we have yet to have a community member come on a call seeking support.
rjones (Tue, 16 Mar 2021 18:52:46 GMT):
Another project did drop community calls when there was nobody showing up. I'm not taking a position, just pointing out there is precedent.
rjones (Tue, 16 Mar 2021 18:53:22 GMT):
I would be interested in the uptake of the Cactus pair programming sessions that happen every weekday morning Pacific, and if that might be a model
shemnon (Tue, 16 Mar 2021 18:58:12 GMT):
Which project?
atoulme (Tue, 16 Mar 2021 20:00:47 GMT):
"nobody in ethereum uses email" meh, that's not a great reason, and the folks Besu targets are enterprise devs right?
atoulme (Tue, 16 Mar 2021 20:01:44 GMT):
Those enterprise devs cannot run zoom on their locked down laptops. Or cannot use slack even. Or rocket.chat for that matter. They can maybe use the web interface of rocket.chat, but there's no archives.
Email is still it.
shemnon (Tue, 16 Mar 2021 20:06:54 GMT):
These are enterprise devs, not military contractors. They've come on zoom, are on rocket chat. And nobody uses our email list.
shemnon (Tue, 16 Mar 2021 20:11:09 GMT):
If there was a demand for email there would be more mails being sent. We've had the email list since day 1
shemnon (Tue, 16 Mar 2021 20:11:53 GMT):
(Of course it's always day 1 at amazon... :grin:)
atoulme (Tue, 16 Mar 2021 20:26:19 GMT):
really? I didn't know there was one...
atoulme (Tue, 16 Mar 2021 20:27:31 GMT):
OK I found it on the web page at the bottom. How does one subscribe to it?
atoulme (Tue, 16 Mar 2021 20:29:28 GMT):
Who is currently receiving email sent to this list?
atoulme (Tue, 16 Mar 2021 20:30:38 GMT):
One more question now that I'm on the website: https://www.hyperledger.org/blog/2020/11/04/announcing-hyperledger-besu-20-10-0 - who sends those blog posts? Looks like we're a few releases behind.
shemnon (Tue, 16 Mar 2021 20:36:09 GMT):
That's under the domain of the Hyperledger marketing committe. We probably need a new quarterly post.
shemnon (Tue, 16 Mar 2021 20:37:05 GMT):
doing a post for each bi-weekly seems excessive.
shemnon (Tue, 16 Mar 2021 20:54:21 GMT):
Click the "join this group" button here - https://lists.hyperledger.org/g/besu
shemnon (Tue, 16 Mar 2021 20:54:41 GMT):
or send a mail here - besu@lists.hyperledger.org
atoulme (Wed, 17 Mar 2021 03:42:18 GMT):
I have definitely worked with financial institutions that don’t allow for zoom, slack and/or Rocket.Chat. I am not making this up, it happened this year :shrug:
NicolasMassart (Wed, 17 Mar 2021 09:48:59 GMT):
@shemnon @lehors I created a PR to update the doc repos checks according to TSC linting requirements https://github.com/hyperledger/besu-docs/pull/640
Could you review it please?
lehors (Wed, 17 Mar 2021 09:48:59 GMT):
Has joined the channel.
joshuafernandes (Thu, 18 Mar 2021 04:10:28 GMT):
@here: The head of `master` branch for repository `hyperledger/besu` has been rebased to deal with a DCO issue, please follow the instructions on https://wiki.hyperledger.org/pages/viewpage.action?pageId=24781754 to fix your local repositories"
NicolasMassart (Thu, 18 Mar 2021 10:36:34 GMT):
Hi, I created a PR to add @alexandratran as a Besu doc maintainer https://github.com/hyperledger/besu-docs/pull/642
Please vote!
alexandratran (Thu, 18 Mar 2021 10:36:34 GMT):
Has joined the channel.
shemnon (Sat, 20 Mar 2021 01:20:44 GMT):
Proposal: Move 4 maintainers to Emeritus status: https://github.com/hyperledger/besu/pull/2048
GarySchulte (Mon, 22 Mar 2021 18:46:32 GMT):
Has joined the channel.
lucassaldanha (Tue, 23 Mar 2021 07:03:51 GMT):
I'll revert commit a43e848e0 because we suspect it might have introduced a bug in Besu. If you are using snapshots from master, use commit bf54b6c0 until we merge the revert PR.
matkt (Tue, 23 Mar 2021 16:47:27 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
matkt (Tue, 23 Mar 2021 17:21:23 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
atoulme (Wed, 24 Mar 2021 07:47:34 GMT):
https://github.com/hyperledger/besu/pull/1980 is ready for review
alexandratran (Wed, 24 Mar 2021 16:12:36 GMT):
Hi, can someone review this PR? https://github.com/hyperledger/besu-docs/pull/645
shemnon (Sun, 28 Mar 2021 00:57:25 GMT):
Who is running tuesday's office hours?
rhart92 (Mon, 29 Mar 2021 17:16:36 GMT):
Hello! I'm Richard
rhart92 (Mon, 29 Mar 2021 17:17:31 GMT):
I started working on this issue https://github.com/hyperledger/besu/issues/1631
rhart92 (Mon, 29 Mar 2021 17:17:57 GMT):
Is it fine to put up a WIP PR to validate that I am going in the right direction?
rhart92 (Mon, 29 Mar 2021 17:19:22 GMT):
(sorry it's taken so long for me to get around to it)
rhart92 (Mon, 29 Mar 2021 17:19:36 GMT):
I'm excited to help out though if I can
shemnon (Mon, 29 Mar 2021 18:46:59 GMT):
Looks like you got the workflow right. I want the docs team's opinion on the particular CLI flag name. Looks fairly straightforward otherwise.
rhart92 (Mon, 29 Mar 2021 18:48:52 GMT):
Awesome, thanks!
macfarla (Mon, 29 Mar 2021 22:27:28 GMT):
I can run the office hours - rough agenda here if anyone wants to add anything https://wiki.hyperledger.org/display/BESU/2021-03-30+Besu+Contributor+Call
shemnon (Mon, 29 Mar 2021 23:57:17 GMT):
I won't be able to make the call (evening appointment conflict). I would like a roadmap caliber update as to what is going on with QBFT and GoQuorum compatibility. Now that I am no longer at ConsenSys I see just how little is communicated about ConsenSys's intentions in that area and when and what should be expected. If the outcome is to point to some EEA discussions that is fine, but the need to be accessible to people who are not members of the EEA.
macfarla (Tue, 30 Mar 2021 01:02:13 GMT):
Office hours call - has zoom info changed? I'm getting an error 1001
macfarla (Tue, 30 Mar 2021 01:04:25 GMT):
Of course it has because it would have been Danno's meeting... Here's a zoom if anyone wants to office hours with me right now Location:
https://consensys.zoom.us/j/98345085456
macfarla (Tue, 30 Mar 2021 01:27:33 GMT):
Ok I'm calling the office hours done. I've made that zoom meeting recurring and set up auto-record, so it should be fine to use on an ongoing basis.
macfarla (Tue, 30 Mar 2021 03:13:13 GMT):
@shemnon I will follow up on the roadmap update.
arash009 (Wed, 31 Mar 2021 06:36:20 GMT):
@shemnon thanks for raising this. I think we need to do a better job at communicating updates, perhapse the contributor calls are a good place to provide that. Regarding the specific question on QBFT interop with GoQuorum, its still work in progress, but getting closer. I expect this to take another month or so. The two current issues being investigated are: https://github.com/hyperledger/besu/issues/2095 and https://github.com/hyperledger/besu/issues/2094
alexandratran (Thu, 01 Apr 2021 21:21:29 GMT):
Hi everyone, when looking at this issue (https://github.com/hyperledger/besu-docs/issues/614) to document a new metric, I found that only the four standard Ethereum metrics are documented on the relevant page (https://besu.hyperledger.org/en/stable/HowTo/Monitor/Metrics/#view-prometheus-graphical-interface). Looking on Prometheus, there are almost 100 total metrics. Some may be self-explanatory. Is this something that should eventually be documented? Do users need this?
NicolasMassart (Fri, 02 Apr 2021 08:53:15 GMT):
Hi doc maintainers. We have an issue with Javadoc.io not being available for some hours, now more than a day, so it may not be temporary.
It prevents PRs to be merged in doc repos because of the link checking CI test. I will provide a PR to remove these links to be able to keep working on other subjects. Meanwhile I will look for a way to host the javadoc by ourselves as it's obviously not working when we rely on third parties. Will keep you posted on this.
NicolasMassart (Fri, 02 Apr 2021 11:11:42 GMT):
Generating the javadoc in our CI process would enable to have it hosted on github pages and look like this https://nicolasmassart.github.io/besu/latest/
Is this an acceptable solution for you?
NicolasMassart (Fri, 02 Apr 2021 11:14:16 GMT):
by the way I had to fix some html in the javadoc for the p2p module as when running gradle it generates the javadoc for all modules. Here I only published the plugins api one as the others are not relevant.
rjones (Fri, 02 Apr 2021 14:08:01 GMT):
@NicolasMassart could it be a part of https://besu.hyperledger.org/en/stable/ ?
NicolasMassart (Fri, 02 Apr 2021 14:21:16 GMT):
Yes we can do that. But it has some drawbacks too:
- the javadoc has to be merged in the doc repos, requires the besu repos CI to push a PR to the besu-docs repos.
- It requires to find a way decorate the javadoc with header and footer to make the user able to go back to the doc. Probably not impossible. But requires to think a bit about it.
Hosting the javadoc on the besu repos github pages is more straightforward and could be a temporary solution. For now we have no javadoc available at all except if you build it by yourself or download the artifact on jfrog.
rjones (Fri, 02 Apr 2021 14:24:06 GMT):
Hmm, perhaps only by reference then.
NicolasMassart (Fri, 02 Apr 2021 14:24:32 GMT):
It will be the fastest way yes. Will let me time to think about something better...
NicolasMassart (Fri, 02 Apr 2021 14:25:21 GMT):
Also we can't update existing doc versions, only the latest (master)... so links will be broken for stable and all past versions... It was a mistake to rely on such a service...
NicolasMassart (Fri, 02 Apr 2021 14:25:56 GMT):
But we rely on too many 3rd party services anyway, starting with Github.
NicolasMassart (Fri, 02 Apr 2021 14:26:06 GMT):
We should IPFT everything..
NicolasMassart (Fri, 02 Apr 2021 14:26:06 GMT):
We should IPFS everything...
rjones (Fri, 02 Apr 2021 14:31:09 GMT):
regardless, I could make a javadocs.besu... or something point somewhere
NicolasMassart (Fri, 02 Apr 2021 16:18:42 GMT):
I had an answer from the dev who created javadoc.io and the site is back online.
So it will enable us to find something to replace it and having it integrated properly with the current doc site. I created an issue for that. :fingers_crossed:
NicolasMassart (Fri, 02 Apr 2021 16:29:15 GMT):
https://github.com/hyperledger/besu/issues/2102
NicolasMassart (Fri, 02 Apr 2021 16:32:35 GMT):
I'm merging PR to add @alexandratran as a Besu doc maintainer. Can an admin add her in the Github team/group so she has all the maintainer rights please?
NicolasMassart (Fri, 02 Apr 2021 16:46:53 GMT):
@rjones ?
shemnon (Fri, 02 Apr 2021 16:47:46 GMT):
is she already in the hyperledger org?
rjones (Fri, 02 Apr 2021 16:49:05 GMT):
Invite sent
NicolasMassart (Fri, 02 Apr 2021 16:50:47 GMT):
@shemnon no that's the point, I should have been more clear you are right. I can add her in the doc maintainers team but only if she is in the HL org already.
@rjones thanks! I will add her as soon as she accepts the invite.
NicolasMassart (Fri, 02 Apr 2021 16:57:29 GMT):
Done, she is part of the gang now. Thanks!
rjones (Fri, 02 Apr 2021 17:12:13 GMT):
since you guys have team something and team other - who is team `ops`?
alexandratran (Fri, 02 Apr 2021 17:12:41 GMT):
Hi all, can I get an additional review on any of these 4 besu-docs PRs? https://github.com/hyperledger/besu-docs/pulls
atoulme (Fri, 02 Apr 2021 17:19:42 GMT):
I'm not familiar enough to review those, sorry.
NicolasMassart (Tue, 06 Apr 2021 09:15:25 GMT):
What to you need to know from them?
NicolasMassart (Tue, 06 Apr 2021 09:28:45 GMT):
CI for https://github.com/hyperledger/besu/pull/2088 is not running. Could anyone have a look?
VijayMichalik (Tue, 06 Apr 2021 12:12:57 GMT):
What's the correct process for being added to the Hyperledger github org? I think I can't be tagged in a PR on the Besu repo unless I am?
NicolasMassart (Tue, 06 Apr 2021 13:27:04 GMT):
You can be mentioned using @vmichalik but you can't be assigned or approve a PR.
NicolasMassart (Tue, 06 Apr 2021 13:27:40 GMT):
It only requires a Github account be mentioned.
NicolasMassart (Tue, 06 Apr 2021 13:28:39 GMT):
But when you are part of the org, mentioning you is easier as you appear in the suggestion list while you type the mention name.
VijayMichalik (Tue, 06 Apr 2021 13:30:47 GMT):
ok got it, will retry next time this comes up
shemnon (Tue, 06 Apr 2021 14:39:06 GMT):
request to join https://github.com/orgs/hyperledger/teams/besu-triage/members
shemnon (Tue, 06 Apr 2021 14:39:20 GMT):
Madeline will have to apporve it.
rjones (Tue, 06 Apr 2021 15:09:47 GMT):
I was just trying to figure out what the team names mean, I don't need to talk to them
alexandratran (Tue, 06 Apr 2021 16:10:44 GMT):
taken care of. Thx Sally!
VijayMichalik (Tue, 06 Apr 2021 17:39:12 GMT):
I just get a 404 on that page @shemnon!
shemnon (Tue, 06 Apr 2021 18:10:22 GMT):
oh, you need to be in the hyperledger org in general maybe? First see if Madeline can add you manually then failing that we will need to involve RyJones.
NicolasMassart (Wed, 07 Apr 2021 13:46:39 GMT):
Ok, so our devops team is called team Yeti ;)
ricardo.silva (Wed, 07 Apr 2021 15:22:19 GMT):
Has joined the channel.
ricardo.silva (Wed, 07 Apr 2021 15:22:19 GMT):
Hello. `eth_estimateGas` on Besu rejects if we send the `nonce` argument with an error `Caused by: com.fasterxml.jackson.databind.exc.UnrecognizedPropertyException: Unrecognized field "nonce" (class org.hyperledger.besu.ethereum.api.jsonrpc.internal.parameters.JsonCallParameter), not marked as ignorable (11 known properties: "gasPrice", "value", "gasLimit", "gas", "from", "to", "feeCap", "gasPremium", "strict", "data", "payload"])`.
the spec https://eth.wiki/json-rpc/API doesn't specify anything about `nonce`. however, other platforms like go-ethereum/GoQuorum ignore if that argument is sent. This is useful when using Web3j Transaction request factory methods to build the Transaction object (code https://github.com/web3j/web3j/blob/master/core/src/main/java/org/web3j/protocol/core/methods/request/Transaction.java), as we can re-use the same Transaction object to estimate and send. Having the error from Besu means that we need to build the transaction twice.
Would it make sense for Besu to do the same behaviour as Get/GoQuorum? ofc, other option is also create a change on Web3j to force the nonce as optional and/or allow to easily deep copy a transaction object. Thanks!
shemnon (Wed, 07 Apr 2021 16:00:02 GMT):
`eth_call` has a 'strict' option. This is something I think we should add to `eth_estimateGas` and add ignoring nonce checking to the non-strict checks. https://besu.hyperledger.org/en/stable/Reference/API-Methods/#eth_call
ricardo.silva (Thu, 08 Apr 2021 08:59:06 GMT):
@shemnon and would you say that the option should be default `strict` or non-strict?
ricardo.silva (Thu, 08 Apr 2021 09:05:16 GMT):
also, here https://besu.hyperledger.org/en/stable/Reference/API-Objects/#transaction-call-object it mentions that the object for `eth_call` and `eth_estimateGas` already has a `strict` option. so what `eth_estimateGas` does with the struct param at the moment?
ricardo.silva (Thu, 08 Apr 2021 09:05:48 GMT):
User User_1 added by ricardo.silva.
GigoJoseph (Thu, 08 Apr 2021 09:56:34 GMT):
Has joined the channel.
shemnon (Thu, 08 Apr 2021 13:18:26 GMT):
eth_estimateGas should do the same validations, which right now is just skipping the balance check. The bug would be to add nonce checking to the strivt-only checks.
shemnon (Thu, 08 Apr 2021 13:18:26 GMT):
eth_estimateGas should do the same validations, which right now is just skipping the balance check. The bug would be to add nonce checking to the strict-only checks.
ricardo.silva (Thu, 08 Apr 2021 14:02:05 GMT):
thanks @shemnon. I will create a bug issue and explain it on the Besu repo
ricardo.silva (Thu, 08 Apr 2021 14:15:11 GMT):
https://github.com/hyperledger/besu/issues/2125
ricardo.silva (Thu, 08 Apr 2021 14:15:12 GMT):
thank you
rai (Thu, 08 Apr 2021 15:37:00 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
alexandratran (Thu, 08 Apr 2021 15:40:59 GMT):
Hi, can I get an additional review on these two doc PRs? https://github.com/hyperledger/besu-docs/pull/662 and https://github.com/hyperledger/besu-docs/pull/665
rai (Thu, 08 Apr 2021 16:17:41 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
madelinemurray (Thu, 08 Apr 2021 23:30:16 GMT):
@VijayMichalik - have asked Ry to add you to the org. When that's done, I can add you to the triage group.
@SajidaZ - you've got a pending invite to the triage group. When you accept that, you be able to assign, move, etc.
VijayMichalik (Fri, 09 Apr 2021 09:05:32 GMT):
thank you! I've accepted
macfarla (Mon, 12 Apr 2021 05:56:57 GMT):
PR for review - ignore nonce in eth_call and eth_estimateGas params https://github.com/hyperledger/besu/pull/2133
madelinemurray (Mon, 12 Apr 2021 07:49:32 GMT):
@VijayMichalik - are you able to assign and move Besu issues now? @SajidaZ is still getting errors even though she's part of the group.
VijayMichalik (Mon, 12 Apr 2021 14:00:17 GMT):
Yes I believe so!
rjones (Mon, 12 Apr 2021 14:12:02 GMT):
@madelinemurray it looks like both have the same access to four repos
alexandratran (Wed, 14 Apr 2021 17:53:43 GMT):
Can I get additional reviews on these straightforward doc PRs?
https://github.com/hyperledger/besu-docs/pull/674
https://github.com/hyperledger/besu-docs/pull/675
https://github.com/hyperledger/besu-docs/pull/677
atoulme (Wed, 14 Apr 2021 20:32:08 GMT):
looks like they've all been approved :)
alexandratran (Wed, 14 Apr 2021 23:11:42 GMT):
a second tech review from someone outside the docs team is the usual procedure :)
alexandratran (Thu, 15 Apr 2021 00:17:33 GMT):
These two are also ready for review:
https://github.com/hyperledger/besu-docs/pull/678
https://github.com/hyperledger/besu-docs/pull/679
atoulme (Thu, 15 Apr 2021 03:39:23 GMT):
ok I can do that
alexandratran (Thu, 15 Apr 2021 16:08:55 GMT):
Bumping this message here. Does anyone have thoughts on the necessity of documenting metrics?
atoulme (Thu, 15 Apr 2021 17:21:37 GMT):
Absolutely. I would even want to standardize them.
shemnon (Thu, 15 Apr 2021 17:42:44 GMT):
Done - https://eips.ethereum.org/EIPS/eip-2159
atoulme (Thu, 15 Apr 2021 18:28:06 GMT):
all 4 of them?
atoulme (Thu, 15 Apr 2021 18:29:47 GMT):
are there any Besu metrics that are private network specific that would be worth putting out as well?
shemnon (Thu, 15 Apr 2021 18:33:47 GMT):
Those are the only standardized metrics. The EIP review process bordered on punitive (one prolific contributor couldn't restrain himself) so Adrian didn't push for any more.
atoulme (Thu, 15 Apr 2021 18:51:32 GMT):
to be clear - I would not standardize in EIPs. I would standardize with the EEA.
shemnon (Thu, 15 Apr 2021 20:48:10 GMT):
You should start from those four then.
atoulme (Fri, 16 Apr 2021 00:08:05 GMT):
Right, I appreciate the pointer!
NicolasMassart (Fri, 16 Apr 2021 13:26:11 GMT):
Hi there is a new failure en besu-doc repos CI since today.
We have the node js jobs (reposlint, markdownlint, and linkchecker) that generate an error.
```
npm ERR! npm ERR! code 128
npm ERR! npm ERR! command failed
npm ERR! npm ERR! command git ls-remote ssh://git@github.com/stevenvachon/chai-subset.git
npm ERR! npm ERR! command-line line 0: unsupported option "accept-new".
npm ERR! npm ERR! fatal: Could not read from remote repository.
```
See for instance https://app.circleci.com/pipelines/github/hyperledger/besu-docs/2460/workflows/5f215f39-a81e-4083-ae93-b7cda3b0c1bc
But the exact same CI steps markdownlint and linkchecker on repos that are not in the HL org are fine, see for instance checks on https://github.com/ConsenSys/doc.quorumplugins/pull/93
It there anything that changed on HL besu-doc CI config?
NicolasMassart (Fri, 16 Apr 2021 13:26:11 GMT):
Hi there is a new failure en besu-doc repos CI since today.
We have the node js jobs (reposlint, markdownlint, and linkchecker) that generate an error.
```
npm ERR! npm ERR! code 128
npm ERR! npm ERR! command failed
npm ERR! npm ERR! command git ls-remote ssh://git@github.com/stevenvachon/chai-subset.git
npm ERR! npm ERR! command-line line 0: unsupported option "accept-new".
npm ERR! npm ERR! fatal: Could not read from remote repository.
```
See for instance https://app.circleci.com/pipelines/github/hyperledger/besu-docs/2460/workflows/5f215f39-a81e-4083-ae93-b7cda3b0c1bc (a recent PR) or even on master https://app.circleci.com/pipelines/github/hyperledger/besu-docs?branch=master so it's not linked to being in a fork or not.
But the exact same CI steps markdownlint and linkchecker on repos that are not in the HL org are fine, see for instance checks on https://github.com/ConsenSys/doc.quorumplugins/pull/93
It there anything that changed on HL besu-doc CI config?
NicolasMassart (Fri, 16 Apr 2021 13:32:30 GMT):
The only difference I can see at CI level is that we have Docker hub credentials passed as env in Consensys Circle CI config but not on besu-doc. But I'm not usre it's about docker here, seems more related to Git or NPM.
Also when running `git ls-remote ssh://git@github.com/stevenvachon/chai-subset.git` locally it works fine, so why is this command returning an error on CI? :thinking:
rjones (Fri, 16 Apr 2021 16:15:42 GMT):
I might have broken something by accident
rjones (Fri, 16 Apr 2021 16:16:24 GMT):
@NicolasMassart time for a quick zoom call?
bgravenorst (Mon, 19 Apr 2021 06:20:19 GMT):
Has joined the channel.
bgravenorst (Mon, 19 Apr 2021 06:20:20 GMT):
If you enable metrics on Besu, then you can view the available list (with descriptions) at http://
NicolasMassart (Mon, 19 Apr 2021 07:53:17 GMT):
Sorry Ray, time zones... I was already in WE...
Let me know if you are available today.
rjones (Mon, 19 Apr 2021 14:44:43 GMT):
Ok
rjones (Mon, 19 Apr 2021 14:55:20 GMT):
@NicolasMassart I don't think I have access to the CCI stuff, but I also don't think I changed docker hub credentials
ricardo.silva (Mon, 19 Apr 2021 15:58:12 GMT):
Hi. As part of the work of privacy interoperability with Quorum, I have created this PR that runs the Quorum Acceptance tests against Besu. It's added on this PR: https://github.com/hyperledger/besu/pull/2122
at the moment, even if the tests are failing, the job will result in success (I'm ignoring the result). I suggested to merge as it is as then we can work on the fixes in small chunks (on PRs) instead of work on top of this PR.
when all issues are solved, we can remove the line that makes the job succeed.
do you agree with this approach? thank you!
macfarla (Mon, 19 Apr 2021 23:33:27 GMT):
I think it would be good to get this PR merged sooner rather than later. It does add an extra step to the end of the build process (~8min) so will extend build times.
ajsutton (Mon, 19 Apr 2021 23:36:16 GMT):
Given it runs in parallel with acceptanceTests it shouldn't affect build times.
ajsutton (Mon, 19 Apr 2021 23:37:07 GMT):
Sorry I misread, it's the one chained after buildDocker. My suggestion would be to make buildDocker not depend on unitTests to avoid extending the build times.
ajsutton (Mon, 19 Apr 2021 23:37:46 GMT):
But otherwise I agree getting it merged in smaller steps makes a lot of sense.
macfarla (Mon, 19 Apr 2021 23:39:34 GMT):
Ok if it makes sense to make buildDocker run parallel to the tests
ajsutton (Mon, 19 Apr 2021 23:40:02 GMT):
Yeah, as long as we don't publish until we see all tests passing I don't see any reason to delay building the docker image and getting started on the quorum tests.
macfarla (Mon, 19 Apr 2021 23:48:04 GMT):
ok fair enough - when does publish happen? it's not part of the buildDocker step?
ajsutton (Mon, 19 Apr 2021 23:49:00 GMT):
It's done as part of the publishDocker step which doesn't run on PRs at all. It already depends on unitTests explicitly anyway.
macfarla (Mon, 19 Apr 2021 23:49:11 GMT):
got it
ajsutton (Mon, 19 Apr 2021 23:50:08 GMT):
We should add acceptanceTestsQuorum to the list of tasks to wait for before publishing at some point as well. Possibly now or just when we make it actually fail.
macfarla (Mon, 19 Apr 2021 23:52:58 GMT):
ok that makes sense. Will make a note
macfarla (Tue, 20 Apr 2021 00:46:35 GMT):
PR for review - ignore `privatefor` in eth_call and eth_estimateGas params https://github.com/hyperledger/besu/pull/2160
macfarla (Tue, 20 Apr 2021 03:23:48 GMT):
PR for review - move buildDocker to run in parallel with tests https://github.com/hyperledger/besu/pull/2161
ricardo.silva (Tue, 20 Apr 2021 08:52:35 GMT):
thank you! I updated with the latest master https://github.com/hyperledger/besu/pull/2122
NicolasMassart (Tue, 20 Apr 2021 15:30:57 GMT):
I don't know what you did, but now it works... :fingers_crossed:
rjones (Tue, 20 Apr 2021 16:43:03 GMT):
I... did nothing. :)
NicolasMassart (Tue, 20 Apr 2021 19:19:25 GMT):
And it worked :D
matkt (Wed, 21 Apr 2021 15:37:44 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
matkt (Wed, 21 Apr 2021 17:46:56 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
shemnon (Thu, 22 Apr 2021 14:11:14 GMT):
Now that Berlin has deployed, what are maintainer's thoughts on the timing of moving main to master? Perhaps not this weekend but next weekend?
alexandratran (Thu, 22 Apr 2021 19:31:53 GMT):
These docs PR are ready for review, they are all minor changes:
https://github.com/hyperledger/besu-docs/pull/689
https://github.com/hyperledger/besu-docs/pull/688
https://github.com/hyperledger/besu-docs/pull/686
https://github.com/hyperledger/besu-docs/pull/685
ajsutton (Thu, 22 Apr 2021 22:22:07 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=fMhg4rS5bYW7fsDFy) Renaming the git branch master to main you mean? I thought I read somewhere that github could do it in a way that supported redirects so it should be pretty seamless.
ajsutton (Fri, 23 Apr 2021 00:35:34 GMT):
It redirects URLs but not `git pull` so not seamless. It does provide instructions for updating your local checkout though. https://docs.github.com/en/github/administering-a-repository/renaming-a-branch
shemnon (Fri, 23 Apr 2021 00:56:19 GMT):
So the main issue is that going master->main is a TSC directive because of D&I stuff. It needs to happen by the end of Q2.
rjones (Fri, 23 Apr 2021 15:52:59 GMT):
as long as a new branch isn't created with the same name, it does a fixup on the backend.
rjones (Fri, 23 Apr 2021 15:53:46 GMT):
I would go here: https://github.com/hyperledger/besu/branches and edit the branch name directly.
rjones (Fri, 23 Apr 2021 15:54:28 GMT):
If you can't do it due to permissions, I will do it once you tell me to.
shemnon (Fri, 23 Apr 2021 16:24:28 GMT):
We also need to tweak some of the CI automation. And for a little bit until CI settles we may need both. And check branch protection rules tied to master.
rjones (Fri, 23 Apr 2021 16:31:28 GMT):
those move magically
shemnon (Fri, 23 Apr 2021 16:32:18 GMT):
circle-ci has a reference to the master branch.
rjones (Fri, 23 Apr 2021 16:32:22 GMT):
(they will now since I just edited it) :)
rjones (Fri, 23 Apr 2021 16:33:26 GMT):
OK. If the repos aren't long lived, that may be an issue. Let me test it out elsewhere.
rjones (Fri, 23 Apr 2021 16:37:00 GMT):
My test shows it doesn't work the way I thought it would, so... belay that.
atoulme (Tue, 27 Apr 2021 00:04:05 GMT):
Contributor call started here: https://consensys.zoom.us/j/94665568300 (I think, I'm by myself right now)
atoulme (Tue, 27 Apr 2021 00:08:09 GMT):
Going to give it another 5 minutes.
atoulme (Tue, 27 Apr 2021 00:11:32 GMT):
ah, I'm an hour early (curses in timezones) https://wiki.hyperledger.org/display/BESU/2021-04-26+Besu+Office+Hours
shemnon (Tue, 27 Apr 2021 15:30:57 GMT):
@macfarla @trent.mohay The containerTests CI integration has gone very consistently flakey over the past 24 hours. Can this be disabled until it's stable?
alexandratran (Tue, 27 Apr 2021 20:03:52 GMT):
Can I get an additional review on this PR doc'ing `--static-nodes-file`? https://github.com/hyperledger/besu-docs/pull/692
ajsutton (Wed, 28 Apr 2021 00:46:56 GMT):
You've probably seen this through github updates but just to complete the loop back here, the container tests should now have been disabled/removed in CI (https://github.com/hyperledger/besu/pull/2196)
macfarla (Wed, 28 Apr 2021 03:40:37 GMT):
added some comments
rai (Thu, 29 Apr 2021 14:56:33 GMT):
PR to add garyschulte as a maintainer:
rai (Thu, 29 Apr 2021 14:56:35 GMT):
https://github.com/hyperledger/besu/pull/2206
grace.hartley (Tue, 04 May 2021 13:48:20 GMT):
@rjones would you be able to help with setting @GarySchulte and @matkt as admins for the Besu repo?
rjones (Tue, 04 May 2021 19:39:50 GMT):
@grace.hartley done
grace.hartley (Thu, 06 May 2021 14:54:59 GMT):
thank you!!!
antonydenyer (Thu, 06 May 2021 16:54:13 GMT):
Ahoy there, I'm looking to further extend besu in a couple of areas, namely adding/modifying the behaviour of some json-rpc endpoints. In and of itself this shouldn't cause any problems, just implement `JsonRpcMethod` and provide a way to register it via the plugin. The problems start when you want to do something vaguely interesting on the endpoint, let's say you wanted to override `eth_getTransactionCount` to have an optional to parameter you would need to get hold of an instance to `BlockchainQueries`. To do this we add something into the `PluginContext` via `addService` maybe a class with a bunch of things we deem acceptable? But it feels a bit messy, open to ideas.
shemnon (Thu, 06 May 2021 17:03:38 GMT):
That was the design path I was going down. But we wouldn't expose all of the Besu internals, just specific services. BlockchainQueries is somehting I would push against becoming an API.
shemnon (Thu, 06 May 2021 17:05:07 GMT):
Rather, I would prefer an in-process call to other exposed endpoints. Possibly directly through a service, where the service corresponds to standardized eth endpoints. That should get most of what want from BlockchainQueries
shemnon (Thu, 06 May 2021 17:10:34 GMT):
I once had arbitrary RPCs prototyped, but that must have been pre-hyperledger. I'll have to check my old machine to see if I still have it.
shemnon (Thu, 06 May 2021 17:10:34 GMT):
I once had arbitrary RPCs via plugin prototyped, but that must have been pre-hyperledger. I'll have to check my old machine to see if I still have it.
shemnon (Thu, 06 May 2021 17:11:37 GMT):
I was working on how to interact with the CLI flags to turn RPCs on and off, interacting with the built in RPCs. But then Hyperledger integration became my top task.
antonydenyer (Thu, 06 May 2021 17:34:02 GMT):
Yeah, I guess you lose control with what you expose at that point. What do you think would be needed for the example above? Just`WorldStateArchive` I guess?
shemnon (Thu, 06 May 2021 18:17:35 GMT):
Whatever you use has to be through the `plugin-api` module. WorldStateArchive isn't a good candidate for that keyhole.
shemnon (Thu, 06 May 2021 18:20:23 GMT):
PoAQueryService may be the model you need.
atoulme (Sat, 08 May 2021 23:06:16 GMT):
Review please :pray: https://github.com/hyperledger/besu/pull/2252
rjones (Sun, 09 May 2021 05:42:54 GMT):
Howdy - who owns CI/CD for Besu these days?
atoulme (Sun, 09 May 2021 06:16:05 GMT):
not me. Thought it was you?
rjones (Sun, 09 May 2021 16:35:07 GMT):
lol
rjones (Sun, 09 May 2021 16:35:42 GMT):
Who is responsible for the CircleCI setup? I delegated that to @shemnon but I suspect it was handed off
atoulme (Sun, 09 May 2021 20:53:25 GMT):
sorry still not me, but looking, keep pressing on
antonydenyer (Mon, 10 May 2021 08:24:17 GMT):
I ended up paying for CircleCI to get things moving on my fork. It ended up costing about $15 per couple of builds!
antonydenyer (Mon, 10 May 2021 08:24:53 GMT):
speaking of which ... is anyone able to look at these two?
https://github.com/hyperledger/besu/pull/2235
https://github.com/hyperledger/besu/pull/2218
rjones (Mon, 10 May 2021 11:53:47 GMT):
@antonydenyer another good reason to migrate to GitHub Actions for CI :)
rjones (Mon, 10 May 2021 11:59:49 GMT):
@ajsutton I can do it
rjones (Mon, 10 May 2021 12:01:46 GMT):
@matkt I installed this app for Cactus; if you're interested, I can enable it for your Besu repos. https://www.mergefreeze.com/
atoulme (Mon, 10 May 2021 23:00:25 GMT):
@rjones I'm sorry, is that a thing this project is doing, moving to github actions?
atoulme (Mon, 10 May 2021 23:00:29 GMT):
Is there an issue for it?
matkt (Tue, 11 May 2021 15:28:45 GMT):
Oh yes cool. I'll discuss this with the team but IMO this is a very good thing to add
rjones (Tue, 11 May 2021 17:24:57 GMT):
Ok, let me know
macfarla (Tue, 11 May 2021 22:33:25 GMT):
Flaky tests update: 3 changes in Besu which address the causes of recent instability:
1. install `haveged` on AT executor.
2. set the property `securerandom.source=file:/dev/urandom` explicitly for ATs
3. disable ATs for SECP256R1 because there is a static reference to the signature algorithm in KeyPairUtil so this needs some refactoring to handle multiple algos
gortaire (Wed, 12 May 2021 04:52:06 GMT):
Has joined the channel.
roshan13046 (Sat, 15 May 2021 04:21:31 GMT):
Has joined the channel.
roshan13046 (Sat, 15 May 2021 04:21:31 GMT):
Hi! Everyone. Joining this chat to explore and contribute to this Hyperledger-Besu Project.
VijayMichalik (Mon, 17 May 2021 10:14:08 GMT):
Hi Besu contributors - I have a question for discussion:
As a PM and non-code contributor (responding to community questions, attending office hours, adding documentation, triaging issues and improving scoping detail on them), I'd like to get write access so I can be even more useful - particularly because this is required to manage Besu tickets in Zenhub.
It doesn't look like I will be able to achieve this under the current maintainer rules.
Is there any appetite to more formally recognise non-code contributions? As an example of a system for qualifying them - here's Kubernetes' non-code contributor guide: https://github.com/kubernetes/community/blob/master/contributors/guide/non-code-contributions.md
VijayMichalik (Mon, 17 May 2021 10:14:08 GMT):
Hi Besu contributors - I have a question for discussion:
As a PM and non-code contributor (responding to community questions, attending office hours, adding documentation, triaging issues and improving scoping detail and prioritization of them), I'd like to get write access so I can be even more useful - particularly because this is required to manage Besu tickets in Zenhub.
It doesn't look like I will be able to achieve this under the current maintainer rules.
Is there any appetite to more formally recognise non-code contributions? As an example of a system for qualifying them - here's Kubernetes' non-code contributor guide: https://github.com/kubernetes/community/blob/master/contributors/guide/non-code-contributions.md
NicolasMassart (Mon, 17 May 2021 10:38:06 GMT):
Just my opinion on Circle CI vs Github Actions:
- migration will require work (rewrite all workflow to GHActions, use new images, find actions to replace Circle Orbs,...)
- we should move doc CI to Github CI too if we move the code repos -> consistent contrib UX.
- I see the pros: enables contributors outside of HL to run CI. No need to pay. Slightly more flexible and modern tooling. Any others?
- what will be the cons of this? What about centralisation in Github?
- how is it compatible with DCO? (I don't see any reason why not, but as DCO always finds a way to make things complicated, we'd better check)
mdwright88 (Mon, 17 May 2021 15:02:51 GMT):
Has joined the channel.
mdwright88 (Mon, 17 May 2021 15:04:29 GMT):
Amazing! How would you like to contribute? You should join our contributor calls to kick things off!
https://wiki.hyperledger.org/display/BESU/2021+Agendas
shemnon (Mon, 17 May 2021 15:05:52 GMT):
I thought "Triage" level access allowed Github issues editing? IIRC that doesn't require maintainer level access since you can't merge PRs. I also think @ryjones needs to invite you to the Hyperledger org before we can add that.
rjones (Mon, 17 May 2021 15:07:31 GMT):
yup. git me a GitHub username to invite
VijayMichalik (Mon, 17 May 2021 15:22:03 GMT):
I have both the above but the triage permissions are unfortunately not compatible with Zenhub
VijayMichalik (Mon, 17 May 2021 15:22:36 GMT):
https://help.zenhub.com/support/solutions/articles/43000037261-github-permissions-and-access-in-zenhub
shemnon (Mon, 17 May 2021 15:28:03 GMT):
hm... that's a bigger issue then.
VijayMichalik (Mon, 17 May 2021 15:31:03 GMT):
let me know if you have any creative ideas :slight_smile:
shemnon (Mon, 17 May 2021 15:31:50 GMT):
Is this just a zenhub issue? [Github permissions](https://docs.github.com/en/organizations/managing-access-to-your-organizations-repositories/repository-permission-levels-for-an-organization#repository-access-for-each-permission-level) say triage has "Close, reopen, and assign all issues and pull requests"
shemnon (Mon, 17 May 2021 15:32:30 GMT):
Also, "Apply Labels," but not "create, edit delete labels"
rjones (Mon, 17 May 2021 15:34:43 GMT):
My guess is that ZenHub doesn't support the (relatively) new Triage permissions
rjones (Mon, 17 May 2021 15:34:43 GMT):
My guess is that ZenHub doesn't support the (relatively) new Triage or Maintain permissions
atoulme (Mon, 17 May 2021 16:21:26 GMT):
Vijay, we certainly need you. We want folks to help with all the non-code aspects of the project.
atoulme (Mon, 17 May 2021 16:22:27 GMT):
The cons are many for github actions, as I'm discovering through a parallel experience at the ASF. It's not a silver bullet.
atoulme (Mon, 17 May 2021 16:23:03 GMT):
Here is some documentation: https://cwiki.apache.org/confluence/display/BUILDS/GitHub+Actions+status
atoulme (Mon, 17 May 2021 16:25:36 GMT):
Look at the mitigations section especially. You will need to keep a tight lid on what 3rd party actions are made available.
atoulme (Mon, 17 May 2021 16:27:08 GMT):
Show us you want in, it's a simple 1-2-3 process:
1. propose a change to the rules if necessary, we can vote that in
2. Send to this channel 2-3 times you helped with community. Should be the easy part.
3. We will then open a PR and vote you in.
VijayMichalik (Mon, 17 May 2021 16:55:15 GMT):
I submitted a request to Zenhub to ask about aligning the two categories better @shemnon, will update when I hear back
VijayMichalik (Mon, 17 May 2021 16:57:14 GMT):
Thank you for the suggestion @atoulme! I will spend some time trying to formulate the change I am imagining to allow a more specific discussion - but I'll be continuing to support in any case while trying to resolve this :thumbsup:
atoulme (Mon, 17 May 2021 17:01:19 GMT):
for every dev we need two of you ;) this project cannot succeed on lines of code alone
matkt (Tue, 18 May 2021 08:13:05 GMT):
To inform you, we will try to switch to a "tag-based release" mode in the next sprint. With this mode we will no longer need to freeze the master branch. We'll know more in a few weeks
NicolasMassart (Tue, 18 May 2021 15:03:08 GMT):
Thanks Antoine!
rjones (Tue, 18 May 2021 15:21:59 GMT):
ok, cool
antonydenyer (Tue, 18 May 2021 17:52:08 GMT):
Does anyone know why the DCO check checks all branches and not just the branch you're building?
shemnon (Tue, 18 May 2021 18:12:14 GMT):
bad programming? we should only have the master branch and quarterly release branches on the main hyperledger repo. If there's another branch breaking things for people I can delete it.
shemnon (Tue, 18 May 2021 18:13:52 GMT):
`rayonism` :/ @matkt @rai will this be a long lived branch?
shemnon (Tue, 18 May 2021 18:15:39 GMT):
If so HLP has a "Real Name" policy that DCO is enforcing, A force push can fix it.
shemnon (Tue, 18 May 2021 18:16:05 GMT):
(not a fan of DCO, but LF is quite form on requiring it)
antonydenyer (Tue, 18 May 2021 18:26:11 GMT):
If it submit a pr it checks all the branches in my fork, not just the pr branch.
rjones (Tue, 18 May 2021 19:36:26 GMT):
hmm. It's only supposed to check back to a common ancestor.
atoulme (Wed, 19 May 2021 07:55:46 GMT):
What is the difference between #2292 and #2296? What's going on with those draft PRs? I don't quite get how to keep up with them or if I should pay attention. The code in those PRs seems really intricate, so happy to help review.
NicolasMassart (Wed, 19 May 2021 11:13:48 GMT):
I totally agree for a PR to update maintainers entry conditions to enable non code contributors who helped the project. @SajidaZ and @VijayMichalik are clearly good candidates and we need them for this project to suceed.
VijayMichalik (Wed, 19 May 2021 14:51:34 GMT):
this is @rai and @GarySchulte
shemnon (Wed, 19 May 2021 15:13:12 GMT):
I don't think we should use draft PRs. If you need a pre-review we can always look at your cloned repo.
atoulme (Wed, 19 May 2021 18:55:22 GMT):
I mean, color me entertained. I just don't get it. I'm really just trying to see what's up.
atoulme (Wed, 19 May 2021 20:30:32 GMT):
not even sure where the contribution guidelines contain info about what constitutes participation
shemnon (Wed, 19 May 2021 20:39:10 GMT):
Hyperledger Charter section 4 subsection b defines contributor and maintainer. It doesn't perscribe github access. https://www.hyperledger.org/about/charter
shemnon (Wed, 19 May 2021 20:39:10 GMT):
Hyperledger Charter section 4 subsection b defines contributor and maintainer. https://www.hyperledger.org/about/charter
atoulme (Wed, 19 May 2021 20:39:56 GMT):
so
atoulme (Wed, 19 May 2021 20:40:01 GMT):
let's open a PR and get them voted in?
atoulme (Wed, 19 May 2021 20:40:38 GMT):
I'm happy to get that rolling.
rai (Wed, 19 May 2021 20:52:57 GMT):
@shemnon That's what we were thinking. It felt like it was better to have it there than in our of our individual forks for discovery purposes
GarySchulte (Wed, 19 May 2021 20:54:45 GMT):
it might be worth dropping the branch and renaming it to 'the merge' after we fix any DCO issues
GarySchulte (Wed, 19 May 2021 20:54:51 GMT):
since rayonism is "done"
rai (Wed, 19 May 2021 21:03:23 GMT):
That sounds good too
GarySchulte (Wed, 19 May 2021 21:04:05 GMT):
I will drop it and recreate a long lived branch.
rai (Wed, 19 May 2021 21:45:12 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
rai (Wed, 19 May 2021 22:18:04 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
GarySchulte (Thu, 20 May 2021 04:31:11 GMT):
@SajidaZ just pointed out this thread. those two PRs are addressing pending transaction ordering for 1559.
GarySchulte (Thu, 20 May 2021 04:32:28 GMT):
2292 implements an idea from one of the openethereum devs: https://gist.github.com/sunce86/7755a6360ef4885146163c218df29ccb
GarySchulte (Thu, 20 May 2021 04:33:51 GMT):
2296 is mostly about trying to clean up what has become pretty unwieldy and tightly coupled code. we actually merged 2292 in the 21.1.6 release, and I am working on adapting that implementation into the interface in 2296.
@atoulme I will definitely take you up on your offer to review :)
atoulme (Thu, 20 May 2021 04:34:20 GMT):
yeah, looks like you guys are working really hard
GarySchulte (Thu, 20 May 2021 04:36:21 GMT):
we'd like to be solid candidates for execution engines in the merge ;)
GarySchulte (Thu, 20 May 2021 04:36:21 GMT):
we'd like to be solid candidates for execution engine in the merge ;)
atoulme (Thu, 20 May 2021 07:11:29 GMT):
Please vote folks: https://github.com/hyperledger/besu/pull/2303
atoulme (Thu, 20 May 2021 07:18:52 GMT):
This one too: https://github.com/hyperledger/besu/pull/2304
NicolasMassart (Thu, 20 May 2021 08:58:11 GMT):
Should we update the rules to become a maintainer first? It would be a simple PR on MAINTAINERS.md to add non code maintainers with explicit criteria.
NicolasMassart (Thu, 20 May 2021 09:00:06 GMT):
Note: I'm for having @SajidaZ and @VijayMichalik join the band, it's only that I don't want anyone to request maintainership because they just answered a question on RC and be frustrated or annoyed because we refuse... So we need to have objectively enforceable rules.
NicolasMassart (Thu, 20 May 2021 09:03:44 GMT):
@shemnon Contributors are not a specific status in this doc, it's just someone who can create a PR or help in the wiki. Here we really need them, for technical reasons, to be maintainers to be able to handle issues properly.
atoulme (Thu, 20 May 2021 10:28:29 GMT):
Feel free
NicolasMassart (Thu, 20 May 2021 14:15:09 GMT):
I experience the same DCO issue on my new PR https://github.com/hyperledger/besu/pull/2307
NicolasMassart (Thu, 20 May 2021 14:16:09 GMT):
I created this PR https://github.com/hyperledger/besu/pull/2307 to update the maintainer rules. Welcome for comments!
NicolasMassart (Thu, 20 May 2021 14:23:52 GMT):
It seems to be caused by https://github.com/hyperledger/besu/commit/9dca04e014f42b83b0ce0c85404feff712774d76
NicolasMassart (Thu, 20 May 2021 14:24:02 GMT):
How did this commit end in master?
rjones (Thu, 20 May 2021 14:25:04 GMT):
no idea
grace.hartley (Thu, 20 May 2021 15:00:49 GMT):
User User_2 added by grace.hartley.
NicolasMassart (Thu, 20 May 2021 15:21:31 GMT):
Fixed.
atoulme (Thu, 20 May 2021 15:23:38 GMT):
My point of view on this whole thing is this: the project should greedily feel ok to accept contributors as it best serves its interests. Do not worry about the next time or the precedent - the real danger would be if there's no next time.
shemnon (Thu, 20 May 2021 15:31:22 GMT):
As a TSC member I need to advocate for precedent and process. Without it we are just a bunch of github repos.
shemnon (Thu, 20 May 2021 15:32:21 GMT):
resolving #2307 before #2303 and #2304 is the course of action that will keep other TSC members uninterested.
atoulme (Thu, 20 May 2021 19:52:29 GMT):
yep
NicolasMassart (Fri, 21 May 2021 08:54:41 GMT):
I explicitly made #2303 and #2304 dependent of #2307 for being merged so there's no doubt about the process.
But please continue to cast your vote (approve PR) so as soon as we reach the approval (either way) we can merge. The sooner the better as we really need @SajidaZ and @VijayMichalik to be fully empowered to help.
atoulme (Sat, 22 May 2021 03:11:37 GMT):
Review please https://github.com/hyperledger/besu/pull/2313
atoulme (Sat, 22 May 2021 09:21:05 GMT):
Review please https://github.com/hyperledger/besu/pull/2315
atoulme (Sat, 22 May 2021 21:22:57 GMT):
This looks like a regression with the latest release: https://github.com/hyperledger/besu/issues/2316
NicolasMassart (Mon, 24 May 2021 13:05:47 GMT):
@VijayMichalik and @SajidaZ are now in the merged maintainer list https://github.com/hyperledger/besu/blob/master/MAINTAINERS.md.
Who is admin and can add them to the Github maintainer group so they have all the access they need to manage the repos?
Thanks!
matkt (Mon, 24 May 2021 13:10:24 GMT):
I will do that
NicolasMassart (Mon, 24 May 2021 13:23:39 GMT):
@shemnon @ajsutton @mbaxter @joshuafernandes? You are the ones who are maintainers in https://github.com/orgs/hyperledger/teams/besu-maintainers/members?query=role%3Amaintainer so I guess you can do that.
Could we also have people from the Consensys Besu team as maintainers?
Team is @matkt, @GarySchulte, @rai, and @SajidaZ.
(@abdelhamidbakhta is on leave)
NicolasMassart (Mon, 24 May 2021 13:24:22 GMT):
shemnon ajsutton mbaxter joshuafernandes? You are the ones who are maintainers in https://github.com/orgs/hyperledger/teams/besu-maintainers/members?query=role:maintainer so I guess you can do that.
Could we also have people from the Consensys Besu team as maintainers?
Team is matkt, GarySchulte, rai, and @SajidaZ.
(@abdelhamidbakhta is on leave)
NicolasMassart (Mon, 24 May 2021 13:24:49 GMT):
shemnon ajsutton mbaxter joshuafernandes? You are the ones who are maintainers in https://github.com/orgs/hyperledger/teams/besu-maintainers/members?query=role:maintainer so I guess you can do that.
Could we also have people from the Consensys Besu team as maintainers?
Team is matkt, GarySchulte, rai, and @SajidaZ.
(@abdelhamidbakhta is on leave)
NicolasMassart (Mon, 24 May 2021 13:24:49 GMT):
@shemnon @ajsutton @mbaxter @joshuafernandes? You are the ones who are maintainers in https://github.com/orgs/hyperledger/teams/besu-maintainers/members?query=role:maintainer so I guess you can do that.
Could we also have people from the Consensys Besu team as maintainers?
Team is @matkt, @GarySchulte, @rai, and @SajidaZ.
(@abdelhamidbakhta is on leave)
NicolasMassart (Mon, 24 May 2021 13:24:49 GMT):
@shemnon @ajsutton @mbaxter @joshuafernandes? You are the ones who are maintainers in https://github.com/orgs/hyperledger/teams/besu-maintainers/members?query=role:maintainer so I guess you can do that.
Could we also have people from the Consensys Besu team as maintainers?
Team is @matkt, @GarySchulte, @rai, and @SajidaZ
( @abdelhamidbakhta is on leave)
rjones (Mon, 24 May 2021 13:51:01 GMT):
@NicolasMassart I've added them both
atoulme (Mon, 24 May 2021 17:38:38 GMT):
Tonight is our Besu Contributor Call APAC/AMER at 6pm PST!
atoulme (Mon, 24 May 2021 18:43:45 GMT):
Additional development on github actions gleaned from asf build team:
FYI
https://github.blog/changelog/2021-05-06-github-actions-beta-api-to-approve-actions-from-forks/
https://github.blog/changelog/2021-04-22-github-actions-maintainers-must-approve-first-time-contributor-workflow-runs/
joshuafernandes (Mon, 24 May 2021 21:17:21 GMT):
Thank you Ry :)
atoulme (Wed, 26 May 2021 16:28:41 GMT):
I have filed the mining issues that I am going to work on, or at least get a keen interest in: https://github.com/hyperledger/besu/labels/Mining
atoulme (Wed, 26 May 2021 17:56:59 GMT):
How does one go about having a new Besu release out around here? I'd like to make sure we get one out for Magneto soonish.
shemnon (Wed, 26 May 2021 18:29:16 GMT):
It looks like this old release schedule has been abandoned since I left: https://wiki.hyperledger.org/display/BESU/Future+Release+Dates
shemnon (Wed, 26 May 2021 18:31:05 GMT):
If one of the Chupacabras can give us the release checklist from GAR-e ( @rai @matkt @GarySchulte @SajidaZ ) it can be pushed manually instead of by the GAR-e tool they use. It takes two maintainers to release: one to make the PRs and one to approve and submit. CircleCI does most of the rest of it.
shemnon (Wed, 26 May 2021 18:31:22 GMT):
There was a release last week so if they are still on 2 week schedule there should be one next week.
atoulme (Wed, 26 May 2021 18:32:00 GMT):
no idea what GAR-e tool is?
shemnon (Wed, 26 May 2021 18:32:21 GMT):
ConsenSys has some canaries they would want to update and then possibly some twitter pushes.
atoulme (Wed, 26 May 2021 18:32:33 GMT):
this might be a good enhancement. I'd like anyone on the project to be able to issue release candidates.
atoulme (Wed, 26 May 2021 18:32:50 GMT):
Otherwise, in effect we depend on one of the participants.
GarySchulte (Wed, 26 May 2021 18:36:52 GMT):
I watched Rai do the last deployment, and my takeaway was that the deployment process revolved around
* tagging/branching PRs
* waiting for builds
* pushing a release to github and to homebrew
* updating canaries
I think I can get the checklist from GAR-e, but I am not sure how applicable the checklist is going to be for Magneto.
Are there separate binaries/releases we use for etc? TBH, when @atoulme mentioned ETC on the contributor call that was the first I had heard that besu was seeing use on classic
atoulme (Wed, 26 May 2021 18:37:53 GMT):
no special binaries
atoulme (Wed, 26 May 2021 18:38:00 GMT):
just different flags
GarySchulte (Wed, 26 May 2021 18:38:23 GMT):
cool, so would the release candidate process we are about to begin suffice for magneto?
GarySchulte (Wed, 26 May 2021 18:38:35 GMT):
let me dig up the exact dates
atoulme (Wed, 26 May 2021 18:38:38 GMT):
yeah
shemnon (Wed, 26 May 2021 22:33:56 GMT):
RC would be on-brand for testnet releases.
atoulme (Wed, 26 May 2021 22:41:47 GMT):
Maybe a rephrase then. How does one go about making a release candidate?
GarySchulte (Thu, 27 May 2021 19:47:18 GMT):
sorry for the slow reply here, the current release candidate schedule is:
* first testnet release of Besu is 21.1.7, May 28
* 21.7 RC 1 includes mainnet support, June 8
* 21.7 RC 2 June 22
* 21.7.0 mainnet ready release, July 6
GarySchulte (Thu, 27 May 2021 19:49:16 GMT):
as far as what is involved in a release candidate, the primary difference to a normal release is the creation of a release candidate branch that the release is cut from
GarySchulte (Thu, 27 May 2021 19:50:27 GMT):
seems we also do customer communications for major releases that point release and RC releases do not have
atoulme (Thu, 27 May 2021 20:44:55 GMT):
ok I'll catch that RC
VijayMichalik (Mon, 31 May 2021 09:56:45 GMT):
Hi all, just a PSA here that ConsenSys has merged all Orion functionality into Tessera, which works as a drop-in replacement, so we're entering a deprecation cycle for Orion which will last 6 months to November 30th, 2021.
Further details are all in the changelogs and docs as well as this blog post https://consensys.net/blog/quorum/tessera-the-privacy-manager-of-choice-for-consensys-quorum-networks/.
VijayMichalik (Mon, 31 May 2021 09:56:45 GMT):
Hi all, just a PSA here that ConsenSys has merged all Orion functionality into Tessera, which works as a drop-in replacement, so we're entering a deprecation cycle for Orion which will last 6 months to November 30th, 2021.
Further details are all in the changelogs and docs as well as this blog post https://consensys.net/blog/quorum/tessera-the-privacy-manager-of-choice-for-consensys-quorum-networks/.
VijayMichalik (Mon, 31 May 2021 09:56:45 GMT):
Hi all, just a PSA here that ConsenSys has merged all Orion functionality into Tessera, which works as a drop-in replacement, so we're entering a deprecation cycle for Orion which will last 6 months to November 30th, 2021.
Further details are all in the changelogs and docs as well as this blog post https://consensys.net/blog/quorum/tessera-the-privacy-manager-of-choice-for-consensys-quorum-networks/.
NicolasMassart (Mon, 31 May 2021 12:53:45 GMT):
rai (Mon, 31 May 2021 13:41:59 GMT):
Sorry for the inconsistency! We didn't sufficiently decentralize the release before abdel went on parental leave. With gary recently becoming a maintainer that meant that if either Karim and I had to take leave then we didn't have the reviews to release.
Now the inconsistency is more to accommodate last minute changes to London and lining things up with the testnets :sweat_smile:
rai (Mon, 31 May 2021 13:42:38 GMT):
We're gonna be doing a housekeeping sprint soon that will include a bunch of release-related improvements
shemnon (Mon, 31 May 2021 16:45:13 GMT):
Any maintainer could have approved the release PRs. The next time this occurs ping Antoine or myself and I think we can make it happen. Or any of the Aussie maintainers.
rai (Mon, 31 May 2021 16:48:31 GMT):
For sure, it wasn't clear y'all would be available/willing at the time but we'll keep that in mind in the future.
atoulme (Mon, 31 May 2021 23:04:54 GMT):
I need a RC for etc asap soon folks. How do we make it happen?
atoulme (Mon, 31 May 2021 23:06:11 GMT):
* first testnet release of Besu is 21.1.7, May 28
Is this out yet?
atoulme (Mon, 31 May 2021 23:07:25 GMT):
Nothing here? https://github.com/hyperledger/besu/tags
atoulme (Mon, 31 May 2021 23:08:18 GMT):
I am not going to ping anyone as today is a holiday in the US. I’ll ping tomorrow.
matkt (Tue, 01 Jun 2021 15:27:30 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
matkt (Tue, 01 Jun 2021 16:22:06 GMT):
@here - It's now safe to merge into master for hyperledger/besu again
antonydenyer (Thu, 03 Jun 2021 07:25:50 GMT):
Could I get a second reviewer on https://github.com/hyperledger/besu/pull/2283 please?
atoulme (Sat, 05 Jun 2021 00:33:21 GMT):
anyone feeling like a review for a simple flaky test: https://github.com/hyperledger/besu/pull/2375
Not sure if this is going to fix the flakiness, but it won't hurt to add timeouts.
RoboCopsGoneMad (Mon, 07 Jun 2021 13:34:08 GMT):
Has joined the channel.
JfloConsensys (Mon, 07 Jun 2021 13:35:20 GMT):
Has joined the channel.
mdwright88 (Tue, 08 Jun 2021 03:39:35 GMT):
Hey all!
We have our Besu contributor call tomorrow for AMER/ EMEA.
What topics should we cover?
usmansaleem (Tue, 15 Jun 2021 02:06:41 GMT):
PR review please. https://github.com/hyperledger/besu/pull/2433
usmansaleem (Tue, 15 Jun 2021 04:41:55 GMT):
Thanks @shemnon, moving it to hyperledger specific repo makes sense. Also looks like I've to fix plugin related test which are weirdly failing because I've removed errorprone submodule.
rai (Tue, 15 Jun 2021 16:43:06 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
rai (Tue, 15 Jun 2021 17:50:09 GMT):
@here - It's now safe to merge into 21.7.0 for hyperledger/besu again.
alexandratran (Tue, 15 Jun 2021 23:57:26 GMT):
Can I get a review on this PR documenting transactions types and new fields for EIPs 2930 and 1559? https://github.com/hyperledger/besu-docs/pull/738
gortaire (Thu, 17 Jun 2021 04:38:18 GMT):
Hi all, I'd like to know what are the requirements to join and collaborate on the project. I'd like to get involved and get deeper learning. Thanks in advance!
VijayMichalik (Thu, 17 Jun 2021 09:30:32 GMT):
Hey @gortaire, pleasure to have you on board! Could you let us know a little more about your skills, background and interest areas? That might help people make more concrete suggestions
VijayMichalik (Thu, 17 Jun 2021 15:52:58 GMT):
Hi all, I've updated the Besu release calendar here (and will endeavour to keep it updated better in the future) https://wiki.hyperledger.org/display/BESU/Future+Release+Dates
shemnon (Fri, 18 Jun 2021 15:42:41 GMT):
Have we done the master-> main migrations yet on the Besu repositories?
rjones (Fri, 18 Jun 2021 15:46:25 GMT):
@shemnon no: ```{
"name": "besu",
"full_name": "hyperledger/besu",
"git_url": "git://github.com/hyperledger/besu.git",
"default_branch": "master",
"clone_url": "https://github.com/hyperledger/besu.git",
"ssh_url": "git@github.com:hyperledger/besu.git",
"has_issues": true,
"archived": false
}
{
"name": "besu-docs",
"full_name": "hyperledger/besu-docs",
"git_url": "git://github.com/hyperledger/besu-docs.git",
"default_branch": "master",
"clone_url": "https://github.com/hyperledger/besu-docs.git",
"ssh_url": "git@github.com:hyperledger/besu-docs.git",
"has_issues": true,
"archived": false
}
{
"name": "besu-native",
"full_name": "hyperledger/besu-native",
"git_url": "git://github.com/hyperledger/besu-native.git",
"default_branch": "master",
"clone_url": "https://github.com/hyperledger/besu-native.git",
"ssh_url": "git@github.com:hyperledger/besu-native.git",
"has_issues": true,
"archived": false
}
```
gortaire (Sat, 19 Jun 2021 05:18:22 GMT):
Hi Vijay, I'm a software engineer with more than 15 years of experience working on different projects for the Chilean financial market, I have advanced skills on JAVA mainly, I've worked with the most common enterprise application stacks from Oracle, IBM, and JBoss, as also with the main frameworks for implement integration solutions for mission-critical software. Last years I have been playing the Software Architect role and now I'm involved in some PoCs related to dlt technologies, so I'm very interested on get more involved to get deeper learning and understanding of the technology. Leave you my LinkedIn profile to look for further detail on my experience if you want.
Greetings!
NicolasMassart (Sat, 19 Jun 2021 16:54:42 GMT):
It's on my todo list for the doc.
shemnon (Sat, 19 Jun 2021 21:45:57 GMT):
Also, who has the task to write the quarterly report? (like this - https://wiki.hyperledger.org/display/TSC/2021+Q1+Hyperledger+Besu). I'm too far out of the loop to write it this quarter. It's due this week.
macfarla (Mon, 21 Jun 2021 00:16:36 GMT):
@grace.hartley is this on your list?
grace.hartley (Mon, 21 Jun 2021 12:52:06 GMT):
Yep, this is on me. Will share with TSC in next day or so.
mdwright88 (Mon, 21 Jun 2021 18:54:54 GMT):
Here's the updated Wiki for today's contributor call:
https://wiki.hyperledger.org/display/BESU/2021-06-21+Besu+Contributor+Call
*We're going to cover:*
- QBFT Update
- PKI Update
- London Hard Fork + Besu Release Update
Does anyone have anything to add?
Thinking @shemnon we can chat through Q2 report as well!
VijayMichalik (Mon, 21 Jun 2021 21:12:09 GMT):
Awesome, thanks for the offer of contribution! It would be great if you could join one of our contributor calls to say hi - this one is today but there will be others. https://wiki.hyperledger.org/display/BESU/2021-06-21+Besu+Contributor+Call
VijayMichalik (Mon, 21 Jun 2021 21:14:10 GMT):
https://wiki.hyperledger.org/display/BESU/2021-06-21+Besu+Contributor+Call
grace.hartley (Tue, 22 Jun 2021 00:20:39 GMT):
Hey @mdwright88 I'm not able to join the contributor call today, but here is the draft Besu quarterly report. Please feel free to share on the contributor call for feedback. https://wiki.hyperledger.org/pages/resumedraft.action?draftId=54657092&draftShareId=48951b2a-1cb9-4512-8114-2a8e5dce2210&src=shareui&src.shareui.timestamp=1624321167729
grace.hartley (Tue, 22 Jun 2021 00:21:27 GMT):
I'll plan on publishing to the TSC tomorrow. @shemnon is planning on adding some Badging information
VijayMichalik (Tue, 22 Jun 2021 08:15:35 GMT):
Also we have some good first issues labelled here if you want to jump right in! https://github.com/hyperledger/besu/labels/good%20first%20issue
iamsdas (Tue, 22 Jun 2021 17:41:59 GMT):
Has joined the channel.
gortaire (Thu, 24 Jun 2021 02:07:49 GMT):
Thank you Vijay, sadly i’m late to join, but hope to have a chance later, however i’ll take a look to the available issues to find something suitable to analyze.
VijayMichalik (Tue, 29 Jun 2021 10:30:22 GMT):
Super! Here's a list of good first issues and I'm always happy to discuss options for resolving them or help you find another maintainer to do so. https://github.com/hyperledger/besu/issues?q=is%3Aissue+is%3Aopen+sort%3Acreated-asc+label%3A%22good+first+issue%22
VijayMichalik (Tue, 29 Jun 2021 10:33:29 GMT):
Just linking this here for anyone lurking who is interested to see if they can get stuck in contributing:
https://github.com/hyperledger/besu/issues?q=is%3Aissue+is%3Aopen+sort%3Acreated-asc+label%3A%22good+first+issue%22 is our list of "Good First Issues"
I'm happy to help offer guidance if you want to get stuck in or help you find the right maintainer to support on more technical questions if it isn't me!
VijayMichalik (Tue, 29 Jun 2021 10:33:29 GMT):
Just linking this here for anyone lurking who is interested to see if they can get stuck in contributing:
https://github.com/hyperledger/besu/issues?q=is%3Aissue+is%3Aopen+sort%3Acreated-asc+label%3A%22good+first+issue%22 is our list of "Good First Issues"
I'm happy to help offer guidance if you want to get involved or help you find the right maintainer to support on more technical questions if it isn't me!
VijayMichalik (Mon, 05 Jul 2021 15:10:26 GMT):
if someone can look at/respond to this issue that'd be awesome - deep in the RLP internals https://github.com/hyperledger/besu/issues/2488
mdwright88 (Tue, 06 Jul 2021 14:17:09 GMT):
We have our Besu Contributor call at the top of the hour. Does anybody have any topics to add to the agenda?
https://wiki.hyperledger.org/display/BESU/2021-07-06+Besu+Contributor+Call
VijayMichalik (Wed, 07 Jul 2021 14:29:06 GMT):
I've just updated the roadmap, might still need to make a few tweaks in the next couple of days https://wiki.hyperledger.org/display/BESU/Roadmap
mdwright88 (Wed, 07 Jul 2021 18:12:51 GMT):
@gortaire let us know if we can do anything to make it easier for you. Would love to have your contributions! We also have contributor calls Monthly you should should! Where are you based?
shemnon (Thu, 08 Jul 2021 15:49:59 GMT):
Once the 21.7.0 release crazyness can we get a release checklist or an open version of GAR-e so non-ConsenSys maintainers can cut releases?
VijayMichalik (Thu, 08 Jul 2021 16:27:45 GMT):
Totally agree, this is something we need to do. I have digging into it in my backlog but not sure when I'll get to it - will go hunt around for some other volunteers though!
rai (Thu, 08 Jul 2021 17:00:50 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to release-21.7.0 until further notice
rai (Thu, 08 Jul 2021 17:01:56 GMT):
@VijayMichalik, let's coordinate on doing release changes and issue cleanup soon :)
shemnon (Thu, 08 Jul 2021 19:34:33 GMT):
Did the process finish?
rai (Thu, 08 Jul 2021 19:35:32 GMT):
Still updating nodes and launching syncs but yes. I realized there's no step clearly announcing this because the branch gets deleted.
shemnon (Thu, 08 Jul 2021 19:58:41 GMT):
For a non-main branch I wonder if we even need to announce the build.
shemnon (Thu, 08 Jul 2021 19:59:52 GMT):
Or change the chat text to "we're beginning the release of
ritvij14 (Sun, 11 Jul 2021 03:38:12 GMT):
Has joined the channel.
atoulme (Mon, 12 Jul 2021 13:21:35 GMT):
https://github.com/hyperledger/besu/pull/2376 is up for review
Helen_Garneau (Tue, 13 Jul 2021 13:27:40 GMT):
Hello Besu Contributors! Reminder to please join the DevRel Marketing Committee call at 9am PT tomorrow- 7/14. Take a look at the agenda and add items if you'd like here: https://wiki.hyperledger.org/x/sANCAw
atoulme (Wed, 14 Jul 2021 12:13:41 GMT):
This is very important +1
shemnon (Wed, 14 Jul 2021 17:11:18 GMT):
@here - We're beginning the release process for hyperledger/besu, please don't merge anything to master until further notice
shemnon (Wed, 14 Jul 2021 18:16:53 GMT):
@here - It's now safe to merge into master for hyperledger/besu again.
atoulme (Sun, 25 Jul 2021 18:09:05 GMT):
hey, this PR is up for review https://github.com/hyperledger/besu/pull/2576/files but please also let whoever opened the issue review first if you can wait. It is not a trivial change but very specific to mining.
shemnon (Mon, 26 Jul 2021 20:36:18 GMT):
i don't think homebrew got the latest 21.7.1 release
shemnon (Tue, 27 Jul 2021 02:21:59 GMT):
Review Request: 21.7.1 on homebew
shemnon (Tue, 27 Jul 2021 02:22:18 GMT):
Review Request: 21.7.1 on homebrew:
https://github.com/hyperledger/homebrew-besu/pull/53
atoulme (Tue, 27 Jul 2021 03:35:45 GMT):
A fun review for another mining related change: https://github.com/hyperledger/besu/pull/2581
atoulme (Tue, 27 Jul 2021 17:53:53 GMT):
2583 and 2584 are open for review too
atoulme (Wed, 28 Jul 2021 20:50:02 GMT):
Nobody up for reviewing https://github.com/hyperledger/besu/pull/2581 ? Danno reviewed my last 3 PRs, hat tip to him, anyone else feeling like reviewing my PRs?
rjones (Wed, 28 Jul 2021 21:56:59 GMT):
I could approve it, but it would make folks angerous :)
atoulme (Wed, 28 Jul 2021 23:19:15 GMT):
@rjones think so? wanna try?
rjones (Thu, 29 Jul 2021 00:42:13 GMT):
nah, last time I did that, people were unhappy :)
atoulme (Thu, 29 Jul 2021 00:42:30 GMT):
but that was _last time_
atoulme (Thu, 29 Jul 2021 00:42:50 GMT):
maybe on average folks are ok, and so to balance it out this time they'd be happy about it
atoulme (Thu, 29 Jul 2021 00:44:16 GMT):
worst case scenario, they might have a bigger incentive in actively reviewing and participating, lest they see PRs merged without them weighing in :)
atoulme (Thu, 29 Jul 2021 00:44:51 GMT):
Either way, if you're a committer, then you can review and approve. I'll do the responsible thing and reach out to the crowd here anyway :)
shemnon (Thu, 29 Jul 2021 00:55:52 GMT):
Ry is not on the maintainers list.
atoulme (Thu, 29 Jul 2021 00:56:29 GMT):
well in that case, ok, you definitely should not approve PRs.
shemnon (Thu, 29 Jul 2021 02:49:42 GMT):
Another step to Gradle 7: updating spotless and google-java-format -- https://github.com/hyperledger/besu/pull/2593
rjones (Thu, 29 Jul 2021 13:51:58 GMT):
as I said :)
atoulme (Thu, 29 Jul 2021 16:39:08 GMT):
Huge thanks to Danno for reviewing my PR https://github.com/hyperledger/besu/pull/2581
shemnon (Thu, 29 Jul 2021 19:57:56 GMT):
Heads up ConsenSys Devs: I'm going to merge the Gradle 7.1.1 upgrade and dependncy upgrade PR tomorrow unless I hear objections. https://github.com/hyperledger/besu/pull/2596
shemnon (Thu, 29 Jul 2021 19:58:08 GMT):
This gets Besu in a place it can run Java 16.
shemnon (Thu, 29 Jul 2021 19:58:08 GMT):
This gets Besu in a place it can run/build Java 16.
shemnon (Thu, 29 Jul 2021 23:28:10 GMT):
Where does libnss3.dylib need to live on my mac for unit tests to pick it up? (spoiler alert: M1 Mac running a roseta vm)
shemnon (Fri, 30 Jul 2021 14:47:57 GMT):
Also, when will 21.7.2 be cut? If it's next week I'll hold off on the gradle upgrade. If it's two weeks I'll commit.
atoulme (Fri, 30 Jul 2021 15:56:27 GMT):
Folks - please don't link to private resources like the consensys slack on a public issue tracker. It just really isn't going to help this project build a community. https://github.com/hyperledger/besu/issues/2601
NicolasMassart (Mon, 02 Aug 2021 15:41:40 GMT):
Hi team!
I updated the besu-doc repos, now default branch is `main`. I updated also ReadTheDocs configuration and CI.
Update the contributing doc is in progress.
If you notice any issue due to this change, please tell me.
Thanks.
NicolasMassart (Mon, 02 Aug 2021 15:43:30 GMT):
NOTE: it will require that you rename your forks default branch to `main` too. You should have an alert that we renamed when you go on your fork in the Github website. You may experience error if you try to fetch or pull without renaming.
NicolasMassart (Mon, 02 Aug 2021 15:59:15 GMT):
I updated the wiki, mostly in the doc contribution sections, so please feel free to update to "main" if you see any remaining "master".
Thanks.
macfarla (Tue, 03 Aug 2021 00:40:02 GMT):
@GarySchulte do you know the answer to this one?
madelinemurray (Tue, 03 Aug 2021 04:54:35 GMT):
hi all - Besu Contributor is on tomorrow (or today depending on your location :-)). EMEA/AMER friendly time:
8 AM Tuesday San Francisco
11 AM Tuesday New York
3 PM Tuesday UTC
5 PM Tuesday Paris/Berlin
1 AM Wednesday Brisbane
https://wiki.hyperledger.org/display/BESU/2021-08-03+Besu+Contributor+Call
Please add anything else you'd like to discuss.
NicolasMassart (Tue, 03 Aug 2021 18:30:08 GMT):
Hi, I forgot to push this PR https://github.com/hyperledger/besu-docs/pull/762 for the renaming changes to take effect in CI.
Can anyone review and approve please?
Thanks!
shemnon (Tue, 03 Aug 2021 18:50:21 GMT):
Here's three more on that theme... Once they are in I'll flip configs:
* https://github.com/hyperledger/besu/pull/2608
* https://github.com/hyperledger/besu-native/pull/42
* https://github.com/hyperledger/homebrew-besu/pull/54
NicolasMassart (Tue, 03 Aug 2021 18:56:20 GMT):
Thanks for your review @shemnon !
shemnon (Tue, 03 Aug 2021 19:34:34 GMT):
The `besu`, `besu-docs`, and `besu-native` repos have all flipped from using the `master` branch as the default branch to using the `main` branch as the default branch. To keep up with these changes
(a) rename the default branch in your forked repositories to `main`. Github will provide reminders when you navigate to your repos.
(b) update your working copies to use `main`. This script will work for remotes named `origin`
```
git branch -m master main
git fetch origin
git branch -u origin/main main
git remote set-head origin -a
```
shemnon (Tue, 03 Aug 2021 19:37:48 GMT):
If you have a homebrew installation simply run `brew tap --repair`
rjones (Tue, 03 Aug 2021 19:42:46 GMT):
thanks!
GarySchulte (Tue, 03 Aug 2021 22:37:57 GMT):
@shemnon re: libnss, For rosetta it is easiest to use the x86 homebrew. (it installs in /usr/local/homebrew, whereas the m1 homebrew is /opt/homebrew)
once you have x86 homebrew installed, you can:
```arch --X86_64 /usr/local/homebrew/bin/brew install nss```
and then an x86 jvm should find the dep
atoulme (Tue, 03 Aug 2021 22:39:20 GMT):
https://github.com/hyperledger/besu/issues/2606 seems like a critical issue
GarySchulte (Tue, 03 Aug 2021 22:40:56 GMT):
barring that, you can always add this to build.gradle:
```test {
systemProperty "java.library.path", "/opt/homebrew/lib"
}```
GarySchulte (Wed, 04 Aug 2021 01:52:04 GMT):
here - We're beginning the release process for hyperledger/besu
macfarla (Wed, 04 Aug 2021 02:08:27 GMT):
fixed!
atoulme (Wed, 04 Aug 2021 05:44:42 GMT):
good stuff.
GarySchulte (Wed, 04 Aug 2021 07:05:57 GMT):
hotfix release is complete for hyperledger/besu
jlochman (Wed, 04 Aug 2021 09:25:45 GMT):
Has joined the channel.
IvanHomoliak (Wed, 04 Aug 2021 12:21:27 GMT):
Has joined the channel.
lucassaldanha (Thu, 05 Aug 2021 01:59:47 GMT):
Here is the incident report re y'day hotfix: https://wiki.hyperledger.org/display/BESU/04-08-2021+Value+transfer+public+transactions+being+rejected+by+Besu
VijayMichalik (Tue, 10 Aug 2021 09:12:40 GMT):
I proposed to add @antonydenyer as a maintainer, if you get a chance to review! https://github.com/hyperledger/besu/pull/2622
kkaur01 (Wed, 11 Aug 2021 06:16:14 GMT):
Has joined the channel.
atoulme (Wed, 11 Aug 2021 19:06:28 GMT):
hey folks, please pay attention and be careful to engage in discussions in public: https://github.com/hyperledger/besu/pull/2626 this PR was closed after a "discussion". Please have discussions in the open.
RoboCopsGoneMad (Thu, 12 Aug 2021 18:28:25 GMT):
heya folks, I'm about to merge https://github.com/hyperledger/besu/pull/2505 and i am looking for suggestions for testing. It's already got unit tests, and I'm planning on deploying a miner with it and watching it. Anyone got any additional testing ideas?
rjones (Thu, 12 Aug 2021 18:32:14 GMT):
Hi, maintainers. GitHub.dev is now public: https://chat.hyperledger.org/channel/fabric-maintainers?msg=bqkZHi6KprzmajRvZ
atoulme (Fri, 13 Aug 2021 06:15:24 GMT):
Test coverage looks ok. Maybe could just test the compare method itself since it's doing a bunch. You won't need a miner with --dev, it will be able to mine with CPU
shemnon (Fri, 13 Aug 2021 21:14:09 GMT):
Two things for next weeks Office Hours:
* Discuss my proposal to make EVM support a library: https://wiki.hyperledger.org/display/BESU/Refactor+EVM+into+a+stand+alone+library
* What's our Java 17 strategy? (Java 17 is the next long term stable version, when do we switch over to that as a base?)
GarySchulte (Fri, 13 Aug 2021 22:47:16 GMT):
PR to plumb GasLimitCalculator into ProtocolSchedule, so we can remove 1559-specific logic from GasLimitRangeAndDeltaValidationRule and AbstractBlockCreator. This has been a bit of a chore to re-plumb, but I am ok with the result. I am sure this breaks myriad tests and mocks, so I'd appreciate any feedback before I put a bunch of work in on that:
https://github.com/hyperledger/besu/pull/2632
TIA :pray:
atoulme (Sat, 14 Aug 2021 18:07:20 GMT):
hey, I'm talking with the folks from EIP-3074 and I put together quickly a draft PR for their review. https://github.com/hyperledger/besu/pull/2636/files Feel free to take a look but it's nowhere near ready for merge, mostly just me trying to have fun on a weekend.
VijayMichalik (Sun, 15 Aug 2021 19:52:02 GMT):
I created this epic to hold tasks required for updating to JDK 16, cc @shemnon @atoulme https://github.com/hyperledger/besu/issues/2638
VijayMichalik (Sun, 15 Aug 2021 19:52:02 GMT):
I created this epic to hold tasks required for updating to JDK 17, cc @shemnon @atoulme https://github.com/hyperledger/besu/issues/2638
VijayMichalik (Sun, 15 Aug 2021 19:52:09 GMT):
Nice!
atoulme (Sun, 15 Aug 2021 21:44:33 GMT):
FWIW latest geth master now removed for support for eth/62 to eth/65, it's now eth/66 or bust. There is no flag to bring back eth/65.
mdwright88 (Mon, 16 Aug 2021 14:02:29 GMT):
@shemnon @atoulme @rai + all —
Any updates we would like to add to the contributor call agenda today at 6pm PST?
shemnon (Mon, 16 Aug 2021 17:27:00 GMT):
Just the EVM componentization and Java version upgrade.
VijayMichalik (Mon, 16 Aug 2021 19:40:35 GMT):
I would love to get a summary after the chat if anyone's able to share here - it'll be at midnight BST so I'm not going to be awake
VijayMichalik (Mon, 16 Aug 2021 19:41:56 GMT):
if modularization helps us get towards better performance or would bring in more users and contributors I could try and rustle up some more support for it
VijayMichalik (Mon, 16 Aug 2021 19:47:29 GMT):
the desirability is not totally clear to me beyond the use cases below (which seem fairly speculative)
shemnon (Mon, 16 Aug 2021 19:51:52 GMT):
To be clear, I would be doing the work, not ConsenSys
shemnon (Mon, 16 Aug 2021 19:52:27 GMT):
And my current employer would be a user of the library.
VijayMichalik (Mon, 16 Aug 2021 20:17:20 GMT):
yes, we are on the same page - my questions still stand.
knowing that you have a use case for it within your employer is a data point!
as you mention, the EVM work is one part of modularizing Besu, I don't expect you are planning to do the whole lot solo :wink:
VijayMichalik (Mon, 16 Aug 2021 20:21:24 GMT):
I've got a draft in google sheets of some feature proposals to break out into the Besu Wiki, if anyone wants an early peek. open to comments and questions! https://docs.google.com/spreadsheets/d/16z9xKoHKR12Qom0A3B7TVd2Y0n7h0_xQ8dOUeDvCyio/edit?usp=sharing
atoulme (Tue, 17 Aug 2021 00:51:37 GMT):
going to attend the call in 10
atoulme (Tue, 17 Aug 2021 04:51:04 GMT):
https://docs.gradle.org/current/userguide/compatibility.html
> A Java version between 8 and 16 is required to execute Gradle. Java 17 and later versions are not yet supported.
VijayMichalik (Tue, 17 Aug 2021 09:20:01 GMT):
ah, that makes sense! I'll update the docs ticket
VijayMichalik (Tue, 17 Aug 2021 10:11:13 GMT):
congrats to @antonydenyer, our newest maintainer!
VijayMichalik (Tue, 17 Aug 2021 10:12:39 GMT):
if anyone has the rights to add him to the maintainers github group, please can you do so? thanks!
VijayMichalik (Tue, 17 Aug 2021 10:59:33 GMT):
how is the governance of the besu maintainers github user list supposed to work? I'm seeing this sub-list which contains a mix of people who are not all super engaged here right now ->
https://github.com/orgs/hyperledger/teams/besu-maintainers/members?query=role%3Amaintainer
shemnon (Tue, 17 Aug 2021 13:35:32 GMT):
First we need to have @ryjones send `antonydenyer` an invite to the Hyperledger organization, once Antony accepts the invite he can be aded to the group.
shemnon (Tue, 17 Aug 2021 13:40:22 GMT):
Also a heads up, about the first week of September I will be proposing another round of maintainer inactivity cleanup. I'll be using the Linux Foundation Insights too to guide this, so if everyone could go to https://myprofile.lfx.linuxfoundation.org/edit/profile and update their current organization that would make the stats cleaner.
rjones (Tue, 17 Aug 2021 14:16:25 GMT):
@shemnon invite extended
shemnon (Tue, 17 Aug 2021 15:01:55 GMT):
invite to https://github.com/orgs/hyperledger/teams/besu-maintainers/members also extended
atoulme (Tue, 17 Aug 2021 15:25:10 GMT):
also means right now I can't update Tuweni on 17, and neither can Besu probably
atoulme (Tue, 17 Aug 2021 15:25:25 GMT):
maybe there's a beta version of gradle to try, didn't have time to dig in
shemnon (Tue, 17 Aug 2021 23:22:57 GMT):
@lucassaldanha @macfarla Here's my proposed solution to pull the privacy data out of the EVM code: add a notion of "context variables" that can be set/removed from the message frame and used by precompiled contracts. The mainnet and privacy transaction processor push the privacy metadata handler and data into the context and the precompiles pull the data out.
https://github.com/shemnon/besu/commit/5a649cff8a5c9ce7946770b796bd7d1daf1cb698
lucassaldanha (Wed, 18 Aug 2021 07:04:58 GMT):
I'll take a look!
shemnon (Wed, 18 Aug 2021 14:50:59 GMT):
I'm contemplating ripping out the EIP-1702 plumbing (account versioning) that was put in place for Istanbul (it was in before it was out). I don't think we will be going do that for mainnet and there were no differences for enterprise chains. Most importantly there was no way to trigger it's use. Thoughts?
shemnon (Wed, 18 Aug 2021 16:35:09 GMT):
Here's what it would look like. https://github.com/hyperledger/besu/pull/2657 @atoulme @macfarla @lucassaldanha
shemnon (Wed, 18 Aug 2021 16:44:07 GMT):
I also cataloged the needed changes to modularize the EVM. Nothing too exotic yet. https://wiki.hyperledger.org/display/BESU/Refactor+EVM+into+a+stand+alone+library
GarySchulte (Wed, 18 Aug 2021 23:19:36 GMT):
removing EIP1559 object and folding into BaseFeeMarket and ProtocolSchedule:
https://github.com/hyperledger/besu/pull/2658
shemnon (Wed, 18 Aug 2021 23:42:08 GMT):
review request: Split out transaction gas calculations from EVM execution related gas calculations: https://github.com/hyperledger/besu/pull/2659
atoulme (Thu, 19 Aug 2021 00:53:04 GMT):
@shemnon rationale to move to Tuweni Wei? Do you want to move to Tuweni Gas too? Just trying to understand how that works.
atoulme (Thu, 19 Aug 2021 00:53:28 GMT):
Very likely the Gas class of Tuweni is too crass for the use case
shemnon (Thu, 19 Aug 2021 02:41:32 GMT):
Re: Wei... trying not to duplicate too many classes. I may just factor it out completely, not sure. It's only _needed_ for the call operations, which has a cascading impact in the world state updater. May just use a UInt256.
shemnon (Thu, 19 Aug 2021 02:42:20 GMT):
But it does have an embedded "unit" which makes type checking easier. I may just copy it into the EVM as it's fairly lightweight, the type is the main value add.
atoulme (Thu, 19 Aug 2021 04:26:21 GMT):
ok no strong opinions there
atoulme (Thu, 19 Aug 2021 04:26:52 GMT):
FWIW I managed to sync up besu to puxi to the latest block, so EIP-3074 works on my branch right now.
macfarla (Thu, 19 Aug 2021 06:03:01 GMT):
LGTM
GarySchulte (Thu, 19 Aug 2021 17:13:37 GMT):
@atoulme - I clarified with Rai about eth/64 support. He did not disable as part of the clean-up:
https://github.com/hyperledger/besu/pull/2278
IDK why I had that impression
atoulme (Thu, 19 Aug 2021 17:32:15 GMT):
no worries
atoulme (Thu, 19 Aug 2021 17:34:23 GMT):
and that said - it would be totally ok to have a flag to run besu with eth/66 only, come to think of it
atoulme (Thu, 19 Aug 2021 17:34:29 GMT):
It would help with testing.
shemnon (Thu, 19 Aug 2021 17:42:50 GMT):
Should be a simple operation, once the CLI flag is plumed down to the right place just only add ETH/66 to the capabilities. Plumbing the flag down is the real work.
shemnon (Thu, 19 Aug 2021 17:45:29 GMT):
I got some comments about the Docs to my `TransactionGasCalculator` PR, but any about the content? If not can I get an approval? https://github.com/hyperledger/besu/pull/2659
antonydenyer (Thu, 19 Aug 2021 19:08:27 GMT):
I need to get hold besu config from within a plugin. Thinking of wiring up a `ConfigQueryService`? The other option is to get rid of the validation in `PicoCLIOptionsImpl` so that in the plugin you could specify args that you are interested in. Thoughts?
shemnon (Thu, 19 Aug 2021 19:10:39 GMT):
What flags in particular are you looking for? Making a generic flag searching service concerns me a bit.
antonydenyer (Thu, 19 Aug 2021 19:11:45 GMT):
`-rpc-http-port` etc
antonydenyer (Thu, 19 Aug 2021 19:11:45 GMT):
`--rpc-http-port` etc
antonydenyer (Thu, 19 Aug 2021 19:12:33 GMT):
I want to call back into besu using rpc
antonydenyer (Thu, 19 Aug 2021 19:13:03 GMT):
But also want to make sure it's enabled on startup
shemnon (Thu, 19 Aug 2021 19:15:56 GMT):
Which call? Two things (A) I would rather create a RPCSercvice that provides that config, and would be where adding RPCs would happen. There could also be hooks to determine if it's currently up
shemnon (Thu, 19 Aug 2021 19:16:19 GMT):
and (b) Wouldn't it make more sense to do the call directly? Even if we have to expose a new Service interface?
antonydenyer (Thu, 19 Aug 2021 19:21:01 GMT):
Hang on, I think we're getting wires crossed. In the plugin I want to use web3j so I can leverage all the existing tooling, but need to know if json rpc is running, if it's https. on what port and if the api is enabled that I want to use.
antonydenyer (Thu, 19 Aug 2021 19:23:10 GMT):
If you mean `RPCConfigService` that just returns the config then I'm up for that :smile: but figured whilst I was there I'd expose all the args.
antonydenyer (Thu, 19 Aug 2021 19:23:10 GMT):
If you mean `RPCService` is something that just returns the config then I'm up for that :smile: but figured whilst I was there I'd expose all the args.
shemnon (Thu, 19 Aug 2021 19:44:09 GMT):
I don't think having a plugin call it's own process via the network port is a good design pattern. If web3j tooling is key it sounds like a microservice design pattern.
shemnon (Thu, 19 Aug 2021 19:44:49 GMT):
But yes, I would have RPCService report the RPC configurations, but not edit them.
antonydenyer (Thu, 19 Aug 2021 19:45:35 GMT):
100% not edit them - was going to be read only
antonydenyer (Thu, 19 Aug 2021 19:48:07 GMT):
the alternative to using jsonrpc is to basically re-implement all the jsonrpc endpoints with methods calls and expose them to the plugin via a service. Whilst this would get you half way there you'd still need to do all the other business with serialization etc I could be wrong, but I think calling back in over jsonrpc is the best option.
shemnon (Thu, 19 Aug 2021 20:00:52 GMT):
Something that complex sounds like a microservice. If you need something from the plugin APIs not in the RPC then the plugin could just be the piece that sends data to the microservice.
shemnon (Thu, 19 Aug 2021 20:01:31 GMT):
Assuming it's a triggerede action and not something on a timer.
shemnon (Thu, 19 Aug 2021 20:01:31 GMT):
Assuming it's a triggered action and not something on a timer.
lucassaldanha (Fri, 20 Aug 2021 02:37:18 GMT):
If I remember correctly we write the ports used by Besu during startup in a file somewhere. Maybe it could be useful?
atoulme (Fri, 20 Aug 2021 05:30:53 GMT):
that looks like you're trying to do too much. Do you mean to make this a k8s deployment down the road? A helm chart can paper over this stuff
antonydenyer (Fri, 20 Aug 2021 06:14:50 GMT):
Writing a custom onchain permissioning implementation.
antonydenyer (Fri, 20 Aug 2021 06:16:36 GMT):
permissioning -> plugin -> query bespoke contract -> true/false
antonydenyer (Fri, 20 Aug 2021 09:33:16 GMT):
This is what I had in mind https://github.com/hyperledger/besu/pull/2661/files
atoulme (Fri, 20 Aug 2021 12:28:26 GMT):
yeah, that's hard to make elegant
atoulme (Fri, 20 Aug 2021 12:29:35 GMT):
so in the tooling you want to use web3j, and therefore that's why you need access to http
atoulme (Fri, 20 Aug 2021 12:30:02 GMT):
so having some sort of native jsonrpc server would serve you way better
antonydenyer (Fri, 20 Aug 2021 12:30:49 GMT):
exactly
atoulme (Fri, 20 Aug 2021 12:31:17 GMT):
I mean, ok that sounds doable to me
atoulme (Fri, 20 Aug 2021 12:31:31 GMT):
I might be way off but it might be easier to do that than anything else
atoulme (Fri, 20 Aug 2021 12:32:08 GMT):
if you torture your code to use http you introduce inherent messes, if it's inside the same process. You'd need to architecture out a different way and not use a plugin as much
atoulme (Fri, 20 Aug 2021 12:32:33 GMT):
either it's integrated or it's not :)
atoulme (Fri, 20 Aug 2021 12:33:03 GMT):
how would web3j provide an interface for a provider?
antonydenyer (Fri, 20 Aug 2021 12:37:49 GMT):
|> if it's inside the same process
that's the beauty - it's in a different thread
atoulme (Fri, 20 Aug 2021 12:38:50 GMT):
I mean, not to rail here, but you're going to call your own process over http and incur the cost of json marshalling/unmarshalling?
atoulme (Fri, 20 Aug 2021 12:38:56 GMT):
That's bonkers to me.
antonydenyer (Fri, 20 Aug 2021 12:39:27 GMT):
yeah it's not ideal ... but
antonydenyer (Fri, 20 Aug 2021 12:39:55 GMT):
it's standards compliant and has a bunch of tooling available and requires no extra stuff in besu
atoulme (Fri, 20 Aug 2021 12:40:29 GMT):
no, but hear me out, that struggle is real
antonydenyer (Fri, 20 Aug 2021 12:40:42 GMT):
~requires no extra stuff in besu~ or more importantly less stuff to maintain
atoulme (Fri, 20 Aug 2021 12:40:47 GMT):
if you open up http just for internals, then you _require_ http to be opened. You also require methods to be opened.
antonydenyer (Fri, 20 Aug 2021 12:41:08 GMT):
true
atoulme (Fri, 20 Aug 2021 12:41:21 GMT):
And then when you go to deploy, you'll find yourself in a pickle because you're opening up potentially internal stuff you wouldn't want to expose
atoulme (Fri, 20 Aug 2021 12:41:27 GMT):
let me go with an example
atoulme (Fri, 20 Aug 2021 12:41:42 GMT):
we have a tool ethlogger that we use to ingest data from evm chains over json-rpc
atoulme (Fri, 20 Aug 2021 12:41:48 GMT):
it likes to get the admin methods
atoulme (Fri, 20 Aug 2021 12:42:15 GMT):
but we found out the hard way we can't just open this for ethlogger. If we do that, then jsonrpc cannot be opened to the rest of the world.
atoulme (Fri, 20 Aug 2021 12:42:26 GMT):
Kinda same quandary for you here
atoulme (Fri, 20 Aug 2021 12:42:50 GMT):
In our case, we set out to introduce a jsonrpc proxy in front of geth to do caching and filtering
atoulme (Fri, 20 Aug 2021 12:43:05 GMT):
but now we're running even more software lol
atoulme (Fri, 20 Aug 2021 12:43:22 GMT):
anyway, maybe this is fine for now
antonydenyer (Fri, 20 Aug 2021 12:43:34 GMT):
yeah - there's some responsibility on the node operator to know not to expose the ports to the world only internally
atoulme (Fri, 20 Aug 2021 12:44:52 GMT):
yeah you think they know?
atoulme (Fri, 20 Aug 2021 12:45:11 GMT):
it's difficult to filter based on request payload
antonydenyer (Fri, 20 Aug 2021 12:45:23 GMT):
they will eventually :laughing:
atoulme (Fri, 20 Aug 2021 12:45:48 GMT):
I don't want to be there for that, or when they find this chat
antonydenyer (Fri, 20 Aug 2021 12:47:17 GMT):
the first lot obviously know - it's the people after the people after!
antonydenyer (Fri, 20 Aug 2021 12:48:17 GMT):
the other option would be ipc ... but no ipc sockets in besu :disappointed:
atoulme (Fri, 20 Aug 2021 12:48:32 GMT):
it's a PR away you know
antonydenyer (Fri, 20 Aug 2021 12:48:44 GMT):
it would at least stop you from making an easy mistake
atoulme (Fri, 20 Aug 2021 12:48:47 GMT):
you could be the hero you need
atoulme (Fri, 20 Aug 2021 12:48:58 GMT):
do it Antony
atoulme (Fri, 20 Aug 2021 12:49:03 GMT):
implement ipc
atoulme (Fri, 20 Aug 2021 12:49:05 GMT):
please
antonydenyer (Fri, 20 Aug 2021 12:50:40 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=jvRxCZe8xMWKkmCvb) https://github.com/lucassaldanha/vertx-jsonrpc
antonydenyer (Fri, 20 Aug 2021 12:50:48 GMT):
if only we knew that guy!
antonydenyer (Fri, 20 Aug 2021 12:51:40 GMT):
seem to remember something about ipc being a pain with vertx
atoulme (Fri, 20 Aug 2021 12:52:48 GMT):
:shrug: I'm not qualified
antonydenyer (Fri, 20 Aug 2021 12:54:04 GMT):
https://github.com/hyperledger/besu/issues/535
atoulme (Fri, 20 Aug 2021 12:54:54 GMT):
ok I can press buttons, reopening
shemnon (Fri, 20 Aug 2021 13:06:37 GMT):
I think it's a bad idea. CLI flags provide no context as to their meaning and if we have plugins that depend on certain flags we restrict our future ability to evolve the CLI options in any way. There are too many layers of organization that are blurred in this method. Plugins should not care what other systems CLI options are, it's bad design.
shemnon (Fri, 20 Aug 2021 13:06:37 GMT):
I think it's a bad idea. CLI flags provide no context as to their meaning and if we have plugins that depend on certain flags we restrict our future ability to evolve the CLI options in any way. There are too many layers of organization that are blurred in this method. Plugins should not care what other system CLI options are, it's bad design. Any data they need should be passed via the systems that are configured with that CLI, not via the CLI.
shemnon (Fri, 20 Aug 2021 13:06:37 GMT):
I think it's a bad idea. CLI flags provide no context as to their meaning and if we have plugins that depend on certain flags we restrict our future ability to evolve the CLI options in any way. There are too many layers of organization that are blurred in this method. Plugins should not care what other system CLI options are, it's bad design. Any data they need should be passed via the systems that are configured with that CLI, not via the CLI configuring that system.
antonydenyer (Fri, 20 Aug 2021 13:09:28 GMT):
that's fair enough - but we're back to exposing `BlockchainQueries`
antonydenyer (Fri, 20 Aug 2021 13:11:32 GMT):
or you end with two params.
`--rpc-http-port` and `--plugin-name-rpc-http-port`
antonydenyer (Fri, 20 Aug 2021 13:18:54 GMT):
or we could get rid of the checks at https://github.com/hyperledger/besu/blob/main/besu/src/main/java/org/hyperledger/besu/services/PicoCLIOptionsImpl.java#L42
shemnon (Fri, 20 Aug 2021 13:20:01 GMT):
What exactly is your plugin trying to do?
antonydenyer (Fri, 20 Aug 2021 13:20:43 GMT):
https://chat.hyperledger.org/channel/besu-contributors?msg=rSpHGAqRk36TLMfGd
antonydenyer (Fri, 20 Aug 2021 13:21:13 GMT):
I think Antoine had similar concerns
antonydenyer (Fri, 20 Aug 2021 13:23:22 GMT):
hmm links are terrible in rocker chat :laughing:
antonydenyer (Fri, 20 Aug 2021 13:23:22 GMT):
hmm links are terrible in rocket chat :laughing:
atoulme (Fri, 20 Aug 2021 13:40:07 GMT):
I'm not concerned, just enjoying the conundrum
antonydenyer (Fri, 20 Aug 2021 13:47:58 GMT):
there's no right answer - just trade offs :smirk:
atoulme (Fri, 20 Aug 2021 13:48:49 GMT):
how does one configure web3j to connect - is there a programmatic provider?
antonydenyer (Fri, 20 Aug 2021 13:50:00 GMT):
yeah `Web3jService`
atoulme (Fri, 20 Aug 2021 13:50:44 GMT):
could besu implement InProcessService?
atoulme (Fri, 20 Aug 2021 13:52:09 GMT):
it could, couldn't it?
antonydenyer (Fri, 20 Aug 2021 13:53:24 GMT):
well, anything is possible right :smile:
atoulme (Fri, 20 Aug 2021 13:55:21 GMT):
yeah I'm not familiar enough as to whether the plugin data access covers enough right now that you could implement a Web3jService with it
atoulme (Fri, 20 Aug 2021 13:55:30 GMT):
or if that requires additional access
atoulme (Fri, 20 Aug 2021 13:55:56 GMT):
ideally your plugin is implementing Web3jService, and Besu offers enough primitives it's a breeze
lucassaldanha (Fri, 20 Aug 2021 23:11:49 GMT):
> if only we knew that guy!
lucassaldanha (Fri, 20 Aug 2021 23:11:50 GMT):
lol
lucassaldanha (Fri, 20 Aug 2021 23:16:17 GMT):
> seem to remember something about ipc being a pain with vertx
Yeah, it can be
lucassaldanha (Fri, 20 Aug 2021 23:17:27 GMT):
However, if you don't mind being locked in into Vertx, you could make use of Vertx eventbus btw processes (usign Vertx cluster)
lucassaldanha (Fri, 20 Aug 2021 23:19:40 GMT):
But unfortunately, our RPC implementation isnt' really making a good use of the eventbus, so maybe it isn't that trivial.
atoulme (Fri, 20 Aug 2021 23:20:33 GMT):
just make a bunch of code changes!
atoulme (Fri, 20 Aug 2021 23:20:47 GMT):
ok I'll be over there.
lucassaldanha (Fri, 20 Aug 2021 23:21:33 GMT):
haha
lucassaldanha (Sat, 21 Aug 2021 05:43:03 GMT):
FYI I have found a bug on the secp256r1 unit tests when running on MacOS. This is relative to the use of the native bindings from besu-native. https://github.com/hyperledger/besu/issues/2665
atoulme (Sat, 21 Aug 2021 14:13:18 GMT):
Review please https://github.com/hyperledger/besu/pull/2666
atoulme (Sun, 22 Aug 2021 18:26:25 GMT):
hey everyone, I'm seeing some internet noise about another Ethereum client developer behavior making the rounds because they are critical, blunt and even could be misconstrued as aggressive. Their message seems worthwhile, but the conveying is a bit harsh. Note I lost interest and didn't read all of it, maybe a better analysis can come along in time.
I wanted to state that I enjoy working with you all as professionals. I am also reassured that we as committers have a code of conduct to address this kind of behavior. Let's just keep grinding.
shemnon (Sun, 22 Aug 2021 21:57:01 GMT):
I've gotten grief from some contributors that CoCs are just there to push leftist ideologies. They are not. The only times I've seen directly CoCs actually become relevant (and enforced!) is in garden variety abusive behavior.
shemnon (Sun, 22 Aug 2021 21:57:01 GMT):
I've gotten grief from some former on-pr contributors that CoCs are just there to push leftist ideologies. They are not. The only times I've seen directly CoCs actually become relevant (and enforced!) is in garden variety abusive behavior.
shemnon (Sun, 22 Aug 2021 21:57:01 GMT):
I've gotten grief from some former one-pr contributors that CoCs are just there to push leftist ideologies. They are not. The only times I've seen directly CoCs actually become relevant (and enforced!) is in garden variety abusive behavior.
shemnon (Sun, 22 Aug 2021 21:57:01 GMT):
I've gotten grief from some former one-pr contributors that CoCs are just there to push leftist ideologies. They are not. The only times I've seen CoCs directly become relevant (and enforced!) is in garden variety abusive behavior.
shemnon (Sun, 22 Aug 2021 21:57:34 GMT):
CoCs are just legally enforceable statements of "Be Excellent To Each Other"
shemnon (Sun, 22 Aug 2021 22:05:33 GMT):
Just thought of this, and I don't have the code handy to inspect. If an enterprise network goes to the london hard fork must they take 1559 with them? And the base fee floor of 7wei? Are there configs that can be flipped to ensure block sizes never go into "too full" mode and up the base fee?
lucassaldanha (Mon, 23 Aug 2021 07:59:13 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=AYtSutzXCqtWKFMX2) Following up on this. We have fixed the issue and it will require a new release of besu-native (0.4.2). I'll update Besu later on with the new version of the library.
lucassaldanha (Mon, 23 Aug 2021 10:39:42 GMT):
And here is the PR updating besu-native version with the MacOS fix: https://github.com/hyperledger/besu/pull/2669
lucassaldanha (Mon, 23 Aug 2021 10:39:42 GMT):
Review request: here is the PR updating besu-native version with the MacOS fix: https://github.com/hyperledger/besu/pull/2669
shemnon (Mon, 23 Aug 2021 13:57:59 GMT):
Is there a reason besu-ntive-ec is a consensys repo rather than a hyperledger repo? I'm not seeing any consensys specific tweaks. Is it just because consensys repos are quicker to spin up?
shemnon (Mon, 23 Aug 2021 20:02:08 GMT):
@here Hyperledger will be moving off of rocket chat onto matrix chat sometime in the next quarter or three. To allow us to migrate earlier than the forced migration I asked RyJones to make a contributor room for us - https://matrix.to/#/#hyperledger-besu-contributors:chat.lfx.linuxfoundation.org
shemnon (Mon, 23 Aug 2021 20:03:15 GMT):
Matrix allows multiple clients to access the chat service. I myself use Element on the desktop and my mobile devices. There are others but I'm not sure which ones are good.
lucassaldanha (Mon, 23 Aug 2021 23:22:39 GMT):
It is easier to spin up. AFAIK there is no reason for not migrating it to HL later on. I'll send a msg to Csy strategy about this.
lucassaldanha (Mon, 23 Aug 2021 23:24:41 GMT):
Should we add this to the next contributor call? If we can come up with a date for the migration, we can start communicating.
lucassaldanha (Mon, 23 Aug 2021 23:24:59 GMT):
Also, we should probably have a channel for "non-contributor" as well.
shemnon (Mon, 23 Aug 2021 23:35:27 GMT):
We do have a users channel - https://matrix.to/#/#hyperledger-besu:chat.lfx.linuxfoundation.org
shemnon (Mon, 23 Aug 2021 23:35:44 GMT):
but agreed, contributor call subject.
shemnon (Mon, 23 Aug 2021 23:36:00 GMT):
And we don't need to rush over, we can set our own timetable (for the time being)
shemnon (Mon, 23 Aug 2021 23:36:43 GMT):
While we are at it we can pull out the errorprone messages too.
lucassaldanha (Tue, 24 Aug 2021 02:22:19 GMT):
I added an item to https://wiki.hyperledger.org/display/BESU/2021-08-31+Besu+Contributor+Call
NicolasMassart (Tue, 24 Aug 2021 09:15:36 GMT):
I'm switching off the HotJar survey and feedback on Besu doc for now as it only provides useless feedback (no comment, spam,...) and makes me think that users are more annoyed than happy of being asked. I will discuss a better way to get feedback with the community management team.
RoboCopsGoneMad (Tue, 24 Aug 2021 17:09:31 GMT):
I am considering adding something to EvmTool to allow it to fail a test because it took too long to complete. This is a shortcut to performance testing. Any thoughts on that?
shemnon (Tue, 24 Aug 2021 17:32:52 GMT):
As a CLI flag? Sounds reasonable.
RoboCopsGoneMad (Tue, 24 Aug 2021 17:44:42 GMT):
I was thinking of making it an optional field on the Post section of the supplied json. if it is present, fail a test run that took longer than the value
RoboCopsGoneMad (Tue, 24 Aug 2021 17:45:56 GMT):
issue being worked, for reference https://github.com/hyperledger/besu/issues/2607
shemnon (Tue, 24 Aug 2021 18:25:26 GMT):
Would the reference tests gain this field?
anuragsingh (Tue, 24 Aug 2021 18:54:58 GMT):
Has joined the channel.
anuragsingh (Tue, 24 Aug 2021 18:56:48 GMT):
Hello everyone, I'm new to hyperledger and blockchain technology itself and want to contribute to Besu as I have some prior experience with Java. Could anyone guide me where should I begin to understand blockchain and hyperledger? It would be really helpful!
atoulme (Wed, 25 Aug 2021 02:53:05 GMT):
https://chat.hyperledger.org/channel/besu?msg=NubbzeHz5YaNRoLEu ;)
anuragsingh (Wed, 25 Aug 2021 04:36:42 GMT):
Thanks a lot!
atoulme (Wed, 25 Aug 2021 07:09:42 GMT):
hey all, I played with besu storage to get better performance, and I would like to benchmark. What's a good use case? Syncing? Or something else? What are good performance markers - CPU/RAM used and how long? Is that documented somewherte?
RoboCopsGoneMad (Wed, 25 Aug 2021 13:23:19 GMT):
A full archive sync is probably best, we did that for the Bonsai trie testing.
RoboCopsGoneMad (Wed, 25 Aug 2021 13:23:48 GMT):
CPU over time is probably more useful than RAM.
RoboCopsGoneMad (Wed, 25 Aug 2021 13:24:16 GMT):
CPU spend on GC is a better indicator than RAM itself.
RoboCopsGoneMad (Wed, 25 Aug 2021 13:25:12 GMT):
:thinking:
RoboCopsGoneMad (Wed, 25 Aug 2021 13:26:10 GMT):
after that, perhaps RPC performance... though thats gonna ask more questions than it answers
shemnon (Wed, 25 Aug 2021 13:48:09 GMT):
And there's two sync modes to consier: fast and full. For full sync you will want to benchmark the latter part of a test network full sync, when there are actual TSes on the blocks. early blocks of most testnets are quite sparse.
shemnon (Wed, 25 Aug 2021 13:48:09 GMT):
And there's two sync modes to consider: fast and full. For full sync you will want to benchmark the latter part of a test network full sync, when there are actual TXes on the blocks. Early blocks of most testnets are quite sparse, just like mainnet.
atoulme (Wed, 25 Aug 2021 14:13:35 GMT):
yeah but if you sync in the wild, your network can crap out, your peers go away, etc.
atoulme (Wed, 25 Aug 2021 14:14:03 GMT):
I will probably create a scenario with 2 nodes on a local network with explicit sync 1:1 with a static peer.
atoulme (Wed, 25 Aug 2021 14:15:21 GMT):
has anyone made archives of goerli nodes? I have a couple mainnet nodes here, looking to get a goerli node, but it's been hell to get good peers.
VijayMichalik (Wed, 25 Aug 2021 14:43:29 GMT):
likewise, thank you all for embodying great community values and keeping things welcoming :raised_hands:
VijayMichalik (Wed, 25 Aug 2021 14:46:02 GMT):
I think the base fee floor is constrained to mainnet but this is definitely something to look into
VijayMichalik (Wed, 25 Aug 2021 14:47:43 GMT):
I'd like to nudge us to upgrade the other enterprise features to support eip1559 type transactions since we want to stay compatible with as much external tooling as possible, and particularly for gas-enabled private networks
VijayMichalik (Wed, 25 Aug 2021 14:50:39 GMT):
I believe we do! @joshuafernandes?
VijayMichalik (Wed, 25 Aug 2021 14:53:45 GMT):
I would love this to be sooner than later! we really want easy connectivity between our existing discord and the hyperledger chat, given how much web3 folks love discord
VijayMichalik (Wed, 25 Aug 2021 14:56:58 GMT):
Just raising my hand - if any contributors want to suggest roadmap items for the next year, I'd love to chat them through with you to help write them up as proposals on the Besu wiki, ahead of discussing them on a contributor call. I'm doing a bunch of others right now anyway!
shemnon (Wed, 25 Aug 2021 15:25:26 GMT):
I don't know if they've set up discord bridges yet. This may be the motivation needed.
grace.hartley (Wed, 25 Aug 2021 15:25:46 GMT):
Hey folks! Hyperledger staff has asked the Hyperledger Besu team to participate in Grace Hopper's Open Source Day on October 1st. It is a great opportunity to showcase Hyperledger Besu and get new developers working on our codebase. To prep for this event, Besu needs to have 30+ good first issues labeled in the project. Besu currently has 14 good first issues opened. Can this group work together to add any additional good first issues in the coming weeks? We need the 30 to be uploaded by September 15th. Thanks!
GarySchulte (Wed, 25 Aug 2021 17:23:56 GMT):
yeah, london == 1559. there are no flags to disable 1559 fee market mechanism, it is all driven by protocolSchedule.
What would an enterprise network stand to gain from enabling london, but not the fee market mechanism? AFAICT a 1559 transaction is essentially an ACCESS_LIST transaction with base fee/tip instead of gasPrice
GarySchulte (Wed, 25 Aug 2021 17:28:42 GMT):
it seems like it would not be terribly difficult to implement a flag for static baseFee mechanism if we were worried about tooling dropping feature support for legacy transactions
GarySchulte (Wed, 25 Aug 2021 17:32:04 GMT):
I think it could be as simple as adjusting max change denominator slack coefficient, and initial base fee to values that cause the basefee computation to always arrive at the same number, see LondonFeeMarket
shemnon (Wed, 25 Aug 2021 19:13:26 GMT):
For zero fee networks I think they want to make sure that the basefee stays at zero. i.e. no levering up if the blocks goes > 50% full.
shemnon (Wed, 25 Aug 2021 19:13:49 GMT):
So if the basefee growth % could be set to zero that should address it.
GarySchulte (Wed, 25 Aug 2021 19:15:15 GMT):
I think if initial basefee = 0, then the arithmetic would prevent it from ever changing. would have to verify that though
GarySchulte (Wed, 25 Aug 2021 19:15:58 GMT):
there might be some divide by zero concerns. but generally speaking I think tweaking those constants would work
atoulme (Thu, 26 Aug 2021 04:18:06 GMT):
added one https://github.com/hyperledger/besu/issues/2678
atoulme (Thu, 26 Aug 2021 06:59:33 GMT):
Do we package the native libs with Besu during assembly?
shemnon (Thu, 26 Aug 2021 12:52:03 GMT):
We should. `besu-native` packages them up into the jar that it builds, and we consume that jar. The privacy and testing have, however, created a dependency on non-packaged jars. Namely libSodium and nss. Those should be better documented as I _still_ can't get `nss` dependent tests running on an intel or m1 mac. `brew install nss` doesn't put the libs in a place the JCE wants to see it.
grace.hartley (Thu, 26 Aug 2021 13:48:15 GMT):
thanks!
atoulme (Thu, 26 Aug 2021 14:23:10 GMT):
Closed it… wasn’t a good one
atoulme (Thu, 26 Aug 2021 15:13:56 GMT):
I am seeing a weird behavior in my test machine. I use a server locally that is running 2 besu nodes. The server reports slowdowns here:
atoulme (Thu, 26 Aug 2021 15:14:02 GMT):
```
2021-08-26 00:00:18.288-07:00 | vertx-blocked-thread-checker | WARN | BlockedThreadChecker | Thread Thread[vert.x-eventloop-thread-1,5,main]=Thread[vert.x-eventloop-thread-1,5,main] has been blocked for 6318 ms, time limit is 2000 ms
io.vertx.core.VertxException: Thread blocked
at java.base@11.0.11/java.util.Arrays.copyOfRange(Arrays.java:4030)
at java.base@11.0.11/sun.security.provider.NativePRNG$RandomIO.implNextBytes(NativePRNG.java:554)
at java.base@11.0.11/sun.security.provider.NativePRNG.engineNextBytes(NativePRNG.java:220)
at java.base@11.0.11/java.security.SecureRandom.nextBytes(SecureRandom.java:751)
at app//org.hyperledger.besu.ethereum.p2p.peers.Peer.randomId(Peer.java:38)
at app//org.hyperledger.besu.ethereum.p2p.discovery.internal.PeerDiscoveryController.refreshTable(PeerDiscoveryController.java:449)
at app//org.hyperledger.besu.ethereum.p2p.discovery.internal.PeerDiscoveryController.refreshTableIfRequired(PeerDiscoveryController.java:429)
at app//org.hyperledger.besu.ethereum.p2p.discovery.internal.PeerDiscoveryController$$Lambda$1041/0x000000084074ec40.handle(Unknown Source)
at app//org.hyperledger.besu.ethereum.p2p.discovery.internal.VertxTimerUtil.lambda$setPeriodic$0(VertxTimerUtil.java:29)
at app//org.hyperledger.besu.ethereum.p2p.discovery.internal.VertxTimerUtil$$Lambda$1042/0x000000084074f040.handle(Unknown Source)
at app//io.vertx.core.impl.VertxImpl$InternalTimerHandler.handle(VertxImpl.java:938)
at app//io.vertx.core.impl.VertxImpl$InternalTimerHandler.handle(VertxImpl.java:906)
at app//io.vertx.core.impl.ContextImpl.executeTask(ContextImpl.java:366)
at app//io.vertx.core.impl.EventLoopContext.execute(EventLoopContext.java:43)
at app//io.vertx.core.impl.ContextImpl.executeFromIO(ContextImpl.java:229)
at app//io.vertx.core.impl.ContextImpl.executeFromIO(ContextImpl.java:221)
```
atoulme (Thu, 26 Aug 2021 15:14:35 GMT):
I see it with 20 minutes intervals of running. Restarting Besu brings it back up for a bit but eventually the behavior reproduces.
atoulme (Thu, 26 Aug 2021 15:15:15 GMT):
I am running haveged on the machine to create entropy. I guess this machine has entropy issues in the first place - it's rather old.
atoulme (Thu, 26 Aug 2021 15:15:40 GMT):
Any immediate fixes that come to mind?
atoulme (Thu, 26 Aug 2021 15:16:49 GMT):
I can try to snapshot threads next time I run into this, to see which other threads are waiting on randomness. In the mean time, is there maybe a way to reduce the need for randomness in the scenario above? I'll take a quick look.
shemnon (Thu, 26 Aug 2021 15:53:56 GMT):
Any other stack traces? Are they all stuck in peer.RandomId? Or is the peer table too large?
shemnon (Thu, 26 Aug 2021 15:54:05 GMT):
Or are peers slow in responding?
atoulme (Thu, 26 Aug 2021 16:00:04 GMT):
No, no others. I’m going to continue monitoring.
shemnon (Thu, 26 Aug 2021 16:20:05 GMT):
We wound up disabling real randomness for our acceptance tests, we kept drying up all the entropy on the circleci machines. https://github.com/hyperledger/besu/blob/6fbc7732760862f960391383c3ff4efb309f2b91/acceptance-tests/tests/build.gradle#L80
shemnon (Thu, 26 Aug 2021 16:21:38 GMT):
but this is already prng, so I"m at a loss why it would take 20 minutes to copy an array.
atoulme (Thu, 26 Aug 2021 19:05:28 GMT):
no
atoulme (Thu, 26 Aug 2021 19:06:01 GMT):
after 20 minutes it takes 6 seconds to get a set of random bytes.
atoulme (Thu, 26 Aug 2021 19:06:30 GMT):
to be exact, 64 bytes.
atoulme (Thu, 26 Aug 2021 23:00:55 GMT):
here is an interesting thread dump. I get several threads locked up here:
```
"vert.x-worker-thread-8" #51 prio=5 os_prio=0 cpu=32946.18ms elapsed=26535.11s tid=0x00007f12b00d1800 nid=0x7d14 waiting for monitor entry [0x00007f14413de000]
java.lang.Thread.State: BLOCKED (on object monitor)
at sun.security.provider.NativePRNG$RandomIO.implNextBytes(java.base@11.0.11/NativePRNG.java:544)
- waiting to lock <0x00000006e8bbeba0> (a java.lang.Object)
at sun.security.provider.NativePRNG.engineNextBytes(java.base@11.0.11/NativePRNG.java:220)
at java.security.SecureRandom.nextBytes(java.base@11.0.11/SecureRandom.java:751)
at org.bouncycastle.util.BigIntegers.createRandom(BigIntegers.java:342)
at org.bouncycastle.util.BigIntegers.createRandomBigInteger(BigIntegers.java:272)
at org.bouncycastle.math.ec.ECCurve$AbstractFp.implRandomFieldElementMult(ECCurve.java:658)
at org.bouncycastle.math.ec.ECCurve$AbstractFp.randomFieldElementMult(ECCurve.java:614)
at org.bouncycastle.math.ec.ECPoint.normalize(ECPoint.java:250)
at org.bouncycastle.math.ec.ECPoint.getEncoded(ECPoint.java:490)
at org.hyperledger.besu.crypto.AbstractSECP256.recoverFromSignature(AbstractSECP256.java:303)
at org.hyperledger.besu.crypto.AbstractSECP256.normaliseSignature(AbstractSECP256.java:109)
at org.hyperledger.besu.crypto.NodeKey.sign(NodeKey.java:34)
at org.hyperledger.besu.ethereum.p2p.discovery.internal.Packet.
atoulme (Thu, 26 Aug 2021 23:04:32 GMT):
jstack.txt
atoulme (Thu, 26 Aug 2021 23:06:37 GMT):
over 10 threads locked waiting for the shared SecureRandom.
atoulme (Thu, 26 Aug 2021 23:20:53 GMT):
unfortunately, most point to this particular code path that was changed recently to support multiple curves:
```
at sun.security.provider.NativePRNG$RandomIO.implNextBytes(java.base@11.0.11/NativePRNG.java:544)
- waiting to lock <0x00000006e8bbeba0> (a java.lang.Object)
at sun.security.provider.NativePRNG.engineNextBytes(java.base@11.0.11/NativePRNG.java:220)
at java.security.SecureRandom.nextBytes(java.base@11.0.11/SecureRandom.java:751)
at org.bouncycastle.util.BigIntegers.createRandom(BigIntegers.java:342)
at org.bouncycastle.util.BigIntegers.createRandomBigInteger(BigIntegers.java:272)
at org.bouncycastle.math.ec.ECCurve$AbstractFp.implRandomFieldElementMult(ECCurve.java:658)
at org.bouncycastle.math.ec.ECCurve$AbstractFp.randomFieldElementMult(ECCurve.java:614)
at org.bouncycastle.math.ec.ECPoint.normalize(ECPoint.java:250)
at org.bouncycastle.math.ec.ECPoint.getEncoded(ECPoint.java:490)
at org.hyperledger.besu.crypto.AbstractSECP256.recoverFromSignature(AbstractSECP256.java:303)
at org.hyperledger.besu.crypto.AbstractSECP256.normaliseSignature(AbstractSECP256.java:109)
at org.hyperledger.besu.crypto.NodeKey.sign(NodeKey.java:34)
at org.hyperledger.besu.ethereum.p2p.discovery.internal.Packet.
macfarla (Fri, 27 Aug 2021 00:06:35 GMT):
@lucassaldanha know anything about this ^
lucassaldanha (Fri, 27 Aug 2021 00:28:34 GMT):
I'll look at it in a second
lucassaldanha (Fri, 27 Aug 2021 00:29:15 GMT):
I'm fixing a commit missing the DCO
atoulme (Fri, 27 Aug 2021 00:29:27 GMT):
there is no hurry.
shemnon (Fri, 27 Aug 2021 00:29:32 GMT):
It's something bouncy castle is doing... *all* normalized points are subject to random blinding. https://github.com/bcgit/bc-java/commit/87ab5d8470829879219e50213912bab6b1ab8fe8
atoulme (Fri, 27 Aug 2021 00:29:39 GMT):
I just see this on that machine.
shemnon (Fri, 27 Aug 2021 00:29:48 GMT):
Because of some security vulnerability report.
atoulme (Fri, 27 Aug 2021 00:30:00 GMT):
Yes, I am not saying BC is wrong.
atoulme (Fri, 27 Aug 2021 00:30:15 GMT):
FWIW I am now trying to use /dev/urandom to get by.
shemnon (Fri, 27 Aug 2021 00:30:16 GMT):
I'm not willing to let them off the hook.
atoulme (Fri, 27 Aug 2021 00:30:41 GMT):
Look, I got mileage for free from them and I never paid them a dollar. They can have this.
shemnon (Fri, 27 Aug 2021 00:31:09 GMT):
There should have been some flag with "export _THIS_IS_INSECURE_DO_NOT_DO_BLIND_NORMALZIATION=true"
shemnon (Fri, 27 Aug 2021 00:31:09 GMT):
There should have been some flag with `export _THIS_IS_INSECURE_DO_NOT_DO_BLIND_NORMALZIATION=true`
atoulme (Fri, 27 Aug 2021 00:31:21 GMT):
sure
atoulme (Fri, 27 Aug 2021 00:32:01 GMT):
looks like we could override their secure random maybe
atoulme (Fri, 27 Aug 2021 00:32:07 GMT):
I spent zero time looking at BC
atoulme (Fri, 27 Aug 2021 00:32:21 GMT):
and was more focused on why we do this weird signature recovery when we're signing
atoulme (Fri, 27 Aug 2021 00:32:40 GMT):
looking at that trace
shemnon (Fri, 27 Aug 2021 00:33:37 GMT):
It's beein in for over a year, so I don't think that's it.
shemnon (Fri, 27 Aug 2021 00:33:37 GMT):
It (bouncy castle > 1.66) has been in in for over a year, so I don't think that's it.
lucassaldanha (Fri, 27 Aug 2021 00:33:38 GMT):
[IMPORTANT] The head of `main` branch for repository `hyperledger/besu` has been rebased to deal with a DCO issue, please follow the instructions on Fixing DCO on main/release to fix your local repositories
lucassaldanha (Fri, 27 Aug 2021 00:33:38 GMT):
[IMPORTANT] The head of `main` branch for repository `hyperledger/besu` has been rebased to deal with a DCO issue, please follow the instructions on Fixing DCO on main/release to fix your local repositories
https://wiki.hyperledger.org/pages/viewpage.action?pageId=24781754
atoulme (Fri, 27 Aug 2021 00:34:00 GMT):
We want to sign something with a node key:
at org.hyperledger.besu.crypto.NodeKey.sign(NodeKey.java:34)
which makes us normalize the signature
at org.hyperledger.besu.crypto.AbstractSECP256.normaliseSignature(AbstractSECP256.java:109)
which triggers a recover?
at org.hyperledger.besu.crypto.AbstractSECP256.recoverFromSignature(AbstractSECP256.java:303)
shemnon (Fri, 27 Aug 2021 00:34:33 GMT):
Signing would trigger the blinding. It's the most important place to blind TBH
shemnon (Fri, 27 Aug 2021 00:34:41 GMT):
recover must be how they do it.
shemnon (Fri, 27 Aug 2021 00:36:13 GMT):
Is this R1? Because K1 should be native. That code is tight and awesome. One thing we can easly credit bitcoin for.
atoulme (Fri, 27 Aug 2021 00:36:22 GMT):
no, we normalize the signature, and that makes us go for a a recovery
atoulme (Fri, 27 Aug 2021 00:37:06 GMT):
ah, now I'm confused.
atoulme (Fri, 27 Aug 2021 00:38:47 GMT):
yeah this is the exact same code that was here: https://github.com/hyperledger/besu/blob/20.10.3/crypto/src/main/java/org/hyperledger/besu/crypto/SECP256K1.java#L270
atoulme (Fri, 27 Aug 2021 00:39:10 GMT):
so the changes to move this code out to AbstractSECP256 didn't affect the logic.
lucassaldanha (Fri, 27 Aug 2021 00:40:05 GMT):
When we introduced the AbstractSECP256 we tried to keep the same code to prevent mistakes
atoulme (Fri, 27 Aug 2021 00:40:22 GMT):
yeah sorry. I'm barking at the wrong thing.
lucassaldanha (Fri, 27 Aug 2021 00:40:31 GMT):
Question: you are using the default K1, aren't you?
atoulme (Fri, 27 Aug 2021 00:40:35 GMT):
yeh
shemnon (Fri, 27 Aug 2021 00:40:48 GMT):
normalzie signature isn't using native facilities. That's a performance regression.
lucassaldanha (Fri, 27 Aug 2021 00:40:53 GMT):
And did you set anything to use BC instead of the native impl?
atoulme (Fri, 27 Aug 2021 00:41:10 GMT):
no I didn't set anything
shemnon (Fri, 27 Aug 2021 00:41:11 GMT):
The code path for normalize never gets a native option.
atoulme (Fri, 27 Aug 2021 00:41:24 GMT):
right
lucassaldanha (Fri, 27 Aug 2021 00:41:29 GMT):
ok that would explain it
atoulme (Fri, 27 Aug 2021 00:41:51 GMT):
so maybe that's a worthwhile improvement, run sig recovery with the native option
lucassaldanha (Fri, 27 Aug 2021 00:42:26 GMT):
@shemnon do you know if this is something that changed? Did we ever use the native option for normalize?
shemnon (Fri, 27 Aug 2021 00:42:44 GMT):
yes. It was the R1 change that created new paths and didn't wire in K1 paths.
lucassaldanha (Fri, 27 Aug 2021 00:42:49 GMT):
got it
lucassaldanha (Fri, 27 Aug 2021 00:43:23 GMT):
@atoulme I'll handle this if you want
lucassaldanha (Fri, 27 Aug 2021 00:43:30 GMT):
Unless you want to fix it :P
shemnon (Fri, 27 Aug 2021 00:44:05 GMT):
`public SECPSignature normaliseSignature(
final BigInteger nativeR,
final BigInteger nativeS,
final SECPPublicKey publicKey,
final Bytes32 dataHash) {` does pure java recovery, and a very unelightend recovery forced mehtod.
atoulme (Fri, 27 Aug 2021 00:44:22 GMT):
right
atoulme (Fri, 27 Aug 2021 00:44:28 GMT):
it runs recovery up to 4 times
shemnon (Fri, 27 Aug 2021 00:44:28 GMT):
Ethereum keys keep the sign bit so we don't need to do the 4x search
atoulme (Fri, 27 Aug 2021 00:44:44 GMT):
:|
shemnon (Fri, 27 Aug 2021 00:51:23 GMT):
It was actually this change that took the native path out... https://github.com/hyperledger/besu/commit/d6fabbda70d65ddac56bdda630a0e4b244cb2762#diff-7446961ea2f50a25525de78f67731cd449744cb25ed1feb026c7722013a68cdfR29
shemnon (Fri, 27 Aug 2021 00:51:35 GMT):
normalzie never had a native path, but create did.
atoulme (Fri, 27 Aug 2021 00:53:47 GMT):
well ain't that an exciting possibility
atoulme (Fri, 27 Aug 2021 00:54:05 GMT):
given how much time we spend signing stuff, that might help with performance ;)
atoulme (Fri, 27 Aug 2021 00:54:37 GMT):
@lucassaldanha I can't commit to getting it done _soon_ but I definitely could use a fix.
lucassaldanha (Fri, 27 Aug 2021 00:54:59 GMT):
I'll look at it. But I can't promise I'll finish before baby pops... :P
atoulme (Fri, 27 Aug 2021 00:55:09 GMT):
so I might make one, and I will post here if I start on it. Please do the same if you get to it first
lucassaldanha (Fri, 27 Aug 2021 00:55:13 GMT):
39.5 weeks and counting...
atoulme (Fri, 27 Aug 2021 00:55:13 GMT):
wait what now
atoulme (Fri, 27 Aug 2021 00:55:14 GMT):
again?
atoulme (Fri, 27 Aug 2021 00:55:22 GMT):
oh great man, congrats!
lucassaldanha (Fri, 27 Aug 2021 00:55:24 GMT):
:D
atoulme (Fri, 27 Aug 2021 00:56:02 GMT):
Thanks for sharing! Completely understand, I'll make sure to follow up if you don't have bandwidth.
atoulme (Fri, 27 Aug 2021 00:58:03 GMT):
fwiw this might explain why we lose peers
atoulme (Fri, 27 Aug 2021 00:58:12 GMT):
they might timeout on us while we're signing frames
shemnon (Fri, 27 Aug 2021 01:00:02 GMT):
So, there never was a native normalize path.
shemnon (Fri, 27 Aug 2021 01:00:39 GMT):
my optimization pass was based on over-the-wire blocks, never looked into p2p stuff, so that's why that method slipped through./
shemnon (Fri, 27 Aug 2021 01:01:20 GMT):
code should be there for a quick recover. And the native code is blinded per process, so it shouldn't drain entrpy
atoulme (Fri, 27 Aug 2021 01:03:58 GMT):
this is fun
atoulme (Fri, 27 Aug 2021 01:04:18 GMT):
I am not sure why I had this pop up so much this time. I think it's because I have 2 besu processes on the machine.
lucassaldanha (Fri, 27 Aug 2021 01:04:34 GMT):
The normalize is only used when signing in "non-native" mode, right?
lucassaldanha (Fri, 27 Aug 2021 01:06:46 GMT):
btw, maybe we move this discussion into a thread?
lucassaldanha (Fri, 27 Aug 2021 01:06:49 GMT):
[IMPORTANT] The head of main branch for repository hyperledger/besu has been rebased to deal with a DCO issue, please follow the instructions on Fixing DCO on main/release to fix your local repositories
https://wiki.hyperledger.org/pages/viewpage.action?pageId=24781754
atoulme (Fri, 27 Aug 2021 01:07:17 GMT):
here
lucassaldanha (Fri, 27 Aug 2021 01:08:24 GMT):
native-libs
lucassaldanha (Fri, 27 Aug 2021 01:08:41 GMT):
ops, didn't see it in time
atoulme (Fri, 27 Aug 2021 06:38:40 GMT):
Is there a native keccak library?
lucassaldanha (Fri, 27 Aug 2021 06:43:17 GMT):
Not yet... https://github.com/hyperledger/besu-native/issues/20
lucassaldanha (Fri, 27 Aug 2021 06:43:28 GMT):
But if you want to do something fun on the weekend...
lucassaldanha (Fri, 27 Aug 2021 06:43:31 GMT):
:)
atoulme (Fri, 27 Aug 2021 15:32:06 GMT):
Ok great. Yeah I can see now that’s the biggest draw for performance
atoulme (Sat, 28 Aug 2021 03:56:05 GMT):
can you generate key pairs natively?
atoulme (Sat, 28 Aug 2021 03:58:11 GMT):
it's pretty impressive how much slowness I get. Right now my machine is topped at 800% CPU because of the randomness starvation...
atoulme (Sat, 28 Aug 2021 03:59:30 GMT):
jstack2.txt
shemnon (Sat, 28 Aug 2021 12:48:41 GMT):
Maybe? I don't know. `calculateECDHKeyAgreement` is the target method and I am not sure how the bitcoin secp256k1 library supports that.
shemnon (Sat, 28 Aug 2021 12:50:43 GMT):
That's at least a full day deep dive I don't have in my time budget at the moment.
shemnon (Sat, 28 Aug 2021 12:58:11 GMT):
Keypair generation may also be a target. `calculateECDHKeyAgreement` appears 6x on your stack and generateKeyPair appears 3x, so both should be done but key agreement will provide the bigger improvement. There is a ECDH module but I didn't work too hard to make it work.
atoulme (Sat, 28 Aug 2021 15:48:51 GMT):
Yes of course. Danno I didn’t mean to imply you had to take it on. Sorry if you thought of it that way.
There is something wrong with this machine and it is not going to provide a good place to benchmark anything.
atoulme (Mon, 30 Aug 2021 06:09:12 GMT):
I had someone transcribe the ETC core dev call. https://hackmd.io/@w1g0/B1PAArNWY
atoulme (Mon, 30 Aug 2021 06:09:20 GMT):
I am going to follow up with issues for besu.
shemnon (Mon, 30 Aug 2021 15:02:49 GMT):
what were the besu issues?
shemnon (Mon, 30 Aug 2021 15:55:08 GMT):
@here contributor call in just over 23 hours - https://wiki.hyperledger.org/display/BESU/2021-08-31+Besu+Contributor+Call
grace.hartley (Mon, 30 Aug 2021 16:28:17 GMT):
User User_3 added by grace.hartley.
grace.hartley (Mon, 30 Aug 2021 16:29:00 GMT):
@andreolf1 FYI
andreolf1 (Mon, 30 Aug 2021 16:30:49 GMT):
great!
mdwright88 (Mon, 30 Aug 2021 20:24:40 GMT):
Thanks Danno! Please meet @andreolf1 who will be leading community around Quorum/Besu. We will both be on the call and look forward to updates.
atoulme (Mon, 30 Aug 2021 22:12:00 GMT):
https://github.com/hyperledger/besu/issues?q=is%3Aissue+is%3Aopen+label%3AETC
macfarla (Mon, 30 Aug 2021 23:17:51 GMT):
PR to ignore all unknown fields in eth_call and eth_estimateGas https://github.com/hyperledger/besu/pull/2690
shemnon (Mon, 30 Aug 2021 23:29:10 GMT):
Looks like a unit test flaked on that pr.
macfarla (Mon, 30 Aug 2021 23:39:54 GMT):
Thanks @shemnon - that's a flake I haven't seen before
macfarla (Tue, 31 Aug 2021 02:04:14 GMT):
For completeness - the issue for the flake https://github.com/hyperledger/besu/issues/2696
lucassaldanha (Tue, 31 Aug 2021 06:48:41 GMT):
It happened again on my PR!
VijayMichalik (Tue, 31 Aug 2021 08:30:54 GMT):
@GarySchulte was looking into this before as well
RoboCopsGoneMad (Tue, 31 Aug 2021 13:29:16 GMT):
do we have any documentation on the configuration subsystem? I'm not seeing anything in the wiki. Alternately, i'm just gonna find a PR that does it and follow that.
RoboCopsGoneMad (Tue, 31 Aug 2021 13:51:00 GMT):
for instance, why would a config option go in the unstable package as opposed to the stable one. how does one stabilize?
shemnon (Tue, 31 Aug 2021 14:12:42 GMT):
If a config option goes into the stable package it becomes an API and really cannot change it's semantics in less than 6 months time. If it's in unstable it can change every release. So options transition from unstable to stable when we think the option won't change in any appreciable way for at least a year and that it will be relevant for years.
shemnon (Tue, 31 Aug 2021 14:13:33 GMT):
It's based off of Java's patterns, except that their timeframe was years and a decade.
RoboCopsGoneMad (Tue, 31 Aug 2021 14:15:35 GMT):
i figured something like that from the X prefix pattern
shemnon (Tue, 31 Aug 2021 14:57:52 GMT):
contributor call in ~ 3 minutes - https://consensys.zoom.us/j/95691099798
shemnon (Tue, 31 Aug 2021 15:32:37 GMT):
Link to new matrix contributors chat room - https://matrix.to/#/#hyperledger-besu-contributors:chat.lfx.linuxfoundation.org
shemnon (Tue, 31 Aug 2021 15:33:18 GMT):
For reference: besu room link: https://matrix.to/#/#hyperledger-besu:chat.lfx.linuxfoundation.org
shemnon (Tue, 31 Aug 2021 15:39:04 GMT):
besu contributors matrix room link - https://chat.lfx.linuxfoundation.org/#/room/#hyperledger-besu-contributors:chat.lfx.linuxfoundation.org
VijayMichalik (Tue, 31 Aug 2021 15:39:13 GMT):
I drafted a potential release calendar for the next 12 months here, any issues with it?: https://docs.google.com/spreadsheets/d/1zNv9tUl8Fill6QX40IDSElPfFA-M0VrV0GaI57awXpQ/edit#gid=0
shemnon (Tue, 31 Aug 2021 15:39:30 GMT):
And general room - https://chat.lfx.linuxfoundation.org/#/room/#hyperledger-besu:chat.lfx.linuxfoundation.org
antonydenyer (Tue, 31 Aug 2021 15:58:39 GMT):
Bringing this one back from the dead ....
https://github.com/hyperledger/besu/pull/2704
shemnon (Tue, 31 Aug 2021 16:17:20 GMT):
Dates look reasonable
atoulme (Wed, 01 Sep 2021 15:54:11 GMT):
Would anyone here be interested in making Besu a clustered tech? Meaning we could have multiple Besu nodes with HA and possible specialization?
shemnon (Wed, 01 Sep 2021 15:58:01 GMT):
how would that differ from a microservices approach? What would the clustering aproach look like?
shemnon (Wed, 01 Sep 2021 15:58:01 GMT):
how would that differ from a microservices approach? What would the clustering approach look like?
atoulme (Wed, 01 Sep 2021 17:38:39 GMT):
it would not differ a whole lot. The clustering approach consists in using an existing technology, Infinispan, which can be configured to run as a single host or cluster with active-active participants.
atoulme (Wed, 01 Sep 2021 17:39:49 GMT):
Infinispan also has the immense advantage of wrapping up I/O access, and can write async and in batches if you're looking for the extra oomph when you sync. That's the stuff I've been testing and could not get clean numbers because my old local server has no entropy.
atoulme (Wed, 01 Sep 2021 17:41:05 GMT):
While I will buy newer hardware to test and get a real baseline number (sorry I really don't want to test this on cloud, I need a reproducible setup) I can tell you from anecdotal evidence that so far writes were just as fast with Infinispan involved. I didn't explore all the options yet.
Reads are even more interesting though.
atoulme (Wed, 01 Sep 2021 17:41:15 GMT):
I didn't get to test reads at this time.
atoulme (Wed, 01 Sep 2021 17:41:53 GMT):
Anyway, I have some crass changes in my fork under the infinispan branch, introducing a new storage engine with infinispan, right now configured to go to a rocksdb.
atoulme (Wed, 01 Sep 2021 17:42:38 GMT):
Infinispan can be a bit higher level and can cache java objects - the current interface wants key value stores to store and retrieve byte arrays, but infinispan can probably work at a higher level and deliver even more performance down the road.
atoulme (Wed, 01 Sep 2021 17:43:23 GMT):
During my testing, when entropy wasn't depleted, I noticed keccak256 was probably a good performance target at this time, and infinispan wasn't a perf bottleneck at this time when writing.
atoulme (Wed, 01 Sep 2021 17:43:54 GMT):
Anyway, I'm just sharing this as I go along. I am just trying to have fun with this :)
shemnon (Wed, 01 Sep 2021 17:47:35 GMT):
I think the real way to scope it out would be to share the branch we can see the scope of the change. PLuggable backend DBs have been a dream of mine for a while.
atoulme (Wed, 01 Sep 2021 17:47:51 GMT):
ok the branch is out there
atoulme (Wed, 01 Sep 2021 17:48:13 GMT):
https://github.com/atoulme/besu/tree/infinispan
atoulme (Wed, 01 Sep 2021 17:49:02 GMT):
here is the crucial place: https://github.com/hyperledger/besu/compare/main...atoulme:infinispan?diff=split&expand=1#diff-e80050d20ec59df3859d1214613525bca8f99ccc21f9523c8cf614c6d48fcc06R70
atoulme (Wed, 01 Sep 2021 17:49:17 GMT):
Now this turns into a very different exercise, because you trust Infinispan with persistence
atoulme (Wed, 01 Sep 2021 17:49:28 GMT):
and Infinispan has a loong list of persistence options and clustering mechanisms
atoulme (Wed, 01 Sep 2021 17:49:48 GMT):
and bear with me here. It can all be configured via XML. Which well isn't sexy at all. But it works.
atoulme (Wed, 01 Sep 2021 17:52:30 GMT):
so for example, if you can make it work for rocksdb, you could also choose to use a different store and swap them out https://infinispan.org/cache-store-implementations
atoulme (Wed, 01 Sep 2021 17:52:40 GMT):
they support JDBC, MongoDB, Redis...
atoulme (Wed, 01 Sep 2021 17:53:49 GMT):
the clustering happens over JGroups, which is super old now but just works. It's using UDP, there's discovery possible, you can do point to point replication, etc.
atoulme (Wed, 01 Sep 2021 17:54:13 GMT):
The best part is Infinispan being somewhat mature, there's an operator manual, and proper resources to make it work.
shemnon (Wed, 01 Sep 2021 17:56:41 GMT):
how does performance measure at mainnet scale?
atoulme (Wed, 01 Sep 2021 19:55:24 GMT):
I don’t know. I’d need a baseline. I could cheat - use an Aurora backend for example, get all nodes using it.
atoulme (Wed, 01 Sep 2021 19:56:01 GMT):
Syncing on mainnet would not be a good perf baseline. What would a good comparison use case look like?
shemnon (Wed, 01 Sep 2021 20:02:16 GMT):
Time to get to sync and time for each block import at mainnet are what I was hoping for to compare. Alternatives would be Goerli or an ETC chain.
shemnon (Wed, 01 Sep 2021 20:02:37 GMT):
But only mainnet has the sustained call volume to compare.
VijayMichalik (Wed, 01 Sep 2021 20:37:57 GMT):
@rjones just tagging you this to dig up a request from a couple of months back - can we get a besu-errorprone-checks repo to place some custom ones we've been using? per this PR https://github.com/hyperledger/besu/pull/2433 cc @usmansaleem
RoboCopsGoneMad (Wed, 01 Sep 2021 20:46:17 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=EbGoc9k6cZWA5Eu1K) would mainnet be a bad baseline because this is really a feature for enterprises?
atoulme (Wed, 01 Sep 2021 20:54:46 GMT):
I was trying to sync etc mainnet fwiw
atoulme (Wed, 01 Sep 2021 20:55:43 GMT):
@RoboCopsGoneMad yeah, not sure if it's a feature for enterprise only, and I have no business having an opinion there, but Erigon is targeting mainnet with their own approach, not enterprise.
atoulme (Wed, 01 Sep 2021 20:56:00 GMT):
And maybe that's where we should start. Could we ask the community what they think?
shemnon (Wed, 01 Sep 2021 20:59:11 GMT):
Strategically (may not be my place to opine) I think appealing to enterprise node operators is better than going to battle with raw node count. That's mutually assured destruction and not a "grow the pie" path. Focusing on a setup enterprises would want to run and not worrying about the "can it run on a RasPi4" crowd seems like a better tradeoff.
RoboCopsGoneMad (Wed, 01 Sep 2021 21:07:47 GMT):
I very much like the notion that "the database is a _detail_ "
RoboCopsGoneMad (Wed, 01 Sep 2021 21:08:28 GMT):
could also come in handy when dealing with state expiry
atoulme (Wed, 01 Sep 2021 21:24:54 GMT):
definitely not going for the RasPi4 discussion
atoulme (Wed, 01 Sep 2021 21:25:23 GMT):
jGroups AFAICT is not gentle. You don't do active/active replication so easy.
rjones (Wed, 01 Sep 2021 21:36:17 GMT):
@VijayMichalik https://github.com/hyperledger/besu-errorprone-checks
atoulme (Wed, 01 Sep 2021 23:39:45 GMT):
I'm getting one of these btw: https://www.crowdsupply.com/13-37/infinite-noise-trng
macfarla (Thu, 02 Sep 2021 00:47:31 GMT):
@usmansaleem
RoboCopsGoneMad (Thu, 02 Sep 2021 22:00:44 GMT):
greetings all! we're about to cut a besu release.
RoboCopsGoneMad (Thu, 02 Sep 2021 22:01:43 GMT):
please refrain from merging to main till further notice.
RoboCopsGoneMad (Thu, 02 Sep 2021 23:04:00 GMT):
It's now safe to merge into main for hyperledger/besu again.
alexandratran (Fri, 03 Sep 2021 17:45:56 GMT):
Hi everyone! I created a PR to add Roland Tyler (recent joinee of ConsenSys protocol docs team) as a besu docs maintainer: https://github.com/hyperledger/besu-docs/pull/785
Please vote! :white_check_mark:
VijayMichalik (Sun, 05 Sep 2021 18:05:16 GMT):
totally also your place to opine as a contributor and maintainer imo!
VijayMichalik (Sun, 05 Sep 2021 18:07:17 GMT):
at some point we expect the enterprises to be running on mainnet :slightly_smiling_face:
but in general I think creating a baseline is easier with a shorter total sync cycle. so picking a testnet probably makes sense
VijayMichalik (Sun, 05 Sep 2021 18:33:38 GMT):
Updated the release calendar here, if you notice anything amiss let me know! https://wiki.hyperledger.org/display/BESU/Future+Release+Dates
macfarla (Mon, 06 Sep 2021 00:10:29 GMT):
@rjones can we get a new team member added to the Hyperledger org - github handle `taccatisid` so they can be assigned tickets? Thanks
rjones (Tue, 07 Sep 2021 00:03:05 GMT):
@macfarla invite sent
macfarla (Tue, 07 Sep 2021 00:13:17 GMT):
thanks!
macfarla (Tue, 07 Sep 2021 00:13:22 GMT):
@taccatisid
taccatisid (Tue, 07 Sep 2021 01:27:12 GMT):
Has joined the channel.
macfarla (Wed, 08 Sep 2021 04:47:22 GMT):
I found an old link to bintray https://github.com/ConsenSys/PluginsAPIDemo/pull/1 - I've added a ticket to update the docs as well `https://github.com/hyperledger/besu-docs/issues/787`
VijayMichalik (Wed, 08 Sep 2021 16:13:50 GMT):
this ETC issue is currently tagged as P2 but it may even be P1 https://github.com/hyperledger/besu/issues/2715
VijayMichalik (Wed, 08 Sep 2021 16:14:06 GMT):
does anyone run ETC nodes?
diegoll (Wed, 08 Sep 2021 16:48:17 GMT):
Hi Vijay, we have several ETC nodes running, but I've never seen the described behavior
diegoll (Wed, 08 Sep 2021 16:48:17 GMT):
Hi Vijay, we have several ETC nodes running, let me check versions
diegoll (Wed, 08 Sep 2021 16:53:09 GMT):
most of our nodes are running versions older than v21.7.2
diegoll (Wed, 08 Sep 2021 16:59:17 GMT):
well, yes, I've upgraded one of our nodes to v21.7.3 and I can confirm the reported issue
diegoll (Wed, 08 Sep 2021 16:59:17 GMT):
well, yes, I've upgraded one of our nodes to v21.7.3 and I can confirm the reported issue also occurs on ETC Mainnet
grace.hartley (Wed, 08 Sep 2021 18:03:38 GMT):
Reposting this note. It looks like we're currently at 17 good first issues. It would be great to get 13 more. :)
_Hey folks! Hyperledger staff has asked the Hyperledger Besu team to participate in Grace Hopper's Open Source Day on October 1st. It is a great opportunity to showcase Hyperledger Besu and get new developers working on our codebase. To prep for this event, Besu needs to have 30+ good first issues labeled in the project. Besu currently has 14 good first issues opened. Can this group work together to add any additional good first issues in the coming weeks? We need the 30 to be uploaded by September 15th. Thanks!_
atoulme (Wed, 08 Sep 2021 19:49:44 GMT):
https://github.com/hyperledger/besu/pull/2634#discussion_r704678780 needs clarification
GarySchulte (Wed, 08 Sep 2021 20:38:12 GMT):
https://matrix.to/#/!CymbmJVrjkuEPrlyaC:chat.lfx.linuxfoundation.org/$ZK-7WioDwqgJvXmHj2b0Cnqo8ilzfDmDpdznEu0-_4Q?via=chat.lfx.linuxfoundation.org&via=matrix.org
GarySchulte (Wed, 08 Sep 2021 20:38:53 GMT):
link to a conversation happening on the new besu-contributors matrix chat regarding modularizing block proposals, in support of MEV strategies
GarySchulte (Wed, 08 Sep 2021 20:41:22 GMT):
TL/DR, the besu mainnet team at Consensys is looking at supporting (at least) flashbots post-merge, and we are starting to lay the architectural work for supporting multiple block proposal strategies. Please weigh in with thoughts, ideas, criticisms. We want to engage early since it will be a large change and possibly politically charged.
atoulme (Wed, 08 Sep 2021 22:02:04 GMT):
ok
atoulme (Thu, 09 Sep 2021 00:28:39 GMT):
Screen Shot 2021-09-08 at 5.28.06 PM.png
atoulme (Thu, 09 Sep 2021 00:28:40 GMT):
@GarySchulte I can't load the conversation:
atoulme (Thu, 09 Sep 2021 00:30:32 GMT):
@rjones are matrix chat convos backed up somewhere?
macfarla (Thu, 09 Sep 2021 00:51:43 GMT):
I was able to load the conversation just now - maybe try again?
atoulme (Thu, 09 Sep 2021 00:53:24 GMT):
ah, worked in incognito
atoulme (Thu, 09 Sep 2021 00:53:32 GMT):
thank you
jan-law (Thu, 09 Sep 2021 19:11:32 GMT):
Has joined the channel.
diegoll (Thu, 09 Sep 2021 20:21:39 GMT):
Hello guys, I just created https://github.com/hyperledger/besu/pull/2746 for https://github.com/hyperledger/besu/issues/2715. I let it as draft b/c I would like to full sync an ETC node just to be sure. Btw, are the specs tested somehow?
atoulme (Thu, 09 Sep 2021 21:23:47 GMT):
Ah, good one @diegoll and no not aware that we do testing for ETC using this approach.
atoulme (Thu, 09 Sep 2021 21:24:12 GMT):
Definitely could use more testing
atoulme (Thu, 09 Sep 2021 21:29:11 GMT):
I approved the tests to run
atoulme (Thu, 09 Sep 2021 21:51:48 GMT):
@diegoll give it a push to get circleci to comply
diegoll (Thu, 09 Sep 2021 22:28:58 GMT):
done
atoulme (Thu, 09 Sep 2021 22:42:51 GMT):
not sure why CircleCI is not running
knagware9 (Sun, 12 Sep 2021 11:11:13 GMT):
Has joined the channel.
shemnon (Mon, 13 Sep 2021 01:06:35 GMT):
Contributor call in ~ 24 hours - add and remove agenda items here - https://wiki.hyperledger.org/display/BESU/2021-09-13+Besu+Contributor+Call
macfarla (Tue, 14 Sep 2021 02:51:34 GMT):
Trying to think of "good first issues" - what about being able to validate a besu config file (ie without actually starting besu)
macfarla (Tue, 14 Sep 2021 04:22:55 GMT):
https://github.com/hyperledger/besu/issues/2759
macfarla (Tue, 14 Sep 2021 06:53:54 GMT):
I'm looking at this https://wiki.hyperledger.org/display/BESU/First+contribution and it seems to skip the first step where you see an issue that you want to work on and get it assigned to you. Have we documented this process anywhere? Is it just "reach out on RocketChat"?
VijayMichalik (Tue, 14 Sep 2021 08:40:26 GMT):
That's a great idea to add!
shemnon (Tue, 14 Sep 2021 17:58:06 GMT):
Changelog looks a little sparse. Are there anythings we need to document for the next release? Maybe the ETC change to their protocol spec? Anything else?
atoulme (Tue, 14 Sep 2021 18:21:19 GMT):
the ETC change is not merged, is it?
shemnon (Tue, 14 Sep 2021 18:47:30 GMT):
It should be before we do the build. I consider it blocking for release.
macfarla (Tue, 14 Sep 2021 22:37:43 GMT):
Is there a reason we have `-V` rather than `-v` as the alias for `--version` in besu? All other CLI options are lower case
shemnon (Tue, 14 Sep 2021 22:48:38 GMT):
`-v` usually means verbose
shemnon (Tue, 14 Sep 2021 22:49:14 GMT):
And usually there is no short option for version-then-quit
macfarla (Tue, 14 Sep 2021 22:52:29 GMT):
true. but `node -v` and `yarn -v` give version info
macfarla (Tue, 14 Sep 2021 22:52:53 GMT):
I'm trying random things I have installed
VijayMichalik (Wed, 15 Sep 2021 10:46:28 GMT):
https://docs.github.com/en/communities/using-templates-to-encourage-useful-issues-and-pull-requests/configuring-issue-templates-for-your-repository these new yml issue templates are nice
VijayMichalik (Wed, 15 Sep 2021 10:46:38 GMT):
maybe we should plan to use them?
grace.hartley (Wed, 15 Sep 2021 13:04:17 GMT):
Thanks to everyone who submitted a good first issue! We hit out goal and now have 33 issues identified! :woo:
shemnon (Wed, 15 Sep 2021 21:30:45 GMT):
Looks like changelog entries were being added to the wrong release. PR to fix - https://github.com/hyperledger/besu/pull/2770
macfarla (Wed, 15 Sep 2021 23:58:57 GMT):
While we're tidying up the changelog - I moved the two "experimental" features under the "Early access" heading https://github.com/hyperledger/besu/pull/2771
alexandratran (Sun, 19 Sep 2021 04:38:10 GMT):
Roland has reached 14 approvals and 2 weeks since opening this PR. Can someone add @rolandtyler to the Hyperledger github organization please?
VijayMichalik (Mon, 20 Sep 2021 10:26:05 GMT):
@rjones please could you add Roland? thanks!
rjones (Mon, 20 Sep 2021 15:00:31 GMT):
invite semt
VijayMichalik (Mon, 20 Sep 2021 15:01:47 GMT):
thanks!
grace.hartley (Wed, 22 Sep 2021 12:29:37 GMT):
Sorry ya'll. Realized I'm delinquent on the Besu quarterly report. Will try to get to it today. cc @shemnon
grace.hartley (Wed, 22 Sep 2021 16:28:02 GMT):
Hey folks - Here is the draft quarterly report to be submitted to the TSC. Let me know if you have any feedback/additions. Otherwise, will share with them later today. https://wiki.hyperledger.org/pages/resumedraft.action?draftId=58855126&draftShareId=44fa6082-4b47-4e28-8bfe-39434cd607ed&src=shareui&src.shareui.timestamp=1632328040922 cc @shemnon @VijayMichalik
shemnon (Wed, 22 Sep 2021 16:45:47 GMT):
I added some bits about what I'm doing at Hedera, and upped the maintainer diversity company count.
grace.hartley (Wed, 22 Sep 2021 16:46:02 GMT):
Thank you very much!
diegoll (Thu, 23 Sep 2021 20:57:57 GMT):
Hello team, I just took one of the oldest `good-first-issue`s and got [this PR](https://github.com/hyperledger/besu/pull/2801). Review is appreciated
atoulme (Thu, 23 Sep 2021 21:23:05 GMT):
looks really good, ventured one comment
VijayMichalik (Mon, 27 Sep 2021 07:49:44 GMT):
Hey @rjones, can you add siladu (simon dudley) to the hyperledger org?
rjones (Mon, 27 Sep 2021 16:23:09 GMT):
done
macfarla (Mon, 27 Sep 2021 23:17:36 GMT):
Refactor (rename) PR for review - it's 74 files and I've shared the sed commands I used - https://github.com/hyperledger/besu/pull/2806
macfarla (Mon, 27 Sep 2021 23:34:29 GMT):
Some Besu source files have ` * Copyright Hyperledger Contributors.` or `* Copyright contributors to Hyperledger Besu` but most have `* Copyright ConsenSys AG.` - is one correct or is it ok to have a mix?
shemnon (Tue, 28 Sep 2021 00:15:00 GMT):
All new file should have `* Copyright contributors to Hyperledger Besu ` https://wiki.hyperledger.org/display/TSC/Copyright+and+License+Policy
shemnon (Tue, 28 Sep 2021 00:15:33 GMT):
Nobody is checking or enforcing it, what is important is a valid copyright claim and the apache license
andreolf1 (Tue, 28 Sep 2021 14:17:22 GMT):
Hey folks - Here is the agenda for our contribution call later: https://wiki.hyperledger.org/display/BESU/2021-09-28+Besu+Contributor+Call
macfarla (Wed, 29 Sep 2021 00:17:15 GMT):
Another rename PR for review - this time "off chain" -> offchain - much smaller, only 13 files - https://github.com/hyperledger/besu/pull/2818
rjones (Thu, 30 Sep 2021 13:37:14 GMT):
thoughts on marking this channel read-only with a pointer to the other chat system?
shemnon (Thu, 30 Sep 2021 13:51:17 GMT):
We need the APAC contributors over on matrix first. In 2 weeks there is an APAC contributor call, I think after that we can label it as read only.
shemnon (Thu, 30 Sep 2021 14:15:18 GMT):
From the quarterly report, a request for more info on QBFT - https://wiki.hyperledger.org/display/TSC/2021+Q3+Hyperledger+Besu?focusedCommentId=58856392#comment-58856392
macfarla (Thu, 30 Sep 2021 20:53:52 GMT):
Updated the "first contribution" page https://wiki.hyperledger.org/display/BESU/First+contribution
RoboCopsGoneMad (Fri, 01 Oct 2021 22:56:23 GMT):
that came in handy today @macfarla thanks!
macfarla (Tue, 05 Oct 2021 01:20:49 GMT):
exciting to see lots of good first issues assigned to folks! is that from the open source day?
macfarla (Tue, 05 Oct 2021 03:11:09 GMT):
besu contributors matrix room link - https://chat.lfx.linuxfoundation.org/#/room/#hyperledger-besu-contributors:chat.lfx.linuxfoundation.org
alexandratran (Mon, 11 Oct 2021 16:26:56 GMT):
Hi, can someone review this small docs PR? https://github.com/hyperledger/besu-docs/pull/820
macfarla (Mon, 11 Oct 2021 23:52:51 GMT):
gm
macfarla (Mon, 11 Oct 2021 23:52:57 GMT):
FYI Preparing to cut 21.10.0-RC2 release in 6hrs!
macfarla (Tue, 12 Oct 2021 06:15:52 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to 21.10.0-RC1-next until further notice
macfarla (Mon, 18 Oct 2021 00:40:14 GMT):
Apologies for not coming back with the completion message here for RC2... RC2 got done but is the same as RC1.
macfarla (Mon, 18 Oct 2021 00:40:23 GMT):
In other news - 21.10.0-RC3 release is complete:
https://github.com/hyperledger/besu/releases/tag/21.10.0-RC3
https://github.com/hyperledger/besu-docs/releases/tag/21.10.0-RC3
andreolf1 (Mon, 25 Oct 2021 10:50:00 GMT):
Hello team, added the agenda items for our contributor call tomorrow: https://wiki.hyperledger.org/display/BESU/2021-10-26+Besu+Contributor+Call
Feel free to add your topics in the agenda that you want to discuss. Looking forward!
macfarla (Tue, 26 Oct 2021 01:41:08 GMT):
Hi @andreolf1 this channel has _mostly_ moved over here -> you should be able to join with this link -> besu contributors matrix room link - https://chat.lfx.linuxfoundation.org/#/room/#hyperledger-besu-contributors:chat.lfx.linuxfoundation.org
macfarla (Tue, 26 Oct 2021 01:41:08 GMT):
Hi @andreolf1 this channel has _mostly_ moved over here -> you should be able to join with this link -> besu contributors matrix room link - https://chat.lfx.linuxfoundation.org/#/room/#hyperledger-besu-contributors:chat.lfx.linuxfoundation.org
macfarla (Tue, 26 Oct 2021 01:41:08 GMT):
Hi @andreolf1 this channel has _mostly_ moved over here -> you should be able to join with this link -> besu contributors matrix room link - https://chat.lfx.linuxfoundation.org/#/room/#hyperledger-besu-contributors:chat.lfx.linuxfoundation.org
rjones (Tue, 26 Oct 2021 02:35:39 GMT):
andreolf1 (Tue, 26 Oct 2021 10:37:22 GMT):
ok cool thanks @rjones
andreolf1 (Tue, 26 Oct 2021 10:37:22 GMT):
ok cool thanks @rjones and @macfarla for the heads up. will post it there
GarySchulte (Thu, 04 Nov 2021 19:44:50 GMT):
:wave:
rjones (Thu, 04 Nov 2021 19:45:16 GMT):
Howdy!
shemnon (Thu, 04 Nov 2021 19:47:07 GMT):
Umm... about that... :stuck_out_tongue_winking_eye:
macfarla (Thu, 04 Nov 2021 20:29:47 GMT):
gm
macfarla (Fri, 05 Nov 2021 01:36:31 GMT):
re the CI usage - we are investigating options. Looks like we might have to get rid of containerTests (third-party lib) which we're currently using to run Besu <> Tessera ATs - we've tried various things to make it more efficient but have run into dead ends
macfarla (Fri, 05 Nov 2021 01:38:53 GMT):
CI restricted context - community PRs have been getting "unauthorized" on unitTest step - I think it's the sonarCloud context. One way around this (this is what we do for the publishDocker step) is to run it only on main branch - but then we wouldn't get the report per PR. Thoughts?
shemnon (Fri, 05 Nov 2021 03:13:53 GMT):
We have issues with sonarcloud not running for non-writers ate Hedera. Since SonarCloud is not gating it won't impact the approval. However some of the issues are nice to see mid-process.
shemnon (Fri, 05 Nov 2021 03:13:53 GMT):
We have issues with sonarcloud not running for non-writers at Hedera. Since SonarCloud is not gating it won't impact the approval. However some of the issues are nice to see mid-process.
lucassaldanha (Fri, 05 Nov 2021 03:33:48 GMT):
lol
macfarla (Fri, 05 Nov 2021 05:04:24 GMT):
maybe it would work to just make it a separate step then? right now it's part of the unitTests step which is required
macfarla (Fri, 05 Nov 2021 05:05:10 GMT):
:wave:
macfarla (Fri, 05 Nov 2021 06:59:37 GMT):
LFR https://github.com/hyperledger/besu/pull/2994
diegoll (Fri, 05 Nov 2021 14:01:36 GMT):
LFR as well https://github.com/hyperledger/besu/pull/2944
diegoll (Fri, 05 Nov 2021 14:01:36 GMT):
LFR as well https://github.com/hyperledger/besu/pull/2944 (idk why the _unitTests_ haven't reported status)
diegoll (Fri, 05 Nov 2021 14:01:36 GMT):
LFR as well https://github.com/hyperledger/besu/pull/2944 (idk why the _unitTests_ hasn't reported status)
macfarla (Sat, 06 Nov 2021 09:22:33 GMT):
it's because of the sonarcloud restricted context
macfarla (Mon, 08 Nov 2021 02:00:38 GMT):
Proposal to add @taccatisid as a Besu maintainer https://github.com/hyperledger/besu/pull/3015
macfarla (Mon, 08 Nov 2021 03:42:45 GMT):
LFR - sonar as a separate step https://github.com/hyperledger/besu/pull/3016
RoboCopsGoneMad (Mon, 08 Nov 2021 13:55:46 GMT):
I had issues in the past with trying it as a separate step, when remounting the project, the previous build wasn't there because it had run in a different executor. You might want to test that this generates test coverage correctly.
GarySchulte (Mon, 08 Nov 2021 21:34:10 GMT):
https://github.com/hyperledger/besu/pull/2985
GarySchulte (Mon, 08 Nov 2021 21:34:47 GMT):
PR to vote for Daniel Lehrner as a contributor
macfarla (Tue, 09 Nov 2021 01:00:40 GMT):
PR to add Taccat Isid as a maintainer: https://github.com/hyperledger/besu/pull/3022
macfarla (Tue, 09 Nov 2021 01:33:51 GMT):
ok will investigate!
macfarla (Tue, 09 Nov 2021 01:56:59 GMT):
@here - We're beginning the release process for Hyperledger/besu, please don't merge anything to main until further notice
macfarla (Tue, 09 Nov 2021 03:16:36 GMT):
Release process is finished! https://github.com/hyperledger/besu/releases/tag/21.10.1
RoboCopsGoneMad (Tue, 09 Nov 2021 03:39:04 GMT):
:tada:
RoboCopsGoneMad (Tue, 09 Nov 2021 14:20:08 GMT):
gm besu
shemnon (Tue, 09 Nov 2021 14:46:44 GMT):
homebrew wasn't updated.
RoboCopsGoneMad (Tue, 09 Nov 2021 15:57:32 GMT):
has anyone tried to upgrade us to latest version of vert.x? we are on 3.9.8 and latest is 4.2.1.
rjones (Tue, 09 Nov 2021 16:13:44 GMT):
I'd be interested in homebrew usage stats. I know Fabric dropped support when the number of downloads was single digits per quarter
shemnon (Tue, 09 Nov 2021 16:35:07 GMT):
I don't think we have a way to distinguish from besu and github downloads.
shemnon (Tue, 09 Nov 2021 16:38:25 GMT):
Judging by repo clones (not sure if that is first time users or upgraders) we are well above that. 35 unique cloners/month right now.
shemnon (Tue, 09 Nov 2021 17:04:07 GMT):
I also think the overhead of maintaining besu's homebrew is much lower than fabric's.
shemnon (Tue, 09 Nov 2021 17:09:44 GMT):
PR for homebrew (warning: auto-merge) https://github.com/hyperledger/homebrew-besu/pull/62
rjones (Tue, 09 Nov 2021 17:15:17 GMT):
Feels like that might be an `action` in the homebrew repo that fires on a tag
shemnon (Tue, 09 Nov 2021 17:15:30 GMT):
Yep, would be handy.
diegoll (Tue, 09 Nov 2021 20:56:03 GMT):
hi all, I think this can be closed https://github.com/hyperledger/besu/issues/1733, right?
RoboCopsGoneMad (Tue, 09 Nov 2021 21:04:02 GMT):
upgrading to Vert.x 4 is gonna be non-trivial. i'm still gonna go for it, but if this spins out of control anymore I may ask for either a buddy to help, or to upgrade a submodule at a time incrementally, to avoid a long lived migration branch.
diegoll (Tue, 09 Nov 2021 21:07:56 GMT):
you can count me in
shemnon (Tue, 09 Nov 2021 21:17:28 GMT):
Sounds good given that vertx4 is a how and when, not an if. Security will bite us one day if we don't.
macfarla (Wed, 10 Nov 2021 01:21:20 GMT):
Will clarify this in the release docs!
RoboCopsGoneMad (Wed, 10 Nov 2021 01:43:06 GMT):
thanks gents, i'm diving in!
macfarla (Wed, 10 Nov 2021 19:35:52 GMT):
Contributor call - I've added the host recording to the wiki page - it starts at the Discord discussion. So @shemnon it probably is worth adding your recording to catch the first half https://wiki.hyperledger.org/display/BESU/2021-11-09+Besu+Contributor+Call
macfarla (Wed, 10 Nov 2021 19:37:01 GMT):
Also from the contributor call - can we get access to the test Discord @rjones - do you want more people to play around in there?
shemnon (Wed, 10 Nov 2021 19:42:50 GMT):
Test discord is here - https://discord.gg/EmZEFWYS - access is wide open at the moment.
macfarla (Wed, 10 Nov 2021 21:11:51 GMT):
I get "invite invalid"
shemnon (Wed, 10 Nov 2021 21:14:01 GMT):
Try this - the link may have had a 7 day expiration - https://discord.gg/UmCGEret
macfarla (Wed, 10 Nov 2021 21:43:36 GMT):
yep that worked
macfarla (Wed, 10 Nov 2021 21:43:39 GMT):
thanks!
macfarla (Wed, 10 Nov 2021 21:43:58 GMT):
LFR https://github.com/hyperledger/besu/pull/3031
GarySchulte (Wed, 10 Nov 2021 21:47:49 GMT):
I am seeing a different sha for the linked zip
daniellehrner (Wed, 10 Nov 2021 22:06:53 GMT):
Has joined the channel.
daniellehrner (Wed, 10 Nov 2021 22:06:54 GMT):
We are currently debugging some peering issues with Besu. In the process we realized that we don't have an easy way to recover the [ENR (Ethereum Node Record)](https://eips.ethereum.org/EIPS/eip-778) of a Besu node. It can only be recovered in the log output, right after the start.
We were thinking to return the ENR via the API. Our first thought was to add it to the existing `admin_nodeInfo`. But other clients, like geth, have the same endpoint and existing tools could have troubles with the response, if there is an unexpected field in it.
So we were thinking about a new endpoint called `admin_enrInfo`, to use a naming pattern similar to node info.
What do you think? Any feedback is welcome.
macfarla (Wed, 10 Nov 2021 22:20:21 GMT):
@GarySchulte I did this:
macfarla (Wed, 10 Nov 2021 22:20:25 GMT):
➜ Downloads shasum -a 256 ~/Downloads/besu-21.10.1.zip
877148a1a13a1065151cf8fb3fbb56ea1d01e3a942fa7ebd6bd4ea4d15101fee /Users/sallymacfarlane/Downloads/besu-21.10.1.zip
macfarla (Wed, 10 Nov 2021 22:20:47 GMT):
is that wrong?
macfarla (Wed, 10 Nov 2021 22:24:40 GMT):
right. I had downloaded the native zip
macfarla (Wed, 10 Nov 2021 22:24:47 GMT):
pushed a fix
macfarla (Wed, 10 Nov 2021 23:55:55 GMT):
LFR take two: https://github.com/hyperledger/besu/pull/3031
macfarla (Thu, 11 Nov 2021 02:05:01 GMT):
Reminder we have 2 proposed maintainers open for voting: https://github.com/hyperledger/besu/pull/3022 and
https://github.com/hyperledger/besu/pull/2985
macfarla (Thu, 11 Nov 2021 03:55:46 GMT):
Moving folks to Emeritus maintainers after at least a quarter of inactivity - do we normally reach out to them before moving them over? Is there a procedure?
shemnon (Thu, 11 Nov 2021 04:01:58 GMT):
There is no procedure. It should be done more regularly. Last year I published a proposed list and at-tagged those being moved. Only two requested a reprieve that I granted. I think the same thing can be done, post a provisional list and then wait a couple of weeks for reactions.
macfarla (Thu, 11 Nov 2021 04:23:23 GMT):
I found your last PR and it had 28 Sep being the start of a quarter - where are the quarters defined? I would have gone with Q3 from 1 Jul - end Oct ?
macfarla (Thu, 11 Nov 2021 04:23:23 GMT):
@shemnon I found your last PR and it had 28 Sep being the start of a quarter - where are the quarters defined? I would have gone with Q3 from 1 Jul - end Oct ?
macfarla (Thu, 11 Nov 2021 04:23:23 GMT):
@shemnon I found your last PR and it had 28 Sep being the start of a quarter - where are the quarters defined? I would have gone with Q3 from 1 Jul - end Oct ? https://github.com/hyperledger/besu/pull/2048
shemnon (Thu, 11 Nov 2021 04:28:28 GMT):
I went by tsc quarterly report dates. Pick any old 3 month stretch. Perhaps be kind and spread it to 6 months?
macfarla (Thu, 11 Nov 2021 04:33:40 GMT):
https://github.com/hyperledger/besu/pull/3037 - it's not _quite_ 6 months but close
macfarla (Thu, 11 Nov 2021 06:58:27 GMT):
LFR https://github.com/hyperledger/besu/pull/3042 this was annoying me today
fab-10 (Thu, 11 Nov 2021 15:12:13 GMT):
Has joined the channel.
RoboCopsGoneMad (Thu, 11 Nov 2021 16:19:12 GMT):
yes it was. haven't seen much retention of their interest though...
RoboCopsGoneMad (Thu, 11 Nov 2021 16:20:01 GMT):
same here
shemnon (Thu, 11 Nov 2021 17:04:33 GMT):
That one expired. This one expires next week - https://discord.gg/UmCGEret
shemnon (Thu, 11 Nov 2021 17:05:10 GMT):
There's a TSC task force looking into chat platforms. Once I see positive signal there will be a permanent link.
grace.hartley (Thu, 11 Nov 2021 19:41:34 GMT):
Hey all - Let's add this to the agenda for next contributor call. I'd love us to evaluate how the Besu community is growing. https://insights.lfx.linuxfoundation.org/projects/hyperledger%2Fbesu/dashboard;quicktime=time_filter_3Y
GarySchulte (Thu, 11 Nov 2021 23:12:51 GMT):
Guillaume is seeking comments on these gas cost changes, in the fork ahead of the verkle trie fork. See especially the `Implementation notes` section:
https://notes.ethereum.org/-fJSOrnYQl-mqoWKpaTIsQ
GarySchulte (Fri, 12 Nov 2021 00:24:38 GMT):
I *think* this is advocating for migrating epoch 0 data into a verkle trie to avoid DoS vector of the gas cost changes. I asked him for clarification since this is still mentioning a frozen MPT for epoch 0: https://notes.ethereum.org/5HDhQXstTaKtVqVbS7S9yw
GarySchulte (Fri, 12 Nov 2021 00:24:38 GMT):
I *think* this is advocating for migrating epoch 0 data into a verkle trie to avoid DoS vector of the gas cost changes. I asked him for clarification since his verkle EIP draft is still mentioning a frozen MPT for epoch 0: https://notes.ethereum.org/5HDhQXstTaKtVqVbS7S9yw
shemnon (Fri, 12 Nov 2021 00:46:47 GMT):
I'm somewhat skeptical the epoch based addresses will be a thing. EIP-1559 was hard enough on the downstream tools and nearly blew up in a bad way. This has a bigger impact (throw in the solidity complier to the mix and 20 byte assumptions) and the coordination costs go up exponentially. Just my opinion however.
shemnon (Fri, 12 Nov 2021 00:47:53 GMT):
I also have this nagging thought in the back of my head that verkle is the next casper. Lots of hype but the first proofs of concept were regressions performance wise. The quicker a full-scale proof of concept happens the better.
macfarla (Fri, 12 Nov 2021 03:11:47 GMT):
We have an absolute majority vote on https://github.com/hyperledger/besu/pull/2985 and https://github.com/hyperledger/besu/pull/3022 so I'll aim to get them merged (and added to the github group) today!
macfarla (Fri, 12 Nov 2021 05:59:17 GMT):
done and done. Congrats @daniellehrner and @taccatisid :tada:
jagpreet (Fri, 12 Nov 2021 08:35:02 GMT):
Has left the channel.
JiriPeinlich (Mon, 15 Nov 2021 13:15:44 GMT):
Has joined the channel.
JiriPeinlich (Mon, 15 Nov 2021 13:15:45 GMT):
Hello. Sorry for bothering you. I started working on Besu in Consensys and had to raise an issue on incubator-tuweni with the dns-discovery module. I also raised a PR that would resolve those issues for me (us) (https://github.com/apache/incubator-tuweni/issues/345). Would someone mind looking at it, or letting me know who might have a look at it? I heard that Mr. @atoulme could help me.
RoboCopsGoneMad (Mon, 15 Nov 2021 14:18:47 GMT):
gm everyone
shemnon (Mon, 15 Nov 2021 19:13:31 GMT):
@here We're beginning the release process for hyperledger/besu, please don't merge anything to main until further notice
shemnon (Mon, 15 Nov 2021 19:16:10 GMT):
PR for 21.10.2 - https://github.com/hyperledger/besu/pull/3066
shemnon (Mon, 15 Nov 2021 20:34:11 GMT):
pr for 21.10.3-SNAPSHOT - https://github.com/hyperledger/besu/pull/3067
shemnon (Mon, 15 Nov 2021 20:36:56 GMT):
PR for homebrew - https://github.com/hyperledger/homebrew-besu/pull/63
lucassaldanha (Mon, 15 Nov 2021 21:26:58 GMT):
Proposal to add Simon Dudley as a Besu maintainer: https://github.com/hyperledger/besu/pull/3068
atoulme (Mon, 15 Nov 2021 21:31:45 GMT):
Yeah I’ll look tonight
shemnon (Mon, 15 Nov 2021 21:32:15 GMT):
@here It's now safe to merge into hyperledger/besu for main again.
atoulme (Mon, 15 Nov 2021 21:38:30 GMT):
Here ya go, fix the pr please
siladu (Tue, 16 Nov 2021 04:33:53 GMT):
Has joined the channel.
JiriPeinlich (Tue, 16 Nov 2021 07:30:27 GMT):
Thank you
fab-10 (Tue, 16 Nov 2021 11:50:49 GMT):
Hi all, I need the review of a gradle expert, for a inefficiency I found on the downloadLicenses task.
Basically I am not able to build my draft PR because the assemble step is always failing due to timeout https://app.circleci.com/pipelines/github/hyperledger/besu?branch=pull%2F3065
and after some debugging I found that the license check is the culprit, and I think the current implementation of the plugin is inefficient, specially for project with many subproject, like Besu, and I submitted a possible fix here https://github.com/hierynomus/license-gradle-plugin/pull/203.
In the meantime we wait for a new release of the plugin, is there any workaround we can use to make the build work?
fab-10 (Tue, 16 Nov 2021 11:57:47 GMT):
Worth to mention that with this fix the build time should decreases of at least 9 minutes
RoboCopsGoneMad (Tue, 16 Nov 2021 16:22:05 GMT):
did something change recently to put us over the timeout?
RoboCopsGoneMad (Tue, 16 Nov 2021 16:23:03 GMT):
there are plenty of previous runs that take 12-13 mins, and still complete... great find though.
fab-10 (Tue, 16 Nov 2021 16:28:28 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=kvGLx8pScj8pPXAfs) I think that since I have added datatypes as dependency in the config subproject, that has raised the build time over the timeout of 10min, that indeed is already very high for this task
JiriPeinlich (Tue, 16 Nov 2021 18:00:26 GMT):
I am wondering, what is the procedure with releasing the library
atoulme (Tue, 16 Nov 2021 22:20:01 GMT):
go ask that at users@tuweni.apache.org
macfarla (Wed, 17 Nov 2021 02:47:07 GMT):
Besu maintainers Admins - Looking at this list https://github.com/orgs/hyperledger/teams/besu-maintainers/members we are a bit light on Admins ie those with the Maintainer label. AFAIK the process for giving these privileges is ad hoc. I propose adding @gary.schulte as another person in the US timezone
macfarla (Wed, 17 Nov 2021 02:47:07 GMT):
Besu maintainers Admins - Looking at this list https://github.com/orgs/hyperledger/teams/besu-maintainers/members we are a bit light on Admins ie those with the Maintainer label. AFAIK the process for giving these privileges is ad hoc. I propose adding @GarySchulte as another person in the US timezone
RoboCopsGoneMad (Wed, 17 Nov 2021 05:33:56 GMT):
seconded
RoboCopsGoneMad (Wed, 17 Nov 2021 14:41:43 GMT):
gm besunians
shemnon (Wed, 17 Nov 2021 15:17:54 GMT):
There are 2 maintainers in the US timezone that have that access, me and Meredith, including two in AUS timezones, Adrian and Josh. Is there a specific example where having limited access to critical keys has been a problem?
shemnon (Wed, 17 Nov 2021 15:18:19 GMT):
I would be more inclined to trade one-for-one rather than add.
RoboCopsGoneMad (Wed, 17 Nov 2021 20:20:52 GMT):
Vert.x upgrade wasn't _that_ bad
RoboCopsGoneMad (Wed, 17 Nov 2021 20:20:55 GMT):
https://github.com/hyperledger/besu/pull/3073
RoboCopsGoneMad (Wed, 17 Nov 2021 21:11:25 GMT):
i think this may have just drafted me into the quest to remove Orion from acceptance tests too
RoboCopsGoneMad (Wed, 17 Nov 2021 21:13:53 GMT):
Although a much bigger problem may be Tessera's dependence on testcontainers. If we go through with this swap of the OrionTestHarness for TesseraTestHarness, we now require a docker stack at unit test time, which seems like.... not what unit test means.
RoboCopsGoneMad (Wed, 17 Nov 2021 21:20:26 GMT):
is there any reason we shouldn't use a Mock enclave from PrivacyReorgTest?
shemnon (Wed, 17 Nov 2021 21:35:36 GMT):
Docker tests at unit test time gets a soft no from me. If we tie it in with a junit `assume` check for local dev and make sure it runs at full speed at checkin that would satisfy me.
RoboCopsGoneMad (Wed, 17 Nov 2021 21:55:05 GMT):
even if we did that we'd have to make a docker stack available on the ci executor used for unit test
RoboCopsGoneMad (Wed, 17 Nov 2021 21:55:18 GMT):
am trying to mock it now. it's not going great.
macfarla (Wed, 17 Nov 2021 22:44:46 GMT):
Revenant is working on getting rid of testcontainers rn https://github.com/hyperledger/besu/issues/2867 - it requires some changes to Tessera so it can be run as a process
RoboCopsGoneMad (Wed, 17 Nov 2021 22:45:32 GMT):
music to my ears!
macfarla (Wed, 17 Nov 2021 22:49:51 GMT):
@mbaxter tagging you in here - I know you're busy with your current project - you're currently one of 4 members with Admin access to the besu-maintainers github group (can add/remove members). What are your thoughts?
RoboCopsGoneMad (Wed, 17 Nov 2021 22:51:59 GMT):
I think PrivacyReorgTest is pretty problematic and likely warrants refactor. Maybe i just don't understand the role of the Enclave during re-orgs though... It was very difficult to mock out and I don't really grok why
RoboCopsGoneMad (Wed, 17 Nov 2021 22:52:20 GMT):
what about the acceptance tests that depend on Orion? Can I remove those?
RoboCopsGoneMad (Wed, 17 Nov 2021 22:52:59 GMT):
Sorry if I sound like I'm in a rush, i'm just eager to avoid the vert.x upgrade becoming a long-lived branch that requires constant rebasing
macfarla (Wed, 17 Nov 2021 22:53:36 GMT):
we have a ticket here to remove Orion ATs - we were holding off until end of Nov since that's when it's officially out of support. But yeah don't spend time upgrading them https://github.com/hyperledger/besu/issues/3030
rjones (Wed, 17 Nov 2021 22:55:07 GMT):
we were working with @peter_somogyvari on a tool to automate this - if we could get some hands to finish it, it would be awesome
rjones (Wed, 17 Nov 2021 22:55:27 GMT):
https://github.com/petermetz/github-organization-automation-tool
RoboCopsGoneMad (Wed, 17 Nov 2021 23:06:40 GMT):
Thanks Sally, I've described my dependency on those issues.
macfarla (Wed, 17 Nov 2021 23:10:20 GMT):
ok cool. @m.terry is working on the Tessera as a process ticket
macfarla (Thu, 18 Nov 2021 05:52:27 GMT):
We also created this ticket for mocking out the enclave in unit tests since that is separate from the ATs and can be done in parallel https://github.com/hyperledger/besu/issues/3083
RoboCopsGoneMad (Thu, 18 Nov 2021 14:14:08 GMT):
gm :coffee:
RoboCopsGoneMad (Thu, 18 Nov 2021 14:33:23 GMT):
yes! awesome, I started doing this but I didn't understand the key rotation at work on the enclave implementation so i wasn't sure what to mock. a static key turned out to be a bad idea. I could totally take this on if someone could describe the minimum complexity needed for the mock.
macfarla (Fri, 19 Nov 2021 01:30:12 GMT):
ok - @frankliawesome is going to start work on this one today, he may reach out to you
macfarla (Sun, 21 Nov 2021 20:14:39 GMT):
There are 4 PRs adding and removing maintainers open for voting https://github.com/hyperledger/besu/pulls?q=is%3Apr+is%3Aopen+maintainers
RoboCopsGoneMad (Mon, 22 Nov 2021 15:10:37 GMT):
I'm cleaning up the agenda for tomorrows call, is there anything we need to still discuss on coding conventions? I believe our previous conclusion was: update the wiki page with examples, and reasoning, and we can reach consensus on it asynchronously.
andreolf1 (Mon, 22 Nov 2021 15:49:36 GMT):
Added here the agenda items for the call tomorrow: https://wiki.hyperledger.org/display/BESU/2021-11-23+Besu+Contributor+Call
Feel free to add your topics in the agenda that you want to discuss. Looking forward
fab-10 (Mon, 22 Nov 2021 17:26:33 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=pB26ttct7jjaTk9sm) Since the license plugin project does not seem to be very active, I am proposing this PR (https://github.com/hyperledger/besu/pull/3095) to temporary use a build with the improvement until an official upstream release is available.
RoboCopsGoneMad (Mon, 22 Nov 2021 18:11:08 GMT):
krufty, but temporary. I'm ok with it.
RoboCopsGoneMad (Mon, 22 Nov 2021 19:14:27 GMT):
I've captured some thoughts and ideas from consensys staff wrt our release process, and how we can open it up wider:
RoboCopsGoneMad (Mon, 22 Nov 2021 19:14:28 GMT):
https://hackmd.io/@OmvOnC6LSSCyzsGX-60MEA/rySMowYdt
JiriPeinlich (Tue, 23 Nov 2021 18:18:20 GMT):
@rjones Sorry for bothering. I was advised to ask you. Could you add me to the hyperledger organisation on github?
rjones (Tue, 23 Nov 2021 18:27:47 GMT):
@JiriPeinlich what is your github ID?
JiriPeinlich (Tue, 23 Nov 2021 18:28:22 GMT):
https://github.com/gezero
rjones (Tue, 23 Nov 2021 18:29:34 GMT):
invite sent
JiriPeinlich (Tue, 23 Nov 2021 18:29:47 GMT):
thank you :-)
sajz (Tue, 23 Nov 2021 19:12:47 GMT):
Has joined the channel.
sajz (Tue, 23 Nov 2021 19:12:47 GMT):
Hi, can we get some reviews for:
`Add Fabio as maintainer`: https://github.com/hyperledger/besu/pull/3052 @fab-10
`Add Simon as maintainer`: https://github.com/hyperledger/besu/pull/3068 @siladu
The 2 weeks window for reviewing ends very soon (in 3 days for Fabio). Thank you!
sajz (Tue, 23 Nov 2021 19:12:47 GMT):
Hi, can we get some reviews for:
`Add Fabio as maintainer`: https://github.com/hyperledger/besu/pull/3052 @fab-10
`Add Simon as maintainer`: https://github.com/hyperledger/besu/pull/3068 @siladu
`Add Jiri as maintainer`: https://github.com/hyperledger/besu/pull/3100 @JiriPeinlich
The 2 weeks window for reviewing ends soon -in 3 days- for Fabio. Thank you!
sajz (Tue, 23 Nov 2021 19:12:47 GMT):
Hi, can we get some reviews for:
`Add Fabio as maintainer`: https://github.com/hyperledger/besu/pull/3052 @fab-10
`Add Simon as maintainer`: https://github.com/hyperledger/besu/pull/3068 @siladu
`Add Jiri as maintainer`: https://github.com/hyperledger/besu/pull/3100 @JiriPeinlich
Thank you!
lucassaldanha (Wed, 24 Nov 2021 00:34:12 GMT):
Thanks for everyone who voted. We have reached 15/28 votes for Simon to be added as a maintainer. I am going to hit the merge button!
lucassaldanha (Wed, 24 Nov 2021 00:34:28 GMT):
PR to add Fabio is almost there :)
lucassaldanha (Wed, 24 Nov 2021 00:34:28 GMT):
PR to add Fabio is already there :)
lucassaldanha (Wed, 24 Nov 2021 00:34:28 GMT):
PR to add Fabio is already merged :)
macfarla (Wed, 24 Nov 2021 11:05:22 GMT):
@shemnon or @mbaxter could you please add 3 new maintainers to the `besu-maintainers` github group: fab-10 siladu and diega - thanks!
macfarla (Wed, 24 Nov 2021 11:48:50 GMT):
@RoboCopsGoneMad are you still stuck on the PrivacyReorgTest?
rjones (Wed, 24 Nov 2021 12:50:24 GMT):
done done done
RoboCopsGoneMad (Wed, 24 Nov 2021 15:31:27 GMT):
Yes I am, but i have shelved the upgrade for a moment while I work on EIP-4399
macfarla (Wed, 24 Nov 2021 20:44:15 GMT):
thanks @rjones !
macfarla (Wed, 24 Nov 2021 20:51:58 GMT):
14 approvals and 14 days - I'm hitting the merge button on https://github.com/hyperledger/besu/pull/3037
shemnon (Wed, 24 Nov 2021 21:11:46 GMT):
It's got a flakey test failure, but I updated the github team membership just now.
macfarla (Wed, 24 Nov 2021 22:29:50 GMT):
Thanks @shemnon !
GarySchulte (Wed, 24 Nov 2021 23:06:05 GMT):
BlockHeader changes for EIP-4399:
https://github.com/hyperledger/besu/pull/3104
GarySchulte (Wed, 24 Nov 2021 23:09:27 GMT):
I am not keen on having to dual-purpose mixHash and random, but the spec does not leave us much option. We can separately encode, but we do not have a way to discern when we decode whether the field is mixHash or random, thus we change the mixHash to Bytes32, use it to back both fields, and differentiate only by accessor
fab-10 (Thu, 25 Nov 2021 16:27:18 GMT):
Hi, need help to review this plugin-api change https://github.com/hyperledger/besu/pull/3065/files#diff-2f04771becb5775fa1004ec41fbd417fcb3e31a9a03fa801c908bab0a868a388L168
the change breaks the compatibility, but since that specific method is tagged as `@Unstable` I think it is ok to change the signature, but want to be sure.
What do you think?
GarySchulte (Thu, 25 Nov 2021 17:45:42 GMT):
https://github.com/hyperledger/besu/pull/3104
is the bare minimum viable PR I can make to move forward with merging the merge branch into main. I would really appreciate some :eyes: on it. merge is blocked on this, TIA
GarySchulte (Thu, 25 Nov 2021 17:54:27 GMT):
Thanks @shemnon
JiriPeinlich (Fri, 26 Nov 2021 12:54:56 GMT):
@atoulme, do you by any chance know if I can do something to get https://github.com/apache/incubator-tuweni/tree/v2.1.0-incubating-rc out? I would like to update the besu dependency. I noticed that the last ci on the branch failed for license checks. https://github.com/apache/incubator-tuweni/runs/4234046724?check_suite_focus=true
JiriPeinlich (Fri, 26 Nov 2021 12:56:55 GMT):
oh I just noticed you replied to me already https://chat.hyperledger.org/channel/besu-contributors?msg=C8YATXDm4xc8KDat9
atoulme (Mon, 29 Nov 2021 21:26:15 GMT):
please ask questions about Tuweni... on Tuweni's lists
atoulme (Mon, 29 Nov 2021 21:26:22 GMT):
Yep. I am happy to help there.
RoboCopsGoneMad (Tue, 30 Nov 2021 14:09:41 GMT):
gm besu frens. :coffee: just out of curiosity, have we had to deal with circular dependencies between projects before? I'm trying to avoid having to factor out MetricsService due to a circular dependency (expressed at acceptance testing) between besu and ethsigner.
RoboCopsGoneMad (Tue, 30 Nov 2021 14:09:43 GMT):
https://github.com/ConsenSys/ethsigner/issues/403
RoboCopsGoneMad (Tue, 30 Nov 2021 14:13:39 GMT):
we could always release with a snapshot dependency, but that is pretty gross
RoboCopsGoneMad (Tue, 30 Nov 2021 14:59:29 GMT):
looks like teku and web3signer are in the same situation. i think we need to break our dep on ethsigner, it's not just metrics but pluginapi also.
RoboCopsGoneMad (Tue, 30 Nov 2021 15:29:36 GMT):
I think @usmansaleem s fix might have sorted this out for everyone..... you rock
macfarla (Tue, 30 Nov 2021 23:24:38 GMT):
Proposal to move 2 more maintainers to emeritus status https://github.com/hyperledger/besu/pull/3119
macfarla (Tue, 30 Nov 2021 23:25:23 GMT):
@mbaxter and @abdelhamidbakhta reach out if you still want to be involved!
macfarla (Wed, 01 Dec 2021 02:48:49 GMT):
One vote shy of an absolute majority on this one https://github.com/hyperledger/besu/pull/3100
pinges (Wed, 01 Dec 2021 04:41:48 GMT):
Proposal to add Frank Li as a maintainer for Besu: https://github.com/hyperledger/besu/pull/3120
lucassaldanha (Wed, 01 Dec 2021 08:29:39 GMT):
:warning: The head of `main` branch for repository `hyperledger/besu` has been rebased to deal with a DCO issue, please follow the instructions on Fixing DCO on main/release to fix your local repositories :warning:
antonydenyer (Wed, 01 Dec 2021 13:17:14 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=zKPgCycX9W4jEPj6A) I don't think you've got a circular dependency, besu doesn't depend on binaries from ethsigner? Why not just wait until besu has done a release (sure depend on snapshot for a pr to get it working)
RoboCopsGoneMad (Wed, 01 Dec 2021 17:05:07 GMT):
You're right, it has been corrected
RoboCopsGoneMad (Wed, 01 Dec 2021 18:32:29 GMT):
@antonydenyer brings up a really good point: https://github.com/hyperledger/besu/commits/main
RoboCopsGoneMad (Wed, 01 Dec 2021 18:32:39 GMT):
that looks bad, and it's all sonarqubes fault.
RoboCopsGoneMad (Wed, 01 Dec 2021 18:33:24 GMT):
it's already a non-required check, but the optics are real bad since github displays the one failure
RoboCopsGoneMad (Wed, 01 Dec 2021 19:55:53 GMT):
for instance, check this out: this pr passes, all green https://github.com/hyperledger/besu/pull/3108
RoboCopsGoneMad (Wed, 01 Dec 2021 19:56:09 GMT):
but when it is merged to main, it failed sonarqube
RoboCopsGoneMad (Wed, 01 Dec 2021 22:50:44 GMT):
I need to put the Vert.x upgrade on the back burner, but anyone else is welcome to pick it up from my draft PR. It's really close, there are just 2 classes that are failing during acceptance test, and I suspect all their tests are failing for the same reason.
macfarla (Thu, 02 Dec 2021 00:58:24 GMT):
14 votes on this one - I'm hitting merge! https://github.com/hyperledger/besu/pull/3100 @ajsutton or @rjones would you add @JiriPeinlich to the maintainers github group - thanks!
RoboCopsGoneMad (Thu, 02 Dec 2021 00:59:13 GMT):
:champagne:
macfarla (Thu, 02 Dec 2021 00:59:30 GMT):
DCO fail on your last commit :crying_cat_face:
RoboCopsGoneMad (Thu, 02 Dec 2021 01:03:18 GMT):
omg that was killing me yesterday. i thought it was complaining about franks commit?
RoboCopsGoneMad (Thu, 02 Dec 2021 01:04:01 GMT):
https://github.com/hyperledger/besu/pull/3073/checks?check_run_id=4371343809
RoboCopsGoneMad (Thu, 02 Dec 2021 01:04:20 GMT):
or are you referring to something else? i checked most recent and it is signed off on
rjones (Thu, 02 Dec 2021 01:16:42 GMT):
@macfarla invite extended. I do have some questions about some things, though.
rjones (Thu, 02 Dec 2021 01:17:29 GMT):
@macfarla [this team has much wider access, and is larger.](https://github.com/orgs/hyperledger/teams/besu-contributors/repositories)
rjones (Thu, 02 Dec 2021 01:19:25 GMT):
@macfarla [I also noticed this person has direct "maintain" access.](https://github.com/siladu)
macfarla (Thu, 02 Dec 2021 02:16:48 GMT):
they should be in the group rather than having access directly
macfarla (Thu, 02 Dec 2021 02:20:00 GMT):
hyperledger contributors - that one has 379 members
macfarla (Thu, 02 Dec 2021 02:20:26 GMT):
thanks
macfarla (Thu, 02 Dec 2021 02:23:31 GMT):
argh sorry you are right
RoboCopsGoneMad (Thu, 02 Dec 2021 02:24:00 GMT):
i'm still unclear how to fix it though
macfarla (Thu, 02 Dec 2021 02:24:46 GMT):
yeah... is that just from merging main branch into yours?
RoboCopsGoneMad (Thu, 02 Dec 2021 02:25:52 GMT):
yes. i think he didn't collapse all his signoffs when squashing his merge. it will go away when i squash, i wonder if i can amend/edit his commit comment
macfarla (Thu, 02 Dec 2021 02:26:04 GMT):
@frankliawesome :wave:
frankliawesome (Thu, 02 Dec 2021 02:27:00 GMT):
Has joined the channel.
frankliawesome (Thu, 02 Dec 2021 02:27:01 GMT):
haha we were talking about this one, it's pretty unclear how this happened?
macfarla (Thu, 02 Dec 2021 02:30:21 GMT):
I'm confused though because that one has been merged into main - if it's wonky wouldn't it affect all PRs that have merged?
macfarla (Thu, 02 Dec 2021 02:31:24 GMT):
is it maybe that it's missing jflo signoff? `frankisawesome authored and jflo committed yesterday` but all the signoffs are frank?
RoboCopsGoneMad (Thu, 02 Dec 2021 02:32:36 GMT):
thats what I thought following the prescribed fix for the DCO would do, but it didn't
macfarla (Thu, 02 Dec 2021 02:32:50 GMT):
oh
macfarla (Thu, 02 Dec 2021 02:33:45 GMT):
does git commit --amend only work on the most recent commit
RoboCopsGoneMad (Thu, 02 Dec 2021 02:34:32 GMT):
ooooh that sounds right. i probably need to specify HEAD~X or the commit sha....
macfarla (Thu, 02 Dec 2021 02:36:26 GMT):
you might need rebase https://www.atlassian.com/git/tutorials/rewriting-history
RoboCopsGoneMad (Thu, 02 Dec 2021 02:36:47 GMT):
yeah interactive rebase
RoboCopsGoneMad (Thu, 02 Dec 2021 02:36:50 GMT):
one sec
RoboCopsGoneMad (Thu, 02 Dec 2021 02:39:44 GMT):
that did it
macfarla (Thu, 02 Dec 2021 02:39:52 GMT):
woo
RoboCopsGoneMad (Thu, 02 Dec 2021 02:39:55 GMT):
you gonna pick this up Sally?
macfarla (Thu, 02 Dec 2021 02:40:03 GMT):
ahhh
macfarla (Thu, 02 Dec 2021 02:40:05 GMT):
maybe
macfarla (Thu, 02 Dec 2021 02:40:09 GMT):
not today!
RoboCopsGoneMad (Thu, 02 Dec 2021 02:40:19 GMT):
no problem
rjones (Thu, 02 Dec 2021 04:00:33 GMT):
Screen Shot 2021-12-01 at 8.00.08 PM.png
rjones (Thu, 02 Dec 2021 04:02:36 GMT):
done
macfarla (Thu, 02 Dec 2021 04:16:30 GMT):
thanks!
macfarla (Thu, 02 Dec 2021 04:19:40 GMT):
does "child team members" include eg besu-docs maintainers?
macfarla (Thu, 02 Dec 2021 04:20:07 GMT):
Screen Shot 2021-12-02 at 2.19.56 pm.png
macfarla (Thu, 02 Dec 2021 04:20:27 GMT):
I see 38 members in the top right but 24 at the top of the list
macfarla (Thu, 02 Dec 2021 11:17:07 GMT):
Re the docker base image vulnerabilities ticket https://github.com/hyperledger/besu/issues/3045 - do we need to stick with jdk11 or can we use 16/17 for the docker image?
RoboCopsGoneMad (Thu, 02 Dec 2021 15:04:06 GMT):
I would be ok with 17 if we pass all tests on 17, which means updating the build executors first
rjones (Thu, 02 Dec 2021 15:34:27 GMT):
if you want to use it that way, the besu contributors group should have read access, and the teams below should have the correct access per repo
rjones (Thu, 02 Dec 2021 15:34:58 GMT):
as it is, any member of any team has write to three repos, and triage for one
fab-10 (Thu, 02 Dec 2021 16:43:22 GMT):
the ticket talks about `openjdk:11.0.7-jre-slim-buster`, but now we are using `openjdk:11-jre-slim-buster` that follow the most recent version of java 11, so it could be that there no security issues already
shemnon (Thu, 02 Dec 2021 17:17:18 GMT):
To fully support Java 17 in build we need to upgrade to Gradle 7.3
RoboCopsGoneMad (Thu, 02 Dec 2021 17:38:22 GMT):
the sonarqube failures are because it considers "new" code to be all commits since the last release. I think those failures are legit :(
macfarla (Thu, 02 Dec 2021 19:52:02 GMT):
I think that's done already
macfarla (Thu, 02 Dec 2021 19:52:06 GMT):
https://github.com/hyperledger/besu/pull/3109
macfarla (Thu, 02 Dec 2021 19:53:48 GMT):
Re moving maintainers to emeritus status - current guide says a quarter of inactivity. Do we want to extend that? 6 months? Longer?
> A general measure of inactivity will be no commits or code review comments for one reporting quarter, although this will not be strictly enforced if the maintainer expresses a reasonable intent to continue contributing.
macfarla (Thu, 02 Dec 2021 19:56:54 GMT):
@matkt
macfarla (Thu, 02 Dec 2021 19:59:45 GMT):
that's this one, right https://github.com/hyperledger/besu/issues/2963
RoboCopsGoneMad (Thu, 02 Dec 2021 20:03:15 GMT):
sounds like that one was to focus on the image itself being end of life, agnostic of jdk versions. @GarySchulte any problem with combining these any more? I'm ok with it.
macfarla (Thu, 02 Dec 2021 20:20:01 GMT):
sonarqube is reporting acceptance-test code as uncovered lines - how do we fix that - eg https://sonarcloud.io/component_measures?id=hyperledger_besu&metric=new_uncovered_lines&pullRequest=3118&view=list
RoboCopsGoneMad (Thu, 02 Dec 2021 20:27:58 GMT):
i think we solved that for integration tests, probably same solution.....
RoboCopsGoneMad (Thu, 02 Dec 2021 20:30:30 GMT):
probably similar to this: https://github.com/hyperledger/besu/blob/5221bd5a7dde942f926d7ae03b8003ca70030450/build.gradle#L798
shemnon (Thu, 02 Dec 2021 21:20:57 GMT):
blink and you miss it.
antonydenyer (Fri, 03 Dec 2021 08:46:58 GMT):
I'm going to move the test plugins that are used in the acceptance tests into the same package namespace
fab-10 (Fri, 03 Dec 2021 11:22:26 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=ocqPxT8oDzDjtN46a) I think we need to support Besu on Java11 for some time, since some organizations are slow to upgrade to the last java version, so the solution here could simply be to use another OpenJDK distribution, that has no vulnerabilities as suggested here https://github.com/hyperledger/besu/issues/3045#issuecomment-985437175
diegoll (Fri, 03 Dec 2021 12:53:15 GMT):
gm all, I would appreciate your thoughts about this PR https://github.com/hyperledger/besu/pull/3123
shemnon (Fri, 03 Dec 2021 15:49:13 GMT):
(a) 11 is losing support in 2023. We need to be ahead of the ball - https://www.oracle.com/java/technologies/java-se-support-roadmap.html
shemnon (Fri, 03 Dec 2021 15:50:33 GMT):
(b) we are long past the age of company wide Java versions. We are well in the age of bring your own JDK. If an enterprise balks then the node is either run in a docker container or a local JRE/Java runtime is deployed with the app. There are now no loger deep OS integrations and stuff like the Apple JDK being bound to the OS version no longer applies.
shemnon (Fri, 03 Dec 2021 15:51:27 GMT):
Creating a third "all in one" distribution with a JDK wrapped in sounds better to me than staying on Java 11 past the release of Java 18.
fab-10 (Fri, 03 Dec 2021 15:57:23 GMT):
I am perfectly fine with upgrading to Java17, if no organization needs to run Besu on Java11, just saying that based on my recent experience , some organizations (big banks for example) do not like to run docker and uses old version of java
shemnon (Fri, 03 Dec 2021 21:57:21 GMT):
Java 17 is a designated LTS release, so I don't expect any flack from a big bank or other enterprise. Java 12-16... different story.
diegoll (Mon, 06 Dec 2021 12:53:00 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=mx4uxKTYp4Ji5cHLi) :grimacing:
RoboCopsGoneMad (Mon, 06 Dec 2021 14:08:45 GMT):
I'm implementing 4488 but I'm not sure what kind of integration test to write. I wasn't able to find anything (yet), but I assume something like London or Berlin should've had an integration test that covered before/transition/after.
i need to cover a gas calculator, a block body validator, and a transaction selector for mining. I'm imagining a test where I start a chain off with a london block, configure the upgrade for block 2, mine an upgraded block, and then assert on the changes. Is this the right way to think about our integration tests, and is there an existing one that does something like this that I should reference?
diegoll (Mon, 06 Dec 2021 21:22:26 GMT):
Hey Justin, I think I got the missing piece for making the acceptance tests to pass. Take a look at commit [4c7b28f](https://github.com/diega/besu/commit/4c7b28f11f2aa05076eb795a0ff715c8743babfc) in my branch. I couldn't run the whole `acceptanceTest` task on my box I don't know why, but I ran each of the reported failures manually and they are passing
diegoll (Mon, 06 Dec 2021 21:22:26 GMT):
Hey Justin, I think I got the missing piece for making the acceptance tests to pass. Take a look at commit [4c7b28f](https://github.com/diega/besu/commit/4c7b28f11f2aa05076eb795a0ff715c8743babfc) in my fork. I couldn't run the whole `acceptanceTest` task on my box I don't know why, but I ran each of the reported failures manually and they are passing
diegoll (Mon, 06 Dec 2021 21:22:26 GMT):
Hey Justin, I think I got the missing piece for making the acceptance tests pass. Take a look at commit [4c7b28f](https://github.com/diega/besu/commit/4c7b28f11f2aa05076eb795a0ff715c8743babfc) in my fork. I couldn't run the whole `acceptanceTest` task on my box I don't know why, but I ran each of the reported failures manually and they are passing
Rp27 (Mon, 06 Dec 2021 23:14:54 GMT):
Has joined the channel.
macfarla (Mon, 06 Dec 2021 23:23:39 GMT):
@rjones can you add @Rp27 to the hyperledger github group? New member of our team - thanks!
rjones (Mon, 06 Dec 2021 23:27:37 GMT):
github username?
macfarla (Mon, 06 Dec 2021 23:47:00 GMT):
Contributor call happening in 15 min https://wiki.hyperledger.org/display/BESU/2021-12-07+Besu+Contributor+Call
macfarla (Mon, 06 Dec 2021 23:47:00 GMT):
Contributor call happening in 1hr 15 min https://wiki.hyperledger.org/display/BESU/2021-12-07+Besu+Contributor+Call
macfarla (Mon, 06 Dec 2021 23:47:58 GMT):
RP27
rjones (Tue, 07 Dec 2021 00:40:27 GMT):
invite extended
macfarla (Tue, 07 Dec 2021 00:50:58 GMT):
thanks
sajz (Tue, 07 Dec 2021 16:41:14 GMT):
Hi all, we are planning to do a release mid next week which will be 22.1.0-RC1. Then no release during the holiday. RC2 early january.
sajz (Tue, 07 Dec 2021 16:41:14 GMT):
Hi all, we are planning to do a release mid next week which will be 22.1.0-RC1. Then no release during the holidays. RC2 early january.
sajz (Tue, 07 Dec 2021 16:43:09 GMT):
We (consensys teams) will also be looking specifically at some issues of bad performances with Besu (around sync, peering, tracing etc) from tomorrow to mid next week. If you have availability to investigate or review, let us know. We can cross post relevant info and github issues in this channel.
RoboCopsGoneMad (Tue, 07 Dec 2021 18:39:18 GMT):
feel free to pull my branch into yours and open a PR to see if it passes CI
diegoll (Tue, 07 Dec 2021 19:04:15 GMT):
sure! I'll rebase it on top of the current `main` as well, b/c there is a little fix that needs to be added into the `build.gradle` for the _merge_ module
GarySchulte (Tue, 07 Dec 2021 22:12:15 GMT):
Hey @shemnon, I know you initially voiced some difference of opinion regarding the repurposing of DIFFICULTY as RANDOM in EIP-4399.
What are your thoughts about doing a rename on whatever hardfork contains EIP-4488? e.g.
https://github.com/hyperledger/besu/pull/3111#pullrequestreview-825808634
GarySchulte (Tue, 07 Dec 2021 22:13:09 GMT):
Are there Operation name dependencies other than how the operations print in traces, if the Operation name changes?
shemnon (Tue, 07 Dec 2021 22:14:04 GMT):
No dependencies. The name should change if the functionality changes.
shemnon (Tue, 07 Dec 2021 22:14:43 GMT):
So one hardfork has the operations array with DifficultyOperation, and the next adds RandomOperation. Actually, if you add both but add Random last it will over-write the initial add.
shemnon (Tue, 07 Dec 2021 22:14:54 GMT):
Since they are all stored in an array.
shemnon (Tue, 07 Dec 2021 22:15:03 GMT):
indexed by opcode.
GarySchulte (Tue, 07 Dec 2021 22:17:42 GMT):
Thanks. Tactically what do you think about the viability of changing the name ahead of the merge ? We could soft-fork that change along with the client release that contains the terminal total difficulty block. But there isn't really a hardfork that we can make that change with.
GarySchulte (Tue, 07 Dec 2021 22:19:59 GMT):
alternately we could make it part of the hardfork that contains eip-4488 (if indeed that happens ahead of the merge), or the first hardfork following the merge.
shemnon (Tue, 07 Dec 2021 22:37:21 GMT):
I should have read to the end of the diff. No, not like that. Difficulty should be one Operation, Random another.
shemnon (Tue, 07 Dec 2021 22:37:36 GMT):
The hard fork protocol spec puts in Random at the merge point.
GarySchulte (Wed, 08 Dec 2021 06:44:35 GMT):
Unfortunately there isn't a merge hardfork, the transition is based on a terminal total difficulty instead :/ The challenge is that we will have to operate within the same protocolSpec with 2 different behaviors for that opCode.
Perhaps we can create a RandomOperation that has the condition that was added to DifficultyOperation, and simply enable it in the EIP-4488 fork ahead of the merge. Then post-merge we can clean it up and remove the condition.
GarySchulte (Wed, 08 Dec 2021 06:44:35 GMT):
Unfortunately there isn't a merge hardfork, the transition is based on a terminal total difficulty instead :/ The challenge is that we will have to operate within the same protocolSpec with 2 different behaviors for that opCode.
Perhaps we can revert the change to DifficultyOperation, and move the condition into a new RandomOperation. Then simply enable it in the EIP-4488 fork ahead of the merge. Then post-merge we can clean it up and remove the condition.
GarySchulte (Wed, 08 Dec 2021 06:44:35 GMT):
Unfortunately there isn't a merge hardfork, the transition is based on a terminal total difficulty instead :/ The challenge is that we will have to operate within the same protocolSpec with 2 different behaviors for that opCode.
Perhaps we can revert the change to DifficultyOperation, and move the condition into a new RandomOperation. Then simply enable it in the EIP-4488 fork ahead of the merge. And finally post-merge we can clean it up and remove the condition.
GarySchulte (Wed, 08 Dec 2021 06:44:35 GMT):
Unfortunately there isn't a merge hardfork, the transition is based on a terminal total difficulty instead :/ The challenge is that we will have to operate within the same protocolSpec with 2 different behaviors for that opCode.
Perhaps we can revert the change to DifficultyOperation, and move the condition into a new RandomOperation. Then simply enable it in the EIP-4488 fork ahead of the merge. And finally post-merge we can remove the condition.
GarySchulte (Wed, 08 Dec 2021 06:44:59 GMT):
(and revert the change to DifficultyOperation)
macfarla (Wed, 08 Dec 2021 08:32:38 GMT):
Did that help? https://sonarcloud.io/component_measures?id=hyperledger_besu&metric=new_coverage&pullRequest=3118&view=list
macfarla (Wed, 08 Dec 2021 08:50:15 GMT):
I added a ticket because this is kind of annoying me https://github.com/hyperledger/besu/issues/3142
antonydenyer (Wed, 08 Dec 2021 09:36:48 GMT):
Sadly not, for some reason it didn't. I moved the files into `org/hyperledger/besu/tests/acceptance/` but still no dice
shemnon (Wed, 08 Dec 2021 15:35:55 GMT):
I would prefer that, a difficulty operation that has no merge logic. Classic, Ropsten, rinkeby, and goerli will appreciate it.
shemnon (Wed, 08 Dec 2021 15:36:06 GMT):
How will the ForkID handle merge signaling?
shemnon (Wed, 08 Dec 2021 15:37:54 GMT):
Would the "fork" be the block where the terminal difficulty is dialed in?
RoboCopsGoneMad (Wed, 08 Dec 2021 18:13:29 GMT):
I've updated comments on this issue here: https://github.com/hyperledger/besu/pull/3111
RoboCopsGoneMad (Wed, 08 Dec 2021 18:18:27 GMT):
sounds like this is best described as a client specific softfork, which is later a hardfork based on TDD
macfarla (Wed, 08 Dec 2021 20:07:12 GMT):
I've hit auto merge on the PR to make @frankliawesome a maintainer (16 votes) https://github.com/hyperledger/besu/pull/3120 - @rjones can you add him to the group? Thanks!
macfarla (Wed, 08 Dec 2021 20:07:28 GMT):
| Frank Li | frankisawesome | frankliawesome |
rjones (Wed, 08 Dec 2021 21:03:29 GMT):
done!
macfarla (Wed, 08 Dec 2021 21:03:44 GMT):
thanks!
RoboCopsGoneMad (Thu, 09 Dec 2021 14:32:22 GMT):
gm besu. Mmmmmmm coffee :coffee:
GarySchulte (Thu, 09 Dec 2021 22:21:45 GMT):
Good afternoon from PNW.
GarySchulte (Thu, 09 Dec 2021 22:23:49 GMT):
question - is ECIP-1049 settled history for eth classic ? I.e. is there any reason to keep the ECIP1049-DEV network definition? genesis file, dev network name, etc. Is that still an active network or was that a long lived test net that is now dead?
GarySchulte (Thu, 09 Dec 2021 22:27:56 GMT):
https://github.com/hyperledger/besu/blob/main/besu/src/main/java/org/hyperledger/besu/cli/config/EthNetworkConfig.java#L63
https://github.com/hyperledger/besu/blob/main/besu/src/main/java/org/hyperledger/besu/cli/config/NetworkName.java#L27
https://github.com/hyperledger/besu/blob/main/config/src/main/resources/ecip1049_dev.json
et cetera
GarySchulte (Thu, 09 Dec 2021 22:28:19 GMT):
@diegoll ?
GarySchulte (Thu, 09 Dec 2021 22:29:36 GMT):
@diegoll ?
diegoll (Thu, 09 Dec 2021 22:54:47 GMT):
no, I think that as long as we keep the possibility to enable keccak mining for testing, we can maintain genesis files outside upstream
diegoll (Thu, 09 Dec 2021 22:56:00 GMT):
once we settle a long lived testnet, with a proper name, we can get it back, but there is no timeline for it yet
frankliawesome (Fri, 10 Dec 2021 00:38:20 GMT):
@rjones hello! looking at this ticket https://github.com/hyperledger/besu/issues/3142 it seems like we just need one config for code coverage on sonar cloud, is it possible to give me access?
rjones (Fri, 10 Dec 2021 00:42:39 GMT):
@frankliawesome I'm not sure what you need me to do. Sonar is mostly handled via in-repo changes
frankliawesome (Fri, 10 Dec 2021 00:50:45 GMT):
I think you're right.. Most stuff can be done via either config file or in UI changes but definitely seems like doing a PR in repo makes a lot more senes!
diegoll (Fri, 10 Dec 2021 01:23:53 GMT):
I'm sorry, I just realized that at any case we gonna need a name to attach the fork to. Right now we are focused on the [upcoming hard fork (Mystique)](https://github.com/ethereumclassic/ECIPs/pull/452) but as soon as we get that out we plan to start testing more consciously ECIP1049
RoboCopsGoneMad (Fri, 10 Dec 2021 16:10:23 GMT):
folks we are preparing a release now to address the log4j zero day
RoboCopsGoneMad (Fri, 10 Dec 2021 16:20:13 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to 21.10.3 until further notice
RoboCopsGoneMad (Fri, 10 Dec 2021 19:17:32 GMT):
release is complete, 21.10.3 now availble
RoboCopsGoneMad (Fri, 10 Dec 2021 19:17:50 GMT):
may god have mercy on log4j users everywhere
rjones (Fri, 10 Dec 2021 19:18:07 GMT):
time is a flat circle.
RoboCopsGoneMad (Fri, 10 Dec 2021 19:24:32 GMT):
pr for new snapshot version: https://github.com/hyperledger/besu/pull/3159
shemnon (Fri, 10 Dec 2021 19:56:25 GMT):
review request - Homebrew update - https://github.com/hyperledger/homebrew-besu/pull/64
shemnon (Fri, 10 Dec 2021 20:43:58 GMT):
This seems anti-climactic - I published the CVE for the SHL/SHR/SAR bug from about a month back - https://github.com/hyperledger/besu/security/advisories/GHSA-7pg2-p5vj-xp5h - CVE-2021-41272
shemnon (Fri, 10 Dec 2021 20:44:11 GMT):
The plan from earlier in the week was to publish it today anyway.
shemnon (Fri, 10 Dec 2021 20:44:28 GMT):
Fix is in 21.10.2, so the log4j fix already has it.
shemnon (Fri, 10 Dec 2021 20:45:24 GMT):
Should this also go out on the besu twitter feed?
rjones (Fri, 10 Dec 2021 21:04:36 GMT):
would you like it to?
shemnon (Fri, 10 Dec 2021 21:18:24 GMT):
Everyone should have updated from the Log4J patch, so publicity isn't needed. I'm on with a non-tweet.
shemnon (Fri, 10 Dec 2021 21:18:24 GMT):
Everyone should have updated from the Log4J patch, so publicity isn't needed. I'm ok with a non-tweet.
macfarla (Mon, 13 Dec 2021 00:36:30 GMT):
LFR correcting download link for lastest release https://github.com/hyperledger/besu/pull/3161
diegoll (Mon, 13 Dec 2021 03:07:00 GMT):
Hello @macfarla, what are the options you have in mind for the `TesseraTestHarness`? ref: https://github.com/hyperledger/besu/pull/3135#issuecomment-992047203
diegoll (Mon, 13 Dec 2021 03:11:26 GMT):
I don't know if would be possible to move the entire environment setup to Docker images handled from outside the tests themselves and use [CircleCI configurations](https://circleci.com/docs/2.0/building-docker-images/) directly
macfarla (Mon, 13 Dec 2021 04:10:50 GMT):
we have work in progress to a) change Tessera so it can be run as a process with runtime port discoverability and b) get rid of containerTests and run tessera as a process for Besu ATs
macfarla (Mon, 13 Dec 2021 04:12:07 GMT):
Tessera PR https://github.com/ConsenSys/tessera/pull/1356 - once this is done, we need a Tessera release and then update Besu to use that for ATs instead of containerTests.
macfarla (Mon, 13 Dec 2021 04:13:04 GMT):
@m.terry is working on this
macfarla (Mon, 13 Dec 2021 04:13:53 GMT):
I'm optimistic we can get this in before christmas :fingers_crossed:
macfarla (Mon, 13 Dec 2021 04:14:50 GMT):
but if we don't want to wait for that, we could ignore those tests for now, OR change integrationTests to run on the large executor (and change it back once that work is finished)
m.terry (Mon, 13 Dec 2021 04:44:09 GMT):
Has joined the channel.
macfarla (Mon, 13 Dec 2021 05:42:11 GMT):
anti-climactic is probably good!
macfarla (Mon, 13 Dec 2021 06:26:07 GMT):
LFR https://github.com/hyperledger/besu/pull/3137
RoboCopsGoneMad (Mon, 13 Dec 2021 14:41:17 GMT):
I'd say change the executor size. It's easy, and won't impact the delivery pipeline (other than price)
diegoll (Mon, 13 Dec 2021 14:41:52 GMT):
ok, I'll give it a try
shemnon (Mon, 13 Dec 2021 15:51:51 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=Wa7BocriH6fEThfK4) changes requested.
diegoll (Mon, 13 Dec 2021 20:54:23 GMT):
I tried changing the `executor` in my fork but it didn't work. I tried large and xlarge, but none worked. Same problem, "Could not find a valid Docker environment. Please see logs and check configuration". E.g [CircleCI#66944](https://app.circleci.com/pipelines/github/hyperledger/besu/12172/workflows/9aca0bea-da05-4b62-94d5-8230325901bb/jobs/66944)
macfarla (Mon, 13 Dec 2021 23:48:58 GMT):
the acceptanceTests use docker and they use this one
macfarla (Mon, 13 Dec 2021 23:49:00 GMT):
acceptanceTests:
parallelism: 6
executor: xl_machine_executor
macfarla (Mon, 13 Dec 2021 23:49:36 GMT):
can you try with that one?
shemnon (Tue, 14 Dec 2021 02:31:18 GMT):
Alternative propsoal: change to ubuntu linux - https://github.com/hyperledger/besu/issues/3171
diegoll (Tue, 14 Dec 2021 07:01:31 GMT):
It worked! Some other ITs have failed that I'll try to tackle tomorrow. Thanks Sally!
macfarla (Tue, 14 Dec 2021 08:29:25 GMT):
:rocket:
macfarla (Tue, 14 Dec 2021 10:54:18 GMT):
I think the difference with those two failing tests is that Tessera gives slightly different error messages to Orion. So to fix those, just update the error message to what Tessera gives and that should be it!
diegoll (Tue, 14 Dec 2021 14:31:40 GMT):
I'll try that. My only concern is that the responses from Tessera seems a little generic and I wouldn't be sure if those are coming from the tested scenario or from some hidden problem. I'm mostly referring to the one returning "Index 9 out of bounds for length 9" instead of "EnclaveDecodePublicKey", which sounds like an `ArrayIndexOutOfBoundsException`
diegoll (Tue, 14 Dec 2021 17:08:57 GMT):
Hey! tests seems to be passing in [#3135](https://github.com/hyperledger/besu/pull/3135). I'm not too convinced by [8ab7942](https://github.com/hyperledger/besu/pull/3135/commits/8ab7942f10c452ca6a62971e8d2806450c4a4f5b) I would appreciate your input on this if you have some time
diegoll (Tue, 14 Dec 2021 17:08:57 GMT):
Hey! tests seems to be passing in [#3135](https://github.com/hyperledger/besu/pull/3135). I'm not too convinced by [8ab7942](https://github.com/hyperledger/besu/pull/3135/commits/8ab7942f10c452ca6a62971e8d2806450c4a4f5b), I would appreciate your input on this if you have some time
atoulme (Tue, 14 Dec 2021 17:48:01 GMT):
https://twitter.com/ayamiyagotchi/status/1470654120180649986?s=21 <— what does this mean that besu received a grant from the EF?
atoulme (Tue, 14 Dec 2021 17:48:32 GMT):
Is it going to HLF? Or is it going to individual committers?
atoulme (Wed, 15 Dec 2021 16:59:51 GMT):
any idea folks? @SajidaZ ?
macfarla (Thu, 16 Dec 2021 00:21:33 GMT):
Besu CI assemble step failing for the last 8+ hrs on this error from the web3 solidity plugin:
FAILURE: Build failed with an exception. Looks like `https://internal.services.web3labs.com/api/solidity/versions/` is down
* What went wrong:
Execution failed for task ':acceptance-tests:tests:compileSolidity'.
> java.lang.Exception: Failed to get solidity version from server
macfarla (Thu, 16 Dec 2021 01:01:12 GMT):
I have opened an issue on the web3j solidity plugin repo and emailed web3labs
sajz (Thu, 16 Dec 2021 08:39:48 GMT):
Hi @atoulme , regarding the Incentive Program that the EF rolled out here are some clarifications: it is not a grant, it is a multi year program designed to incentivize client teams (organizations) to dedicate time and resources to Ethereum roadmap (testnet participation, spec implementation, testing, R&D etc). Specifically to prepare for the Merge and the upcoming milestones, like Sharding. The 4608 ETH that the EF is prodiving to run the validators will not be available for client teams upfront, it will be used to load the validators but the withdrawal keys will be kept by the Ethereum Foundation and disclosed over several years. However, client teams will be able to directly access the rewards generated by those validators over time, if the team keeps contributing (this will be judged by the EF). The program can not be regarded as a grant (where money is directly available) but it will certainly lead to creation of grants/bounties once it's up and running (post-merge, the validators will be generating rewards). The EF just recently made the information public, but for the past 4 months the team at Consensys has worked on setting up the infrastructure to be able to run those validators safely. The infra is ready to receive validators but the handover has not happened yet. Let me know if you have any questions or remarks on this topic.
sajz (Thu, 16 Dec 2021 08:39:48 GMT):
Hi, regarding the Incentive Program that the EF rolled out here are some clarifications: it is not a grant, it is a multi year program designed to incentivize client teams (organizations) to dedicate time and resources to Ethereum roadmap (testnet participation, spec implementation, testing, R&D etc). Specifically to prepare for the Merge and the upcoming milestones, like Sharding. The 4608 ETH that the EF is prodiving to run the validators will not be available for client teams upfront, it will be used to load the validators but the withdrawal keys will be kept by the Ethereum Foundation and disclosed over several years. However, client teams will be able to directly access the rewards generated by those validators over time, if the team keeps contributing (this will be judged by the EF). The program can not be regarded as a grant (where money is directly available) but it will certainly lead to creation of grants/bounties once it's up and running (post-merge, the validators will be generating rewards). The EF just recently made the information public, but for the past 4 months the team at Consensys has worked on setting up the infrastructure to be able to run those validators safely. The infra is ready to receive validators but the handover has not happened yet. Let me know if you have any questions or remarks on this topic.
sajz (Thu, 16 Dec 2021 08:39:48 GMT):
macfarla (Thu, 16 Dec 2021 10:16:09 GMT):
https://github.com/hyperledger/besu/pull/3183
macfarla (Thu, 16 Dec 2021 10:16:09 GMT):
Fix for the web3j solidity plugin issue - install solc on the CI executor https://github.com/hyperledger/besu/pull/3183
macfarla (Thu, 16 Dec 2021 10:18:01 GMT):
Fix for the web3j solidity plugin issue - install solc on the CI executor https://github.com/hyperledger/besu/pull/3183
fab-10 (Thu, 16 Dec 2021 14:19:19 GMT):
FYI starting the quarterly besu release candidate 22.1.0-RC1 from main.
fab-10 (Thu, 16 Dec 2021 14:19:19 GMT):
FYI starting the quarterly besu release candidate 22.1.0-RC1 from main, please don't merge anything to main until further notice
shemnon (Thu, 16 Dec 2021 14:57:53 GMT):
Why was only ConsenSys involved in this discussions? The client is _Hyperledger Besu_ not _ConsenSys Besu_
shemnon (Thu, 16 Dec 2021 14:57:53 GMT):
Why was only ConsenSys involved in the discussions? The client is _Hyperledger Besu_ not _ConsenSys Besu_
shemnon (Thu, 16 Dec 2021 15:13:52 GMT):
2) Who is the client team that will receive the grant? All maintainers or just ConsenSys?
fab-10 (Thu, 16 Dec 2021 15:20:55 GMT):
It's now safe to merge into main for hyperledger/besu again.
fab-10 (Thu, 16 Dec 2021 15:55:19 GMT):
Besu 22.1.0-RC1 release is now complete
RoboCopsGoneMad (Thu, 16 Dec 2021 16:21:22 GMT):
thank you ser. The only concern I have is test coverage is low, but I think we should merge anyway. Let me think on it.
diegoll (Thu, 16 Dec 2021 16:23:51 GMT):
anytime!
shemnon (Thu, 16 Dec 2021 19:07:13 GMT):
Should we consider another 21.10.x release for the latest log4J bug? (the one that prompted 2.16.0)? I don't _think_ we use thread context but the next production release is scheduled for late Jan.
macfarla (Thu, 16 Dec 2021 20:02:54 GMT):
yeah I _think_ that test is ok but we can investigate it separately. Agree with @RoboCopsGoneMad - we should merge
macfarla (Thu, 16 Dec 2021 20:58:34 GMT):
yeah I think we should. We were actually aiming to do it yesterday but got blocked on the solidity gradle plugin issue - but that's fixed now. Leave it with me
shemnon (Thu, 16 Dec 2021 21:11:53 GMT):
My thought is to do a fork off of 21.7.3, and cherry pick PRs 3171 (docker vulnerabilities) and 3175 (log4j)
atoulme (Fri, 17 Dec 2021 04:06:04 GMT):
Sajida, why wasn’t that brought up in the Besu project 4 months ago?
diegoll (Fri, 17 Dec 2021 16:13:28 GMT):
I would need approvals if we want to move forward with [#3135](https://github.com/hyperledger/besu/pull/3135) :upside_down:
SeanBohan (Fri, 17 Dec 2021 19:57:38 GMT):
Has joined the channel.
grace.hartley (Fri, 17 Dec 2021 20:06:02 GMT):
Hey all - I wanted to follow up regarding the Ethereum Foundation announcement earlier this week and wanted to apologize that this was a surprise to the Besu community at large.
I want to emphasize that it is very important to me personally and all Besu maintainers from ConsenSys that Besu is an open, inclusive, vibrant community of contributors. We want Besu to stay that way and to grow to be even better.
With the announcement, we (ConsenSys) wants this incentive program to benefit the Besu community. We are committed to finding a way to ensure contributors are rewarded for their work on the project. We would love this group’s feedback on how best to incentivize the program.
If the group agrees, I'd like to propose we as the Besu community discuss this topic in detail at the next contributor call on December 21st at 11 AM ET. ConsenSys would be happy to answer any questions others have. Tim from the EF is also planning on joining to help answer questions and provide additional context on the program. We'd also like to begin the process of how we can leverage these funds to enhance the Besu community. Also, please feel free to personally reach out to me if you’d like to discuss one on one.
atoulme (Fri, 17 Dec 2021 21:01:46 GMT):
I will likely not be able to attend. I will be skiing with the kids that day and I have to drop them off at 8am, at a ski resort. I will try to join later if I can.
atoulme (Fri, 17 Dec 2021 21:01:50 GMT):
@grace.hartley if you haven't already, please feel free to add to the agenda topics for the call? https://wiki.hyperledger.org/display/BESU/2021-12-21+Besu+Contributor+Call
atoulme (Fri, 17 Dec 2021 21:03:40 GMT):
Would you please document ahead of time as much as possible what is happening - a wiki page sounds like the right place. It would be useful to read this for me ahead of time. It looks like there is a lot going into it and I am just not understanding what is going on.
atoulme (Fri, 17 Dec 2021 21:03:52 GMT):
Would it be possible in any case to record the call? Thank you.
rjones (Fri, 17 Dec 2021 21:07:18 GMT):
I think all calls are recorded
atoulme (Fri, 17 Dec 2021 21:10:36 GMT):
Thanks, so at least I will make sure to watch afterwards.
rjones (Fri, 17 Dec 2021 21:11:19 GMT):
looking at the wiki - it looks like the last several few have _not_ been recorded.
atoulme (Fri, 17 Dec 2021 22:40:16 GMT):
No worries, let's make sure this one is recorded. If that's any trouble, I can provide a Zoom meeting url.
shemnon (Fri, 17 Dec 2021 23:19:30 GMT):
Let's use a hyperledger zoom link
madelinemurray (Sat, 18 Dec 2021 00:16:53 GMT):
@rjones - can you set the hyperledger zoom link to automatically record? My hit rate with remembering to record meetings went up significantly once I started using that feature.
rjones (Sat, 18 Dec 2021 00:32:31 GMT):
@madelinemurray the calendar link is for a consensys zoom link, so I don't have control over the settings
rjones (Sat, 18 Dec 2021 00:35:49 GMT):
@madelinemurray use this link: https://zoom.us/j/97540031823
rjones (Sat, 18 Dec 2021 00:35:49 GMT):
@madelinemurray use this link: https://zoom.us/j/97540031823?pwd=dEJrRW1LakFlWnZPelI3VGxoVjg2dz09
madelinemurray (Sat, 18 Dec 2021 00:51:39 GMT):
Thanks @rjones. @grace.hartley - do you have update access on the invite?
rjones (Sat, 18 Dec 2021 00:56:44 GMT):
I updated it in the calendar
atoulme (Sat, 18 Dec 2021 01:05:57 GMT):
Cool, I updated the wiki and added this item to the agenda
madelinemurray (Sat, 18 Dec 2021 01:10:10 GMT):
Thanks @atoulme - enjoy the skiing. Seems hard to imagine from this side of the world!
atoulme (Sat, 18 Dec 2021 01:11:07 GMT):
kids will ski and I might just put my feet up and code. Appreciate the kind words Madeline, I hope you have a good summer ;)
atoulme (Sat, 18 Dec 2021 05:07:56 GMT):
the network is still live for all I know, can't remember though :)
shemnon (Sat, 18 Dec 2021 15:49:19 GMT):
Looks like there is a _third_ log4j vulnerability: https://logging.apache.org/log4j/2.x/security.html (2.17.0) - Thoughts on wether we should roll another 21.10.x release and another 22.1.0-RCx?
shemnon (Sat, 18 Dec 2021 15:50:32 GMT):
Meta-question: thoughts on dropping log4j-core for just java.util.logging backed?
shemnon (Sat, 18 Dec 2021 16:05:05 GMT):
Here's a patch for 21.10.x - https://github.com/hyperledger/besu/pull/3198
shemnon (Sat, 18 Dec 2021 16:05:12 GMT):
RC isn't as urgent IMHO
atoulme (Sun, 19 Dec 2021 01:30:20 GMT):
Logback is better - use slf4j as api so you don’t force a runtime
atoulme (Sun, 19 Dec 2021 01:30:35 GMT):
java.util.logging is slow
atoulme (Sun, 19 Dec 2021 01:31:10 GMT):
Or even better slf4j can use java.util.logging as backend if you prefer.
shemnon (Sun, 19 Dec 2021 02:00:01 GMT):
The goal would be to get rid of attack surface. We could include instructions on how to wire in your favorite, but I don't want to be spinning releases on weekends.
shemnon (Sun, 19 Dec 2021 02:10:25 GMT):
Reminder for future releases: do not use the `/ui/native/...` URLs from artifactory, in the release notes or changelog. curl will return the intermediate hash pages and your hashes will be all wrong.
atoulme (Sun, 19 Dec 2021 03:25:32 GMT):
Slf4j reduces the surface attack since you can plug any backend. I’m game to help!
atoulme (Sun, 19 Dec 2021 03:25:49 GMT):
And yes big point on weekend releases…
shemnon (Sun, 19 Dec 2021 04:10:12 GMT):
especially when I don't have the time to figure out why circleci hates the branch.
atoulme (Sun, 19 Dec 2021 04:38:20 GMT):
I am supposed to wrap presents tonight. Might not be a great time
atoulme (Sun, 19 Dec 2021 04:41:29 GMT):
I see your pr. Happy to review.
shemnon (Sun, 19 Dec 2021 05:51:16 GMT):
PR to release 21.10.5 with just the log4j changes - https://github.com/hyperledger/besu/pull/3199
shemnon (Sun, 19 Dec 2021 05:51:34 GMT):
Then a PR to set the version to .6-SNAPSHOT (will need a rebase) https://github.com/hyperledger/besu/pull/3200
shemnon (Sun, 19 Dec 2021 05:52:14 GMT):
the 3199 PR will push a release once merged, consider my posting it a vote to release.
shemnon (Sun, 19 Dec 2021 05:53:33 GMT):
After that (AUS could do this when the come online monday morning)
[ ] Update Homebrew (21.10.4 was missed, we should skip it)
[ ] Update the release tag in besu
[ ] update the besu-docs tag
shemnon (Sun, 19 Dec 2021 05:53:33 GMT):
After that (not as urgent, AUS could do this when the come online monday morning if we get wrapped up in family stuff)
[ ] Update Homebrew (21.10.4 was missed, we should skip it)
[ ] Update the release tag in besu
[ ] update the besu-docs tag
atoulme (Sun, 19 Dec 2021 07:02:32 GMT):
Both approved
macfarla (Sun, 19 Dec 2021 20:27:43 GMT):
Thanks @shemnon - we (AUS) will finish off the release today!
macfarla (Sun, 19 Dec 2021 20:58:56 GMT):
on it!
macfarla (Sun, 19 Dec 2021 21:56:42 GMT):
Release 21.10.5 completed! https://github.com/hyperledger/besu/releases/tag/21.10.5
shemnon (Mon, 20 Dec 2021 03:07:23 GMT):
@rjones This feels tweet worthy to me. The text in #besu looks well worded (https://chat.hyperledger.org/channel/besu?msg=qe5HWzwptNia9Ak2n)
shemnon (Mon, 20 Dec 2021 03:08:00 GMT):
Follow on question, how do we workflow tweet requests? Is there a better person to ping?
madelinemurray (Mon, 20 Dec 2021 10:46:56 GMT):
@matt.nelson.94 - are you the person for tweet requests?
andreolf1 (Mon, 20 Dec 2021 13:14:07 GMT):
[Besu Contributor Call]
Added here the agenda items for the call tomorrow.
Feel free to add your topics in the agenda that you want to discuss. Looking forward https://wiki.hyperledger.org/display/BESU/2021-12-21+Besu+Contributor+Call
andreolf1 (Mon, 20 Dec 2021 13:14:07 GMT):
[Besu Contributor Call]
Added here the agenda items for the call tomorrow.
Feel free to add your topics in the agenda that you want to discuss. Looking forward
https://wiki.hyperledger.org/display/BESU/2021-12-21+Besu+Contributor+Call
grace.hartley (Mon, 20 Dec 2021 14:33:38 GMT):
I updated to add the Incentive Program to the top of the agenda.
grace.hartley (Mon, 20 Dec 2021 14:35:51 GMT):
We will definitely record and a Wiki page sounds reasonable. We will get started on it. It might not be ready for tomorrow's meeting, but will begin building it out on Wiki. Good suggestion!
rjones (Mon, 20 Dec 2021 16:23:10 GMT):
anyone with the keys to the Consensys account has that power
rjones (Mon, 20 Dec 2021 16:27:30 GMT):
hi, @andreolf1 - let's discuss the calendar in here
rjones (Mon, 20 Dec 2021 16:29:51 GMT):
@madelinemurray it looks like @andreolf1 has edited the invite - I'm not going to have a power struggle over choosing a Zoom account, could you two sort that out?
rjones (Mon, 20 Dec 2021 16:30:49 GMT):
This is the current invite: https://lists.hyperledger.org/g/besu/viewevent?repeatid=22223&eventid=1365677&calstart=2021-12-21
rjones (Mon, 20 Dec 2021 16:31:48 GMT):
@grace.hartley ^^^
andreolf1 (Mon, 20 Dec 2021 16:32:29 GMT):
This is correct @rjones
andreolf1 (Mon, 20 Dec 2021 16:32:49 GMT):
Screenshot 2021-12-20 at 17.32.35.png
andreolf1 (Mon, 20 Dec 2021 16:33:04 GMT):
but i guess all good then.
rjones (Mon, 20 Dec 2021 16:33:40 GMT):
well, no.
rjones (Mon, 20 Dec 2021 16:34:10 GMT):
We had moved the call to a different Zoom account with a public recording and now you have moved it back.
andreolf1 (Mon, 20 Dec 2021 16:37:49 GMT):
Ok my bad. As last we could't record i though I added my zoom but i am totally fine to put the other one then (didn't know was changed, was thinking was still the Matt one). Would it be possible to change it back?
andreolf1 (Mon, 20 Dec 2021 16:37:49 GMT):
Ok my bad. As last time we could't record i though I added my zoom but i am totally fine to put the other one then (didn't know was changed, was thinking was still the Matt one). Would it be possible to change it back?
andreolf1 (Mon, 20 Dec 2021 16:37:49 GMT):
Ok my bad. As last time we could't record i though I added my zoom but i am totally fine to put the other one then (didn't know was changed, was thinking was still the Matt one). Would it be possible to change it back? sorry to create confusion
rjones (Mon, 20 Dec 2021 16:43:11 GMT):
I will let someone from Consensys handle this - it's a community meeting
andreolf1 (Mon, 20 Dec 2021 16:43:51 GMT):
Ok thanks @rjones!
andreolf1 (Mon, 20 Dec 2021 16:46:01 GMT):
Thanks!
shemnon (Mon, 20 Dec 2021 16:47:03 GMT):
I requested a hyperledger zoom account be used. We have been having repeated problems getting the recording happening because it's been locked to a single user's account to record. If, for example, ConsenSys Quorum is off on a company-wide retreat in a bad time zone I would still like the other maintainers to be able to run the call.
andreolf1 (Mon, 20 Dec 2021 16:52:44 GMT):
Sounds good @shemnon , do you have the latest zoom URL so I can edit the invite and add the Hyperledger zoom account?
shemnon (Mon, 20 Dec 2021 16:59:54 GMT):
https://chat.hyperledger.org/channel/besu-contributors?msg=qDbti6aChFtnnJ6dn
andreolf1 (Mon, 20 Dec 2021 17:03:11 GMT):
Thanks @shemnon, I edited the invite again with the right zoom link. Sorry again for creating confusion
rjones (Mon, 20 Dec 2021 20:43:40 GMT):
@grace.hartley @madelinemurray @macfarla is the correct time for the meeting 0700 Pacific or 0800 Pacific?
grace.hartley (Mon, 20 Dec 2021 20:45:04 GMT):
8 AM Pacific
rjones (Mon, 20 Dec 2021 20:48:30 GMT):
thank you
macfarla (Mon, 20 Dec 2021 22:08:57 GMT):
DCO issue on a non-main branch prevents main from building. Can we improve on that?
shemnon (Mon, 20 Dec 2021 22:09:11 GMT):
DCO fix on branch `release-21.10.x` `92c401fda2a162d8e9022b3db5f4cd5d0d0a9889` DCO signed into `c3e01e7bb34dd702018ee2678defe73cd1b5c3df`
shemnon (Mon, 20 Dec 2021 22:41:46 GMT):
When Circle CI says "Minor Service Outage" I think they mean "SRE forgot to rotate the docker keys before Christmas" - https://status.circleci.com/
shemnon (Mon, 20 Dec 2021 22:51:02 GMT):
Apologies to CircleCI: looks to be a full docker meltdown. https://status.docker.com/pages/533c6539221ae15e3f000031
macfarla (Mon, 20 Dec 2021 23:11:22 GMT):
argh
shemnon (Mon, 20 Dec 2021 23:30:19 GMT):
Looks to be operational now.
atoulme (Tue, 21 Dec 2021 00:51:51 GMT):
It would be good to document what exactly it is that all this entails and what was promised to the EF.
macfarla (Tue, 21 Dec 2021 03:35:30 GMT):
Does anyone have strong opinions on _how_ we install solc on the CI executor https://github.com/hyperledger/besu/pull/3204
macfarla (Tue, 21 Dec 2021 03:37:37 GMT):
@RoboCopsGoneMad any final thoughts on this one https://github.com/hyperledger/besu/pull/3135
shemnon (Tue, 21 Dec 2021 04:25:49 GMT):
For build simplicity I'm inclined to look at the compiled solidity as the test data, and putting generate methods in the build.gradle to re-generate when new 0.x versions of solidity ship. While it works against the "don't check in generated code or binaries" ethos it increases the variables of the dependent tooling to create the build.
shemnon (Tue, 21 Dec 2021 04:26:50 GMT):
Plus, we could generate an array of compiled solidity versions if that ever becomes an issue. 0.8 introdues ABIv2 but that's benign outside of multidimensional arrays.
macfarla (Tue, 21 Dec 2021 04:27:41 GMT):
Do you mean get rid of the web3j solidity gradle plugin?
shemnon (Tue, 21 Dec 2021 04:28:01 GMT):
yes. check in the .sol files for reference but run solidity and check in the .bin files.
shemnon (Tue, 21 Dec 2021 04:28:35 GMT):
mac won't let me put solc in /usr/bin, so I already have to hack the build for development.
macfarla (Tue, 21 Dec 2021 04:29:07 GMT):
so - essentially - revert this PR? https://github.com/hyperledger/besu/pull/2872
shemnon (Tue, 21 Dec 2021 04:30:11 GMT):
Not a direct revert. I would put the .bin files as separate sources and have a build target that generates them. In theory a run of the tool will result in zero changes since the timestamp is not part of the .bin
shemnon (Tue, 21 Dec 2021 04:30:11 GMT):
Not a direct revert. I would put the .bin files as separate sources and have a build target that generates them. In theory a rerun of the tool will result in zero changes since the timestamp is not part of the .bin
shemnon (Tue, 21 Dec 2021 04:31:00 GMT):
It's a comprimise, but the two alternatives attempted (web3j service dependency, solc local binary) don't seem to be fairing well.
diegoll (Tue, 21 Dec 2021 14:26:49 GMT):
Wouldn't the bundled `solc` work in CircleCI? I tried removing the `executable` and `version` fields and it downloaded version **0.8.9** on my local (I don't know why it couldn't find **0.8.10**) and the artifacts were generated successfully (ABI and BIN)
diegoll (Tue, 21 Dec 2021 15:08:58 GMT):
I'm sorry, is the Community Call happening right now? I'm connected to the link listed here https://wiki.hyperledger.org/display/BESU/2021-12-21+Besu+Contributor+Call. Is that the right one?
shemnon (Tue, 21 Dec 2021 15:11:37 GMT):
In an hour
shemnon (Tue, 21 Dec 2021 15:13:25 GMT):
It _should_ have been at 7pt, because the meetings were originally pinned to UTC to avoid AUS/NA/EUR daylight issuesbut the notes went up with a daylight time in the US, so we are just sticking with that for this meeting.
shemnon (Tue, 21 Dec 2021 15:13:25 GMT):
It _should_ have been at 7pt, because the meetings were originally pinned to UTC to avoid AUS/NA/EUR daylight issues but the agenda went up with a daylight time in the US, so we are just sticking with that for this meeting.
shemnon (Tue, 21 Dec 2021 15:13:32 GMT):
Historicaly it's been pinned to UTC.
diegoll (Tue, 21 Dec 2021 15:15:56 GMT):
gotcha. So, it still says 15:00UTC at the top, but that's not consistent with, let's say, 11AM in NYC which is listed below
shemnon (Tue, 21 Dec 2021 15:16:09 GMT):
yes
grace.hartley (Tue, 21 Dec 2021 15:36:37 GMT):
Sorry for the confusion Diego! Yes, it begins at 11 AM ET, 4 PM UK.
diegoll (Tue, 21 Dec 2021 15:36:53 GMT):
no worries!
atoulme (Tue, 21 Dec 2021 17:53:10 GMT):
@shemnon there was a release of the splunk java logging lib. I need to open a PR with the new version and jfrog artifactory site. Do you have a PR open for this? I looked and didn't see any
grace.hartley (Tue, 21 Dec 2021 18:38:35 GMT):
Hey All -
Thanks to everyone joining for the contributor call today. You can find the notes linked below. The next steps from the call were:
-More detailed document around Incentive Program on Wiki (note we will likely not have this until first week of Jan)
-Conduct an offcycle contributor call
Agenda for next call (potentially off-cycle):
-Incentive program (30 mins) (Cost of running nodes)
-Besu maintainer progress/community concerns (30 mins)
I've kicked off an agenda for the proposed next Contributor call. Options for timing are Option 1: December 30th at 10 AM ET, Option 2: December 30th at 11 AM ET, Option 3: December 31st at 10 AM ET.
For full transparency, almost all of the ConsenSys-side contributors will be out of office next week (because it is a ConsenSys holiday) so only a select few of us will be able to join the call.
grace.hartley (Tue, 21 Dec 2021 18:38:59 GMT):
Notes from today's call: https://wiki.hyperledger.org/display/BESU/2021-12-21+Besu+Contributor+Call
grace.hartley (Tue, 21 Dec 2021 18:39:13 GMT):
Agenda for next contributor meeting (feel free to edit!): https://wiki.hyperledger.org/pages/viewpage.action?pageId=62232351
grace.hartley (Tue, 21 Dec 2021 18:40:33 GMT):
Please reply to the thread with which option you prefer for the next call.
shemnon (Tue, 21 Dec 2021 19:21:14 GMT):
I am available for all. My vote would be for 1 to get things going but 2 or 3 also works for me without any problem.
RoboCopsGoneMad (Tue, 21 Dec 2021 21:22:52 GMT):
nope, that is good to merge
macfarla (Tue, 21 Dec 2021 22:32:22 GMT):
that's how it used to work but when an internal server was down it did not - this fix should prevent that, and when released we can update the plugin https://github.com/web3j/web3j-sokt/pull/8
shemnon (Tue, 21 Dec 2021 22:37:08 GMT):
No open security PRs. We don't have dependabot turned on. And based on what I'm experiencing at Hedera I wouldn't recommend it, very low signal to noise.
macfarla (Tue, 21 Dec 2021 22:46:46 GMT):
I get what you're saying, still thinking on it
macfarla (Tue, 21 Dec 2021 22:47:41 GMT):
in other news - the web3j solidity plugin should get updated with this fix which should improve things https://github.com/web3j/web3j-sokt/pull/8
diegoll (Tue, 21 Dec 2021 23:28:01 GMT):
gotcha!
atoulme (Wed, 22 Dec 2021 06:29:56 GMT):
no worries. I'll open a PR when I get a chance
atoulme (Wed, 22 Dec 2021 06:55:00 GMT):
All work for me as well.
RoboCopsGoneMad (Wed, 22 Dec 2021 15:30:21 GMT):
Same here, all work for me, no opinion on which
RoboCopsGoneMad (Wed, 22 Dec 2021 15:34:18 GMT):
@shemnon Could you elaborate on the poor noise you're seeing at Hedera? Asking because I've seen the opposite. In my experience, dependabot would open PRs to automatically update dependencies when security fixes are released, which was very helpful.
shemnon (Wed, 22 Dec 2021 16:20:13 GMT):
Not all updates are security updates, and not all projects follow semver. So unless our test cases are designed to test the libraries themselves things like performance regressions and flakey interactions are introduced.
shemnon (Wed, 22 Dec 2021 16:20:47 GMT):
I think it's safer to at each quarterly update hand-upgrade each lib to the latests and let the RC cycle work through the bugs.
shemnon (Wed, 22 Dec 2021 16:21:36 GMT):
at least, that's what my experiences have been.
RoboCopsGoneMad (Wed, 22 Dec 2021 16:23:19 GMT):
cool, makes sense. Not being semver makes the whole thing collapse.
grace.hartley (Wed, 22 Dec 2021 16:36:50 GMT):
Okay, I think I'll schedule it for 10 AM ET on Dec. 30th. Thanks
grace.hartley (Wed, 22 Dec 2021 17:13:06 GMT):
Hi All - I have scheduled the off cycle Besu contributor call to be at December 30th at 10 AM ET. Invite can be found here. https://lists.hyperledger.org/g/besu/viewevent?eventid=1367457&calstart=2021-12-30
atoulme (Thu, 30 Dec 2021 16:28:18 GMT):
Why was that message removed?
atoulme (Thu, 30 Dec 2021 16:30:03 GMT):
hey folks, it looks like Sajida removed her message earlier.
Folks, if we use discord or email, or anything, please note - I don't think we should allow deletion. We are in a public forum and we work in the open. At the very least, providing a rationale for deleting a message sounds the least you can do, especially when you don't respond to questions in thread.
atoulme (Thu, 30 Dec 2021 16:48:40 GMT):
The fact this message was removed raises more questions for me. Sajida had claimed she had mobilized operations and worked towards node hosting. What was done? What work was it?
atoulme (Thu, 30 Dec 2021 16:52:08 GMT):
> Hi, regarding the Incentive Program that the EF rolled out here are some clarifications: it is not a grant, it is a multi year program designed to incentivize client teams (organizations) to dedicate time and resources to Ethereum roadmap (testnet participation, spec implementation, testing, R&D etc). Specifically to prepare for the Merge and the upcoming milestones, like Sharding. The 4608 ETH that the EF is prodiving to run the validators will not be available for client teams upfront, it will be used to load the validators but the withdrawal keys will be kept by the Ethereum Foundation and disclosed over several years. However, client teams will be able to directly access the rewards generated by those validators over time, if the team keeps contributing (this will be judged by the EF). The program can not be regarded as a grant (where money is directly available) but it will certainly lead to creation of grants/bounties once it's up and running (post-merge, the validators will be generating rewards). The EF just recently made the information public, but for the past 4 months the team at Consensys has worked on setting up the infrastructure to be able to run those validators safely. The infra is ready to receive validators but the handover has not happened yet. Let me know if you have any questions or remarks on this topic.
Just found I had Sajida’s message saved up. Here it is.
atoulme (Thu, 30 Dec 2021 16:53:55 GMT):
@GarySchulte the 4 months figure mentioned during the call came from above.
atoulme (Thu, 30 Dec 2021 16:58:14 GMT):
A good reminder here that this project will only fail if we let it. I am not about letting it. Please work in the open with all of us. I need commitment from everyone to work together in the open.
shemnon (Thu, 30 Dec 2021 18:18:46 GMT):
As promised on the call:
Proposal #1 - Bicameral proposal voting https://hackmd.io/MzykBNnpSYmySer_WJzi7Q
Proposal #2 - Disbursement Priority Tiers https://hackmd.io/2A9HkGwgRYyq8vOEntRlHA
grace.hartley (Thu, 30 Dec 2021 19:17:58 GMT):
I own that I requested this message be deleted because it failed to provide important context around the program and did not accurately reflect ConsenSys' view around the EF incentive program. Because it lacked some context, I was not comfortable with it remaining posted because it was perpetuating misinformation. But, your feedback is fair, that we should have provided a detailed reply around why messages are deleted and we will do that moving forward.
grace.hartley (Thu, 30 Dec 2021 19:18:32 GMT):
Here are the notes from the today's call. https://wiki.hyperledger.org/pages/viewpage.action?pageId=62232351
atoulme (Fri, 31 Dec 2021 01:46:09 GMT):
That's not clear to me, sorry. What misinformation was it perpetuating? Did the message contain false information?
Didn't ConsenSys work with the EF for the past 4 months? What is the team at ConsenSys that she refers to? Is it a different team? Are nodes at ConsenSys already set up and ready to go like she says? I don't quite follow.
atoulme (Fri, 31 Dec 2021 02:30:44 GMT):
@shemnon I like proposal #2. I put some thoughts on what I understand of the situation so far and proposals for a way forward here: https://hackmd.io/x1eSt7AyTAWgwt2vTPiK-Q
JiriPeinlich (Fri, 31 Dec 2021 10:37:34 GMT):
I find both proposals interesting. What are current HLF member companies?
shemnon (Fri, 31 Dec 2021 11:31:06 GMT):
Review Request: rollback rocksdb version - https://github.com/hyperledger/besu/pull/3221
**caution: auto-commit enabled**
shemnon (Fri, 31 Dec 2021 12:39:25 GMT):
https://www.hyperledger.org/about/members
diegoll (Sat, 01 Jan 2022 18:43:47 GMT):
Happy new year folks. I wish you all have a great one
GarySchulte (Mon, 03 Jan 2022 00:06:09 GMT):
Not a fan of a bicameral setup for a host of reasons, not the least of which is that it presupposes conflict between "organizations" and forces someone to belong to one of them. Seems like an invitation for discord.
GarySchulte (Mon, 03 Jan 2022 00:08:09 GMT):
From the second proposal, I definitely prefer the endowment model. It seems most in-line with the spirit of the grant to support mainnet.
GarySchulte (Mon, 03 Jan 2022 00:17:49 GMT):
I think both proposals sidestep the issue of distribution and who/what constitutes a valid participant. Do you envision only 2 potential organizations which would or could run validators? If there is an implied short list of potential candidate validators we should specify that, conversely if anybody can run a validator there should be some qualifying criteria.
GarySchulte (Mon, 03 Jan 2022 00:22:36 GMT):
I also think it would be best if there was a fixed ratio schedule for rewards, rather than "if there is still funds left over from prior distribution tiers". If operations are an open-ended cost up-to the total revenue from rewards, that will de-incentivize running an efficient operation.
GarySchulte (Mon, 03 Jan 2022 00:30:32 GMT):
generally I think Danno's proposal 2 as an endowment is viable, with clarifications about who participants are and how decisions are made.
GarySchulte (Mon, 03 Jan 2022 00:31:56 GMT):
@shemnon why did we need to roll back rocksdb? I was hoping to bump it to latest once this was merged:
https://github.com/facebook/rocksdb/pull/9254
shemnon (Mon, 03 Jan 2022 01:11:01 GMT):
Karim said a canary node failed in the 22Q1 updates pr
shemnon (Mon, 03 Jan 2022 01:11:34 GMT):
IMHO there is no presupposition. It reflects the current state of things in the Besu Community. There's ConsenSys, and then everyone else.
shemnon (Mon, 03 Jan 2022 01:52:10 GMT):
There is a history of rocksdb updates causing canary issues, so I had considered doing it in it's own pr. We can get again and tweak versions but I think rocksdb versions should be tweaked in isolation.
shemnon (Mon, 03 Jan 2022 01:52:10 GMT):
There is a history of rocksdb updates causing canary issues, so I had considered doing it in it's own pr. We can try again and tweak versions but I think rocksdb versions should be tweaked in isolation.
shemnon (Mon, 03 Jan 2022 05:33:19 GMT):
For Tomorrows Construbtor call I would like to add a discussion about the vision of where Hyperledger Besu is headed. Here's my red-herring as an attempt to seed discussion: https://wiki.hyperledger.org/display/~shemnon/Besu+2022+Vision
macfarla (Mon, 03 Jan 2022 10:49:26 GMT):
Agenda for tomorrow's APAC/AMER timezone contributor call - https://wiki.hyperledger.org/display/BESU/2022-01-04+Besu+Contributor+Call
macfarla (Mon, 03 Jan 2022 10:49:30 GMT):
added to the agenda
grace.hartley (Mon, 03 Jan 2022 15:24:45 GMT):
it was missing some important information , which is why I thought it was perpetuating some misinformation. Some information includes around how the keys will be distributed, conflated the terms of client team vs. ConsenSys, and how the keys would be distributed, and expectations for running validator nodes. I also think it missed information around how the EF wanted this to be a win for the Besu client team as well as the ConsenSys team working on Besu. We are pulling together a detailed overview of the EF incentive program and the work expected that better captures all of the elements of the program to make sure the Besu team has a complete picture of the program. A clear Wiki page rather than a Rocketchat message will help give everyone all of the information needed.
grace.hartley (Mon, 03 Jan 2022 15:28:26 GMT):
I added a recap from the prior two Besu contributor calls given most of the folks on the call won't have been at the prior ones and it would be helpful for everyone to be up to speed.
atoulme (Mon, 03 Jan 2022 17:38:33 GMT):
ok that's good
atoulme (Mon, 03 Jan 2022 17:38:39 GMT):
thanks Grace
RoboCopsGoneMad (Mon, 03 Jan 2022 19:23:25 GMT):
merge question, probably for @GarySchulte ... is TDD a >= comparison or just > ? If my difficulty is == ttd, am I the last PoW block, or the second to last?
RoboCopsGoneMad (Mon, 03 Jan 2022 19:29:08 GMT):
I was also uncompelled by the bicameral proposal, however I found the tiered revenue proposal very useful. That is a good shorthand for outlining where the revenue would have to go.
macfarla (Mon, 03 Jan 2022 21:30:43 GMT):
Thanks @grace.hartley - will you be there to do the recap?
grace.hartley (Mon, 03 Jan 2022 21:37:26 GMT):
Yes, I will plan on facilitating the call as well.
macfarla (Mon, 03 Jan 2022 22:04:42 GMT):
Great! See you there!
rjones (Mon, 03 Jan 2022 22:41:02 GMT):
is that in a little over two hours?
rjones (Tue, 04 Jan 2022 00:49:02 GMT):
It looks like the calls in the future are now set to use a private zoom, which the community (me) doesn't have access to in order to ensure recordings.
macfarla (Tue, 04 Jan 2022 00:57:47 GMT):
yes @rjones ! Well now it's in 3 minutes!
rjones (Tue, 04 Jan 2022 01:13:14 GMT):
https://zoom.us/my/ry.jones?pwd=RTJ1Q0R2aXVoNi8xajZ6SlM4SzVqQT09
rjones (Tue, 04 Jan 2022 01:13:16 GMT):
https://zoom.us/my/ry.jones?pwd=RTJ1Q0R2aXVoNi8xajZ6SlM4SzVqQT09
grace.hartley (Tue, 04 Jan 2022 02:02:44 GMT):
Here are the notes from today's contributor call. There are several action items that I will share here tomorrow. https://wiki.hyperledger.org/display/BESU/2022-01-04+Besu+Contributor+Call
RoboCopsGoneMad (Tue, 04 Jan 2022 15:39:01 GMT):
gm besunians. every log4j patch makes me want to work on a new release process. I'll start collecting thoughts into a wiki doc.
RoboCopsGoneMad (Tue, 04 Jan 2022 15:40:33 GMT):
please forward me anything you've written down, or even random thoughts
RoboCopsGoneMad (Tue, 04 Jan 2022 16:21:44 GMT):
I'm listening to the recorded contribs call from last night, and I think there is a dilemma I need to square. We want immutable metadata, so to me that means embedded in the artifact*. If it is embedded in the artifact, we now have a modified binary which was only changed via the version number. Strictly speaking, that should be fully retested, which is the source of a lot of our release time pain.
RoboCopsGoneMad (Tue, 04 Jan 2022 16:22:24 GMT):
How can we just assign a version number to an already released/tested artifact, and still keep it immutable?
atoulme (Tue, 04 Jan 2022 16:30:21 GMT):
You assign the version number upfront
shemnon (Tue, 04 Jan 2022 16:32:21 GMT):
The number of bugs i've seen relating to detecting version numbers of software within code or in the named artifacts is too high. I've got some war stories. That is why I feel simply changing the version number triggers a full rebuild and test cycle.
shemnon (Tue, 04 Jan 2022 16:32:42 GMT):
Automation breaks in strange ways when the word "-SNAPSHOT" disappers.
atoulme (Tue, 04 Jan 2022 16:34:26 GMT):
that is fair. The Apache process requires to produce a source distro that can be used to rebuild the release. The source distro is what is trusted, not the git repo.
atoulme (Tue, 04 Jan 2022 16:34:59 GMT):
That's kind of a ham handed way to do it, but it works across languages and everything
shemnon (Tue, 04 Jan 2022 16:35:12 GMT):
Hyperledger has no release standards. I kind of view that as a problem.
atoulme (Tue, 04 Jan 2022 16:35:34 GMT):
Once an Apache release is made, the RC is uploaded to SVN and then is voted on, with everyone testing it.
atoulme (Tue, 04 Jan 2022 16:35:41 GMT):
More often than not we find problems there.
atoulme (Tue, 04 Jan 2022 16:36:08 GMT):
We also use Nexus and stage the release in a repo there before pushing to Maven Central
atoulme (Tue, 04 Jan 2022 16:36:28 GMT):
so you can actually target that repo and update your versions and test the libs that way. It's very convenient.
atoulme (Tue, 04 Jan 2022 16:52:31 GMT):
I can take a look at helping with the release process - listening to the call, looks like there was a need for a volunteer
grace.hartley (Tue, 04 Jan 2022 17:33:42 GMT):
Hey all! Following up on the actions from yesterday's contributor call:
1. Will the contributors please provide feedback/comments to Danno's Besu 2022 Vision doc? We will review again at the next contributor call. https://wiki.hyperledger.org/display/~shemnon/Besu+2022+Vision
2. We are considering a rotating schedule for the Contributor call. Can folks provide their feedback on whether they would like to keep the current schedule of one APAC friendly and one EMEA/AMER friendly OR would you like to shift to a rotating schedule with 3 meeting times throughout the month?
3. Team to work through what is the scope of an ideal release solution and then see what solutions (tag based or otherwise) would work. it looks like @atoulme was nice enough to volunteer above but wanted to double check.
atoulme (Tue, 04 Jan 2022 17:34:16 GMT):
Yes, I'm ok with helping with release. I can bring over some hard lessons I learned from Apache Tuweni
RoboCopsGoneMad (Tue, 04 Jan 2022 17:34:46 GMT):
that is awesome, thanks antoinne
atoulme (Tue, 04 Jan 2022 17:35:07 GMT):
I have to think about the vision for Besu 2022, thanks Danno for putting it together, it's a good year vision and we don't need to do it all at once.
grace.hartley (Tue, 04 Jan 2022 17:39:20 GMT):
Next: I think it is appropriate to schedule another off cycle Besu contributor call for Tuesday (1/11) or Wednesday (1/12) of next week to continue to discuss the EF Incentive Program. ConsenSys will be sharing validator node requirements and the Program Overview by the end of this week to inform the discussion. Any time preferences? I'm looking at 12:30 to 1:30 PM ET on 1/11 or 1/12.
RoboCopsGoneMad (Tue, 04 Jan 2022 17:55:50 GMT):
https://wiki.hyperledger.org/pages/resumedraft.action?draftId=62233978&draftShareId=224da658-de45-4b74-8727-7f2c91cc94e4&src=shareui&src.shareui.timestamp=1641318906102
RoboCopsGoneMad (Tue, 04 Jan 2022 18:02:22 GMT):
very much a draft release doc: https://wiki.hyperledger.org/x/ep21Aw
atoulme (Tue, 04 Jan 2022 19:00:24 GMT):
thanks Justin, reading
RoboCopsGoneMad (Tue, 04 Jan 2022 19:34:28 GMT):
in a past life, i had a really good experience with the gradle git plugin, but that was a semver project. has anyone tried using it with calver? I don't see any reason it wouldn't work, since the dev sets major/minor either way...
GarySchulte (Tue, 04 Jan 2022 20:47:26 GMT):
I haven't used the gradle release plugin - what guarantees can it make about the release artifact?
GarySchulte (Tue, 04 Jan 2022 20:48:44 GMT):
Also, who are the release approvers in this case?
RoboCopsGoneMad (Tue, 04 Jan 2022 20:50:06 GMT):
gradle release plugin is kinda old, not sure how active it is. it was an attempt to provide a maven-like release process for builds
RoboCopsGoneMad (Tue, 04 Jan 2022 20:50:19 GMT):
release approvers remains TBD
RoboCopsGoneMad (Tue, 04 Jan 2022 20:50:49 GMT):
i would suggest just supporting present day approval requirements, then improving on that by tightening it up
GarySchulte (Tue, 04 Jan 2022 20:52:34 GMT):
I keep coming back to the fact that both the artifact we produce and the release notes are mutable.
macfarla (Tue, 04 Jan 2022 21:21:06 GMT):
We are going to do a release 21.10.6 today with just the log4j 2.17.1 update https://github.com/hyperledger/besu/commit/a890e427ca96abe08924dba4a05e4918cb473350
macfarla (Tue, 04 Jan 2022 21:29:34 GMT):
LFR https://github.com/hyperledger/besu/pull/3230
macfarla (Tue, 04 Jan 2022 22:40:00 GMT):
Release 21.10.6 complete! This updates log4j to 2.17.1 https://github.com/hyperledger/besu/releases/tag/21.10.6
macfarla (Tue, 04 Jan 2022 22:40:05 GMT):
Homebrew PR LFR https://github.com/hyperledger/homebrew-besu/pull/66
macfarla (Tue, 04 Jan 2022 22:44:13 GMT):
and snapshot version number LFR - https://github.com/hyperledger/besu/pull/3231
macfarla (Tue, 04 Jan 2022 22:50:12 GMT):
and one more to add 21.10.6 to the main changelog https://github.com/hyperledger/besu/pull/3232
macfarla (Tue, 04 Jan 2022 23:03:42 GMT):
AFAIK we don't have a policy on this - it's been done ad-hoc - but for example - when I release and add the download links for 21.10.6 which supersedes 21.10.5 - should I remove the download links for 21.10.5? With the more serious log4j updates we did but this one doesn't seem as serious
shemnon (Tue, 04 Jan 2022 23:24:39 GMT):
My take is it depends on why we rolled the security update. If it's a bug in Besu code I've pulled the download. Library issues are a bit different, as they can reasonably be live patched (until we go to jpms...)
shemnon (Tue, 04 Jan 2022 23:25:15 GMT):
I would be ok with leaving or pulling the 21.10.5 and older ones that haven't been pulled.
atoulme (Tue, 04 Jan 2022 23:38:01 GMT):
What are the requirements for a release here - at Hyperledger, for this project? What are we trying to achieve?
atoulme (Tue, 04 Jan 2022 23:38:55 GMT):
Even if we go to another process, we're bound to go back to why in some time, so it might be worth listing our constraints. I do see quite a few above listed out. Do they all matter? Are they must-have?
macfarla (Tue, 04 Jan 2022 23:52:51 GMT):
As per action items from recent contributor call: proposal to change guidelines on moving maintainers to emeritus status https://github.com/hyperledger/besu/pull/3233 - I think this PR should follow the same "2 weeks or absolute majority" rule
macfarla (Wed, 05 Jan 2022 10:47:56 GMT):
3
macfarla (Wed, 05 Jan 2022 10:48:17 GMT):
^ cat says gm
macfarla (Wed, 05 Jan 2022 10:51:22 GMT):
Some PRs to update dependencies LFR https://github.com/hyperledger/besu/pull/3234 https://github.com/hyperledger/besu/pull/3235 https://github.com/hyperledger/besu/pull/3236
RoboCopsGoneMad (Wed, 05 Jan 2022 14:12:30 GMT):
gm, cat
JiriPeinlich (Wed, 05 Jan 2022 17:18:19 GMT):
Today I finished working on PR that is changing the algorithm of processing nodes when downloading the world state during fast sync. The PR is now ready for review https://github.com/hyperledger/besu/pull/3202
JiriPeinlich (Wed, 05 Jan 2022 17:26:40 GMT):
Tomorrow from around 2:30 UTC I will be releasing Besu. I am new in the company and never made a release before. If you would like to see how we release a new version, feel free to join. I am not going to record it. I will be on this Zoom call: https://consensys.zoom.us/j/93768570437
JiriPeinlich (Wed, 05 Jan 2022 17:26:40 GMT):
Tomorrow from around 2:30 UTC I will be releasing Besu. I am new in the Consensys company and never made a release before. If you would like to see how we release a new version, feel free to join. I am not going to record it. I will be on this Zoom call: https://consensys.zoom.us/j/93768570437
JiriPeinlich (Wed, 05 Jan 2022 17:26:40 GMT):
Tomorrow from around 2:30 UTC I will be releasing Besu. I am new in the Consensys and never made a release before. If you would like to see how we release a new version, feel free to join. I am not going to record it. I will be on this Zoom call: https://consensys.zoom.us/j/93768570437
JiriPeinlich (Wed, 05 Jan 2022 17:26:40 GMT):
Tomorrow from around 2:30 UTC I will be releasing Besu. I am new in the Consensys and never made a release of Besu before. If you would like to see how we release a new version, feel free to join. I am not going to record it. I will be on this Zoom call: https://consensys.zoom.us/j/93768570437
JiriPeinlich (Wed, 05 Jan 2022 17:26:40 GMT):
Tomorrow from around 16:30 UTC I will be releasing Besu. I am new in the Consensys and never made a release of Besu before. If you would like to see how we release a new version, feel free to join. I am not going to record it. I will be on this Zoom call: https://consensys.zoom.us/j/93768570437
JiriPeinlich (Wed, 05 Jan 2022 17:26:40 GMT):
Tomorrow from around ~16:30 UTC~ 14:30 UTC I will be releasing Besu. I am new in the Consensys and never made a release of Besu before. If you would like to see how we release a new version, feel free to join. I am not going to record it. I will be on this Zoom call: https://consensys.zoom.us/j/93768570437
JiriPeinlich (Wed, 05 Jan 2022 17:27:52 GMT):
In case you would want your commit to be cherry-picked to the release branch, let me know before then
JiriPeinlich (Wed, 05 Jan 2022 17:28:20 GMT):
This will be a RC-2 release I think
atoulme (Wed, 05 Jan 2022 17:37:03 GMT):
2:30... AM? PM?
JiriPeinlich (Wed, 05 Jan 2022 17:52:05 GMT):
2:30 PM, sry
JiriPeinlich (Wed, 05 Jan 2022 17:52:54 GMT):
so 16:30 UTC
JiriPeinlich (Wed, 05 Jan 2022 17:52:54 GMT):
so ~16:30 UTC~ ... Apparently no its not! It is 14:30 UTC....
atoulme (Wed, 05 Jan 2022 17:55:14 GMT):
so what you're telling me is that you're planning on a chill, LoFi beats party tomorrow, where you'll be livestreaming your release work, and you just dropped an exclusive alpha link.
https://www.timeanddate.com/worldclock/fixedtime.html?msg=Release+and+beats+-+LoFi+only&iso=20220106T1630&p1=1440&ah=1
atoulme (Wed, 05 Jan 2022 17:55:38 GMT):
^^ gotta sell it dude ;)
shemnon (Wed, 05 Jan 2022 17:56:04 GMT):
That's during a critical work call for me, otherwise I'd totally drop by.
atoulme (Wed, 05 Jan 2022 17:56:39 GMT):
I'll try and make it
JiriPeinlich (Wed, 05 Jan 2022 17:57:01 GMT):
As I noticed previously, when something goes wrong it might take "longer" so feel free to drop in/out as you feel like
atoulme (Wed, 05 Jan 2022 17:57:19 GMT):
I think you're saying there's an after party
shemnon (Wed, 05 Jan 2022 17:57:30 GMT):
I'll drop in if my call doesn't run too far over, it has a tendency to do so.
RoboCopsGoneMad (Wed, 05 Jan 2022 18:25:20 GMT):
lol antoinne is now in charge of marketing
atoulme (Wed, 05 Jan 2022 18:50:04 GMT):
always has been
atoulme (Wed, 05 Jan 2022 18:50:22 GMT):
which is why, well, hmm ok who wants the job
RoboCopsGoneMad (Wed, 05 Jan 2022 18:57:32 GMT):
hard no. twitter gives me hives.
RoboCopsGoneMad (Wed, 05 Jan 2022 20:37:22 GMT):
there is a question in #besu about private networks, I do not know the answer to
macfarla (Wed, 05 Jan 2022 21:08:12 GMT):
16:30 UTC is 2:30am for me so not party time
macfarla (Wed, 05 Jan 2022 22:35:25 GMT):
I thought we used snyk somewhere in our build pipeline to check for security vulnerabilities? Do I have that wrong?
macfarla (Wed, 05 Jan 2022 22:41:19 GMT):
^ and if I do have that wrong - what do we use to test for vulnerabilities in dependencies?
shemnon (Wed, 05 Jan 2022 22:59:08 GMT):
It was not in the build pipeline. I think I only ever took the email alerts and they went to my consensys email. I tried to get others to sign up but I think all that did also left consensys
macfarla (Wed, 05 Jan 2022 23:06:34 GMT):
hmmm ok so if we don't already have it - is snyk a good option or is something else better? this user mentions nexus IQ which is sonatype https://github.com/hyperledger/besu/issues/3229
macfarla (Wed, 05 Jan 2022 23:09:33 GMT):
also snyk says 0 issues https://snyk.io/test/github/hyperledger/besu/?tab=issues
shemnon (Wed, 05 Jan 2022 23:14:44 GMT):
The report must not be running then. My personal account found 50 highs.
shemnon (Wed, 05 Jan 2022 23:15:15 GMT):
I didn't like snyk because it had issues when we overrode transitive dependencies. It's overly tuned for maven.
shemnon (Wed, 05 Jan 2022 23:16:48 GMT):
also, it's been at least 2 years since I took snyk seriously, so it may have improved.
macfarla (Wed, 05 Jan 2022 23:45:24 GMT):
ok thanks will see what thoughts others have
macfarla (Wed, 05 Jan 2022 23:58:21 GMT):
also I did the snyk import a different way and I can see 50 highs
macfarla (Thu, 06 Jan 2022 06:07:16 GMT):
So I may have gone down a bit of a rabbit hole with investigating reported security vulnerabilities. This one
http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2020-29582
In JetBrains Kotlin before 1.4.21, a vulnerable Java API was used for temporary file and folder creation. An attacker was able to read data from such files and list directories due to insecure permissions.
is being reported by the dependency check gradle plugin and according to ./gradlew -q dependencies it's coming from tuweni but when I look there https://github.com/apache/incubator-tuweni/blob/9e4814ee853fedb22da48545b4f7648c758f112f/dependency-versions.gradle#L116 it seems to already have 1.5.2 - but there are a few different kotlin libs at different versions and my time down the rabbit hole is up. @atoulme I will put extra details in this thread, if you can shed any light on this that would be great
macfarla (Thu, 06 Jan 2022 06:08:11 GMT):
this is an extract from ./gradlew -q dependencies
macfarla (Thu, 06 Jan 2022 06:08:20 GMT):
`| | | | +--- org.apache.tuweni:tuweni-bytes -> 2.0.0
| | | | | +--- com.google.guava:guava:27.0.1-jre -> 31.0.1-jre
| | | | | | +--- com.google.guava:failureaccess:1.0.1
| | | | | | +--- com.google.guava:listenablefuture:9999.0-empty-to-avoid-conflict-with-guava
| | | | | | +--- com.google.code.findbugs:jsr305:3.0.2
| | | | | | +--- org.checkerframework:checker-qual:3.12.0 -> 3.19.0
| | | | | | +--- com.google.errorprone:error_prone_annotations:2.7.1 -> 2.10.0
| | | | | | \--- com.google.j2objc:j2objc-annotations:1.3
| | | | | +--- org.connid:framework:1.3.2
| | | | | +--- org.connid:framework-internal:1.3.2
| | | | | \--- org.jetbrains.kotlin:kotlin-stdlib-jdk8:1.4.20
| | | | | +--- org.jetbrains.kotlin:kotlin-stdlib:1.4.20 -> 1.6.10
| | | | | | +--- org.jetbrains:annotations:13.0
| | | | | | \--- org.jetbrains.kotlin:kotlin-stdlib-common:1.6.10
| | | | | \--- org.jetbrains.kotlin:kotlin-stdlib-jdk7:1.4.20
| | | | | \--- org.jetbrains.kotlin:kotlin-stdlib:1.4.20 -> 1.6.10 (*)`
atoulme (Thu, 06 Jan 2022 06:13:44 GMT):
Hey, you're looking at the latest dependencies, but the 2.0 release dependencies are here: https://github.com/apache/incubator-tuweni/blob/v2.0.0-incubating/dependency-versions.gradle
atoulme (Thu, 06 Jan 2022 06:14:40 GMT):
They certainly seem to have this vulnerability. We've been due for a release for a little while. We need to kick a new one soonish. Feel free to post on the dev list of tuweni that you need a release to get this to come along faster.
macfarla (Thu, 06 Jan 2022 06:23:09 GMT):
ok thanks for the clarification. dev list of tuweni - do you mean like a mailing list?
atoulme (Thu, 06 Jan 2022 06:25:56 GMT):
Yes
macfarla (Thu, 06 Jan 2022 06:27:04 GMT):
I will put it on my list
macfarla (Thu, 06 Jan 2022 06:56:34 GMT):
I don't think this fixes anything in particular but I was on a roll https://github.com/hyperledger/besu/pull/3243
macfarla (Thu, 06 Jan 2022 06:58:39 GMT):
And this one - I made a start at upgrading the version of kubernetes client-java but there's breaking changes and I think this is better done by someone who knows something about kubernetes and NAT https://github.com/hyperledger/besu/pull/3239 - @JiriPeinlich maybe?
macfarla (Thu, 06 Jan 2022 07:38:32 GMT):
done!
madfu (Thu, 06 Jan 2022 07:40:50 GMT):
Has joined the channel.
RoboCopsGoneMad (Thu, 06 Jan 2022 12:43:20 GMT):
gm, besu
JiriPeinlich (Thu, 06 Jan 2022 13:56:27 GMT):
Because I am completely stupid I made the calculation wrongly... 2:30PM ies 14:30 not 16:30
JiriPeinlich (Thu, 06 Jan 2022 13:56:27 GMT):
Because I am completely stupid I made the calculation wrongly... 2:30PM is 14:30 not 16:30
JiriPeinlich (Thu, 06 Jan 2022 13:59:04 GMT):
I am just replying here because I wrongly assumed 2:30 PM is 16:30, it is actually 14:30 UTC. So, probably 2 hours before the link Mr. atoulme posted...
JiriPeinlich (Thu, 06 Jan 2022 14:01:09 GMT):
I am starting with the release in 30 minutes
JiriPeinlich (Thu, 06 Jan 2022 14:34:20 GMT):
I am starting with the release now
JiriPeinlich (Thu, 06 Jan 2022 15:19:11 GMT):
44b4f5e91 (HEAD -> release-22.1.0-RC2, xxx) added release notes for 21.10.6 (#3232)
* 70d06d503 Removing deprecated hash variable which was used to cache the transaction (#3110)
* 29baf9692 update log4j 2.17.1 (#3225)
* 81f05ef63 Stream JSON RPC responses to avoid creating big JSON strings in memory (#3076)
* d828295f6 Vert.x upgrade (with tests) (#3135)
* c877f1491 Update to Log4j 2.17.0 (#3209)
* 48abc1c97 CHANGELOG got a bit out of sync. this fixes it (#3205)
* f49e67027 Start external services before we start plugins (#3118)
* 0abb2c195 Update minimum solc version in all contracts (#3203)
* 1ba921753 fix canDeployContractSignedByPlugin acceptance test by retrieving the transaction hash of the contract deployment programmatically. Removed path for solidity executable to make ot work on MacOS as well. (#3195)
* d1c99180d Make 'to' field optional in eth_call method according to the spec (#3177)
* 97e640922 Update CHANGELOG for 22.1.0-RC1 release and prepare it for next RC (#3193)
JiriPeinlich (Thu, 06 Jan 2022 15:21:00 GMT):
These are the commits that we are including in the build
JiriPeinlich (Thu, 06 Jan 2022 15:47:21 GMT):
We're beginning the release process for hyperledger/besu, please don't merge anything to release-22.1.0-RC2 until further notice
atoulme (Thu, 06 Jan 2022 16:32:58 GMT):
Hmm
atoulme (Thu, 06 Jan 2022 16:33:13 GMT):
Then did I just miss the action?
atoulme (Thu, 06 Jan 2022 16:33:43 GMT):
Oh man ok
JiriPeinlich (Thu, 06 Jan 2022 16:34:20 GMT):
still on going! feel free to join
atoulme (Thu, 06 Jan 2022 16:34:33 GMT):
2 hours in?
matkt (Thu, 06 Jan 2022 17:23:59 GMT):
It's now safe to merge into release-22.1.0-RC2 for hyperledger/besu again.
GarySchulte (Thu, 06 Jan 2022 17:46:09 GMT):
probably one of the most painful releases we have done. that was a poster-child for the need to update our release process ;)
atoulme (Thu, 06 Jan 2022 17:46:27 GMT):
listening to the call, I guess I don't think GAR-E should live on. I saw there was a checklist on it while it ran, that could be cool to get inspired from
GarySchulte (Thu, 06 Jan 2022 17:47:02 GMT):
GAR-e is used for a variety of projects actually, so it would live on. But we certainly can stop using it for besu
atoulme (Thu, 06 Jan 2022 17:47:09 GMT):
ya
GarySchulte (Thu, 06 Jan 2022 17:47:12 GMT):
I think besu is the only project that does branch based releases
shemnon (Thu, 06 Jan 2022 17:47:12 GMT):
where
shemnon (Thu, 06 Jan 2022 17:47:22 GMT):
where is the call link?
atoulme (Thu, 06 Jan 2022 17:47:37 GMT):
oh call is done just now, there wasn't much going on
shemnon (Thu, 06 Jan 2022 17:47:39 GMT):
Besu is the only project Adrian doesn't exert control over
shemnon (Thu, 06 Jan 2022 17:48:03 GMT):
So it makes sense the administrative parts differ
atoulme (Thu, 06 Jan 2022 17:48:06 GMT):
lots of staring at the screen and looking at git issues, and other finnicky things
shemnon (Thu, 06 Jan 2022 17:48:16 GMT):
cherry picks?
atoulme (Thu, 06 Jan 2022 17:48:43 GMT):
more like ssh cloning vs http cloning, and access token scope, and machine setup, and ssh-agent setup
shemnon (Thu, 06 Jan 2022 17:48:57 GMT):
so Jiri hadn't run gar-e before?
atoulme (Thu, 06 Jan 2022 17:49:02 GMT):
that took Jiri on a loop to fix and figure out
atoulme (Thu, 06 Jan 2022 17:49:04 GMT):
guess not
JiriPeinlich (Thu, 06 Jan 2022 17:49:27 GMT):
I did not
shemnon (Thu, 06 Jan 2022 17:49:37 GMT):
We should look into moving these to git actions. Just as soon as any of us get free time.
GarySchulte (Thu, 06 Jan 2022 17:49:59 GMT):
cherry picks were ok I believe, but it took a while to recognize that the local git / hub config were an issue
shemnon (Thu, 06 Jan 2022 17:50:05 GMT):
IIRC gar-e is high up-front setup but very low repeated use cost.
RoboCopsGoneMad (Thu, 06 Jan 2022 17:50:16 GMT):
add some thoughts to the wiki page! https://wiki.hyperledger.org/display/BESU/Release+Process+Improvement
shemnon (Thu, 06 Jan 2022 17:50:33 GMT):
I would like to get away from driving releases on individual machines.
RoboCopsGoneMad (Thu, 06 Jan 2022 17:50:34 GMT):
i am between tasks and am ready to work on this, i can make the time
GarySchulte (Thu, 06 Jan 2022 17:50:53 GMT):
most definitely would love to get this off of individual machines
shemnon (Thu, 06 Jan 2022 17:51:33 GMT):
A model I would propose is that every release gets a branch. First commit is the new version number. Build and release from there. Second commit is the SNAPSHOT number. Then that branch is merged into main so the history is preserved.
shemnon (Thu, 06 Jan 2022 17:52:13 GMT):
Also protects against main merges.
GarySchulte (Thu, 06 Jan 2022 17:52:32 GMT):
our CI build time makes that a fairly long process, but if it is entirely hands off then that reduces most of the pain, apart from not being able to merge into main during the process
shemnon (Thu, 06 Jan 2022 17:52:55 GMT):
But if we build all releases off of a release branch then main remains open.
GarySchulte (Thu, 06 Jan 2022 17:53:23 GMT):
agreed
atoulme (Thu, 06 Jan 2022 17:53:23 GMT):
anything that can break will break, I think a requirement for me is that we can't block main
shemnon (Thu, 06 Jan 2022 17:53:24 GMT):
each release branch comes off of main or the maintinence branch
GarySchulte (Thu, 06 Jan 2022 17:53:49 GMT):
we could potentially have merge conflicts if we use a release brach and allow merges into main
GarySchulte (Thu, 06 Jan 2022 17:54:01 GMT):
so there would need to be an escape hatch
shemnon (Thu, 06 Jan 2022 17:54:12 GMT):
In theory it's only the properties file. But then we do a manual merge if it's more.
GarySchulte (Thu, 06 Jan 2022 17:54:19 GMT):
changelog also
shemnon (Thu, 06 Jan 2022 17:54:32 GMT):
Yea, that wold be cleaned up in the merge commit.
atoulme (Thu, 06 Jan 2022 17:57:56 GMT):
another fun approach if you like is to not use a CHANGELOG file
atoulme (Thu, 06 Jan 2022 17:58:05 GMT):
You can compile changelog from commits
atoulme (Thu, 06 Jan 2022 17:58:27 GMT):
I like and dislike this approach. It works well in tight projects
atoulme (Thu, 06 Jan 2022 17:59:12 GMT):
I used convential commits where you use a different tool called gitcz to create your commit message, and it formats your commit just so it's easier on the eyes
atoulme (Thu, 06 Jan 2022 17:59:31 GMT):
then computing CHANGELOG is easy. We used semantic releases with that.
RoboCopsGoneMad (Thu, 06 Jan 2022 17:59:42 GMT):
There is a github action that allows you to tag a commit message for inclusion in changelog
atoulme (Thu, 06 Jan 2022 17:59:57 GMT):
I'm not 100% behind it though, since it forces one commit == one PR in a way
RoboCopsGoneMad (Thu, 06 Jan 2022 17:59:57 GMT):
i have similar mixed feelings to @atoulme 's
atoulme (Thu, 06 Jan 2022 18:00:23 GMT):
I don't think CHANGELOG files are cool
RoboCopsGoneMad (Thu, 06 Jan 2022 18:01:39 GMT):
in my perfect world, a changelog is human readable, and would be derrived from issues.
RoboCopsGoneMad (Thu, 06 Jan 2022 18:01:55 GMT):
commits are the domain of developers. they are a safe space for shitty puns
atoulme (Thu, 06 Jan 2022 18:14:53 GMT):
As a manager in my $dayjob, I kinda like the idea of using issues and milestones in github
atoulme (Thu, 06 Jan 2022 18:15:04 GMT):
it provides more clarity for changelogs.
atoulme (Thu, 06 Jan 2022 18:15:20 GMT):
What that would change is that we would likely require that an issue is filed for a PR
atoulme (Thu, 06 Jan 2022 18:16:02 GMT):
which is a form of gatekeeping, but maybe less intimidating than having folks open a PR, watch CI go red, and then have to go fix a weird CHANGELOG file to add their PR number to it
atoulme (Thu, 06 Jan 2022 18:21:28 GMT):
it's also a different responsibility imo to write release notes, one that is best left to a PM - tie in the release to a trend, put 3 issues together, add color and commentary
atoulme (Thu, 06 Jan 2022 18:24:52 GMT):
Usually that's why we do votes at Apache, not to use this refrain too much but...
Developer does a RC
Folks try it out
There is a vote thread and we all vote on the RC. That's a good time to edit and review release notes.
When the vote is finished, the release is pushed, and we make an email to announce@apache.org with the contents of the release
The whole process is documented here: https://github.com/apache/incubator-tuweni/blob/main/RELEASE.md
atoulme (Thu, 06 Jan 2022 19:59:01 GMT):
@sajz thoughts on the above?
RoboCopsGoneMad (Thu, 06 Jan 2022 20:41:30 GMT):
I agree that requiring an issue before a PR is a form of gatekeeping, I suspect we could work around that. Changelogs could be filled from both issues and developer-tagged commits, to allow devs to submit prs that don't warrant an issue/changelog inclusion, but if they feel an issueless PR should be included in the changelog, tag the commit.
atoulme (Thu, 06 Jan 2022 21:17:37 GMT):
or write the issue after, map back to the commit, make this a bit more flexible
atoulme (Thu, 06 Jan 2022 21:17:55 GMT):
in all cases, this is only going to be as good as the person doing the release notes
sajz (Fri, 07 Jan 2022 12:46:07 GMT):
I think we need to better define the problem we are trying to solve and that will inform how we go about addressing it. I have read the previous points and there is probably value in adding this convo the next contributors call agenda.
sajz (Fri, 07 Jan 2022 12:46:57 GMT):
Hi, sharing here some updates on the Hyperledger wiki.
1. I have added a “*Programs*” page and within it a “*Besu Execution Client Incentive Program*” page with all the details.
https://wiki.hyperledger.org/display/BESU/Besu+Execution+Client+Incentive+Program
https://wiki.hyperledger.org/display/BESU/Programs (maybe in the future we can add “Bounties Program” and other programs we’d like to launch)
2. I also updated the Mainnet items of the Hyperledger Besu *roadmap* following the template that was in place.
It would be good to add the Enterprise (privacy, permissionning etc) related items to the page.
https://wiki.hyperledger.org/display/BESU/Roadmap
I will be updating and adding more content to the hyperledger wiki next week.
shemnon (Fri, 07 Jan 2022 14:39:50 GMT):
Any info on the node size requirements? Will those also land in the wiki?
shemnon (Fri, 07 Jan 2022 14:40:51 GMT):
There was some notable discussion about it on the last contributor call already, that gives context to this proposal.
JiriPeinlich (Fri, 07 Jan 2022 14:53:43 GMT):
The time we spend downloading world state in Besu should be addressed with the snapsync changes Mr. matkt is working on. The other part of the download process that takes lots of time is download of the block headers. It appears that it might be enough to get “online” without downloading the whole block headers history. Maybe it would be enough to just download the last N block headers and then get online and start working, while in the background we would continue downloading the rest of the block headers. I was thinking, maybe have a startup option to specify from block of height H when downloading headers, and then finish the rest of the headers only in parallel to being already online and working. Do you see any strong reason not to allow for this behaviour?
JiriPeinlich (Fri, 07 Jan 2022 14:53:43 GMT):
The time we spend downloading world state in Besu should be addressed with the snapsync changes Mr. matkt is working on. The other part of the download process that takes lots of time is download of the block headers. It appears that it might be enough to get “online” without downloading the whole block headers history. Maybe it would be enough to just download the last N block headers and then get online and start working, while in the background we would continue downloading the rest of the block headers. I was thinking, perhaps have a startup option to specify from block of height H when downloading headers, and then finish the rest of the headers only in parallel to being already online and working. Do you see any reason not to allow for this behaviour?
shemnon (Fri, 07 Jan 2022 14:58:24 GMT):
An implementation of this aligns with the ACD desire for block and header expiry if we add two params, one number of headers and bodies download before going "live" (most recent X, likely aligned to a boundary) and one listing a number to download before we stop, which right now is functionally is zero.
shemnon (Fri, 07 Jan 2022 14:59:18 GMT):
Another tricky part is our download starts at zero and grows up. This is a leftover design artifact from when we had no fast syncs and only had full sync. "Backward" block downloading would be the biggest architectural change.
JiriPeinlich (Fri, 07 Jan 2022 15:00:52 GMT):
I was recently working on Backward sync kind of implementation that we merged for the kinsugi, I was thinking about leveraging that.
shemnon (Fri, 07 Jan 2022 15:01:23 GMT):
nice
JiriPeinlich (Fri, 07 Jan 2022 15:09:45 GMT):
It will still need some improvements for this use case. If I understand what you are saying correctly: First parameter determines starting block. From there, we first sync forward to be online as soon as possible. Then we start syncing backwards until we reach the block determined by the second parameter.
shemnon (Fri, 07 Jan 2022 15:11:45 GMT):
I was think both halves would be backward, although direction doesn't matter for the online-requred portion. We start at the current head block. Then we sync backwards. Once we pass the first parameter the node can switch to online. Then we continue backwards sync until we hit the stop depth.
shemnon (Fri, 07 Jan 2022 15:12:28 GMT):
We may want to load in a list of chain checkpoints and require that the stop depth be below one of the checkpoints for the particular chain. To validate we are not syncing the wrong chain.
shemnon (Fri, 07 Jan 2022 15:13:17 GMT):
That was always the argument for downloading all headers, you would validate against genesis. By putting in checkpoints we are making assertions those are off of the genesis chain. If a user doesn't trust our checkpoints their historical block downloads should go to zero.
JiriPeinlich (Fri, 07 Jan 2022 15:14:33 GMT):
list of checkpoints - do we already have something like this?
shemnon (Fri, 07 Jan 2022 15:14:53 GMT):
Nope. We have a quasi check for the classic fork.
shemnon (Fri, 07 Jan 2022 15:15:18 GMT):
So we would need a generic structure for it. The classic validation is loaded into the validations as a special case.
JiriPeinlich (Fri, 07 Jan 2022 15:16:22 GMT):
Ok, that makes sense, thank you
JiriPeinlich (Fri, 07 Jan 2022 15:17:46 GMT):
Would trying it with clasic start block but mainnet endblock and vise versa made a good test if the validation of being on a proper chain works?
JiriPeinlich (Fri, 07 Jan 2022 15:18:13 GMT):
or is clasic and mainnet already too different?
shemnon (Fri, 07 Jan 2022 15:18:39 GMT):
That would be a good test, as they share the network number. Most chains signal their chainID as their networkID, but not classic and mainnet.
shemnon (Fri, 07 Jan 2022 15:19:11 GMT):
However one peer validation that is hard loaded is to query the classic fork block and require the hash to be correct. I would expect peers to do that to us too.
JiriPeinlich (Fri, 07 Jan 2022 15:19:39 GMT):
hmm, makes sense
JiriPeinlich (Fri, 07 Jan 2022 15:20:38 GMT):
I am still on collecting thoughts phase on it now, but I will work on this as my next task probably
atoulme (Fri, 07 Jan 2022 16:56:02 GMT):
@RoboCopsGoneMad FYI would your wiki page help here?
atoulme (Fri, 07 Jan 2022 16:58:08 GMT):
@sajz do you want to add to the next contributors call agenda please?
sajz (Fri, 07 Jan 2022 18:19:48 GMT):
sure, just added it to the 2022-01-18 BCC agenda.
grace.hartley (Fri, 07 Jan 2022 19:50:54 GMT):
Hey All,
I wanted to provide a quick update to the Besu community before heading into the weekend. There were a few action items that I wanted to follow up on:
1. Share the details of the EF's Incentive program on a Wiki page: Sajida drafted up an overview of the program. It can be found here. https://wiki.hyperledger.org/display/BESU/Besu+Execution+Client+Incentive+Program
2. Share the Validator Nodes requirements: I just published the recommendations here. The ConsenSys team pulled together its high-level requirements. We are happy to have a conversation about this with folks who are interested. https://wiki.hyperledger.org/pages/viewpage.action?pageId=62234516
3. ConsenSys team to review Danno and Antoine's proposals regarding decentralizing management of funds. Several of the ConsenSys maintainers reviewed the proposals. I also reviewed. We've come up with a revised proposal to try to address our near-term needs while giving us time to fully formulate a long-term solution. We tried to incorporate elements from the other proposals and also wanted to consolidate our feedback into one proposal rather than sharing disjointed comments. https://wiki.hyperledger.org/display/BESU/Near-Term+Besu+Incentive+Program+proposal
4. Community Practices Improvements: Apologies, I have been out sick a few days this week so I haven't made as much progress on implementing some of these as I hoped this week. I will be following up on them early next week. However, Sajida has gone ahead and published a draft Besu Roadmap for 2022, which is a great start. https://wiki.hyperledger.org/display/BESU/Roadmap
Next Contributor Call: I am going to schedule an off-cycle Contributor call for either Tuesday or Wednesday of next week. Please let me know if you have feedback on timing. Looking at 12:30 PM ET on either day. I would like to propose this meeting be focused on reviewing the Near-Term Proposal I shared above around the Incentive Program.
Have a good weekend!
Grace
atoulme (Fri, 07 Jan 2022 20:07:01 GMT):
Thanks Grace, I hope you get better soon
atoulme (Fri, 07 Jan 2022 20:11:03 GMT):
@grace.hartley I placed comments on the near-term proposal
shemnon (Fri, 07 Jan 2022 21:28:52 GMT):
I have a lot of thoughts. I'll post it as a reply to the email and link the listserve link here.
diegoll (Fri, 07 Jan 2022 21:41:43 GMT):
Hello folks, ETC has decided on blocks for the upcoming upgrade of its different networks in the [ECIP-1104](https://ecips.ethereumclassic.org/ECIPs/ecip-1104). I implemented it in this PR https://github.com/hyperledger/besu/pull/3256. We have already tested private networks with the other client (core-geth) and they connect and sync flawlessly before and after the activation
diegoll (Sat, 08 Jan 2022 12:37:58 GMT):
Thanks for the approval @Antoine. What's the next step? Who should merge it? Is it ok if I squash+merge it myself or should I wait for someone else to do it?
atoulme (Sat, 08 Jan 2022 20:15:21 GMT):
You should merge it
atoulme (Sat, 08 Jan 2022 20:15:39 GMT):
It’s your pr and you can squash it if you like
timbeiko (Sat, 08 Jan 2022 22:33:43 GMT):
One small comment re: what it takes to run nodes
timbeiko (Sat, 08 Jan 2022 22:33:54 GMT):
(Woah, I don't think I should still be a moderator here!)
timbeiko (Sat, 08 Jan 2022 22:34:16 GMT):
Back to the nodes: I think you want to test, either on the canary validators, but at the _very_ least on testnets, that Besu works with multiple CL clients
timbeiko (Sat, 08 Jan 2022 22:35:02 GMT):
If there is an issue on the CL side, we do expect EL teams to be able to switch clients. You don't want the first time you run through that process to be while a major issue is happening on mainnet.
timbeiko (Sat, 08 Jan 2022 22:35:38 GMT):
I suspect this doesn't change "ongoing" costs much, but it will probably change setup efforts, given you need to make sure you have a secure setup with 2+ CL clients.
shemnon (Sat, 08 Jan 2022 22:52:37 GMT):
timbeiko
shemnon (Sat, 08 Jan 2022 22:53:17 GMT):
Changed. Generally speaking maintainers get moderator. If this does not match anyone's status DM me and I'll adjust.
timbeiko (Sat, 08 Jan 2022 22:53:48 GMT):
Might be worth double-checking the emeritus list just to be sure.
diegoll (Sun, 09 Jan 2022 11:49:00 GMT):
Done, apparently squash+merge is the only enabled option
shemnon (Sun, 09 Jan 2022 13:19:43 GMT):
Either day is fine with me. Prefer Tuesday to do it sooner.
shemnon (Mon, 10 Jan 2022 05:08:30 GMT):
Like I said, I have lots of thoughts about the proposal on Friday, and here are some (but not all) of them on a wiki page - https://wiki.hyperledger.org/display/~shemnon/Response+to+Consensys%27+First+Incentive+Proposal
atoulme (Mon, 10 Jan 2022 06:06:43 GMT):
That's a very good write-up, and I agree with its points. I commented on the near term proposal myself with some similar points. I am excited for the great development this can bring to the project. Remember, it will only fail if we let it :)
RoboCopsGoneMad (Mon, 10 Jan 2022 13:46:24 GMT):
what is this snyk security scan check that has appeared on PRs? how do I get added to whatever org I need to to see the results?
shemnon (Mon, 10 Jan 2022 15:00:39 GMT):
I was working with @macfarla to try and see where the old snyk reports went. Looks like in that process snyk.io "helpfully" added a webhook without any direct prompting.
shemnon (Mon, 10 Jan 2022 15:00:45 GMT):
I hate snyk for one more reason now.
RoboCopsGoneMad (Mon, 10 Jan 2022 15:01:10 GMT):
oh, so this is something we used to have enabled?
shemnon (Mon, 10 Jan 2022 15:01:36 GMT):
It never had build integration before.
RoboCopsGoneMad (Mon, 10 Jan 2022 15:01:45 GMT):
Am willing to act on the results in the PR that got flagged, if I can see them
shemnon (Mon, 10 Jan 2022 15:01:48 GMT):
and if it does gain build interation it shouldn't be via my personal github account
shemnon (Mon, 10 Jan 2022 15:07:27 GMT):
My main concern with adding snyk integration is that these should not gate individual PRs unless the finding is caused by the PR. Most findings are prompted by external events (jackson getting yet another CVE, for example).
shemnon (Mon, 10 Jan 2022 15:08:50 GMT):
So really these would gate releases, but without some "burn in" time a library upgrade is risky.
RoboCopsGoneMad (Mon, 10 Jan 2022 15:09:16 GMT):
Do: prevent devs from introducing new vulns
Don't: require devs to fix unrelated but new vuln "while they are in there"
RoboCopsGoneMad (Mon, 10 Jan 2022 15:09:28 GMT):
sounds reasonable
RoboCopsGoneMad (Mon, 10 Jan 2022 16:06:47 GMT):
Our quality profile in sonarcloud has been changed from The Besu Way back to Sonar way. Was that done on purpose?
RoboCopsGoneMad (Mon, 10 Jan 2022 16:09:47 GMT):
our gates are correct (80% test coverage, don't scan for duplicates)
RoboCopsGoneMad (Mon, 10 Jan 2022 16:10:33 GMT):
but our rules are not being enforced, and it looks like they never have, so I'm not gonna touch it right now. Just know that we are still operating with the default Sonar Way rules.
RoboCopsGoneMad (Mon, 10 Jan 2022 16:11:30 GMT):
Here is the Besu Way rules: https://sonarcloud.io/organizations/hyperledger/quality_profiles/show?language=java&name=Besu+way
shemnon (Mon, 10 Jan 2022 18:55:45 GMT):
I thought the decision was sonar was advisory and not gating for commits?
shemnon (Mon, 10 Jan 2022 18:57:15 GMT):
Also, besu way lacks duplicate block detection. Duplicate code blocks are a key aspect to how performance was improved in the EVM.
RoboCopsGoneMad (Mon, 10 Jan 2022 19:51:08 GMT):
yes, correct, sonar is not gating. we are also skipping duplicate block detection.
RoboCopsGoneMad (Mon, 10 Jan 2022 19:52:55 GMT):
it's really the ruleset that i'm trying to clarify. we have a set of rules that is based on the Sonar Way but has 11 deactivated
RoboCopsGoneMad (Mon, 10 Jan 2022 19:53:08 GMT):
thing is, that ruleset isn't being applied
RoboCopsGoneMad (Mon, 10 Jan 2022 20:24:23 GMT):
anyhoo, am asking a bunch of sonarqube questions because I was writing this: https://wiki.hyperledger.org/display/BESU/Usage+of+SonarQube
diegoll (Mon, 10 Jan 2022 22:35:05 GMT):
I started to think about migrating to slf4j using logback as the backend, but now I'm seeing this feature of log4j2, [Garbage-free logging](https://logging.apache.org/log4j/2.x/manual/garbagefree.html) and I'm wondering if it's worth to take the risk of potentially increasing the memory footprint of the node b/c of this migration. I couldn't find any reference to a feature like that in the logback docs. Would it make sense to migrate to slf4j and keep using log4j2 as the backend?
atoulme (Mon, 10 Jan 2022 22:35:59 GMT):
it's definitely a possibility
atoulme (Mon, 10 Jan 2022 22:36:18 GMT):
as we're moving towards modules, we'd want that anyway, so people can plug whatever logging backend they want
diegoll (Mon, 10 Jan 2022 22:36:53 GMT):
ok, cool. I'll start the migration and then we can decide which implementation we include by default in the distribution
atoulme (Mon, 10 Jan 2022 22:53:36 GMT):
awesome!
atoulme (Tue, 11 Jan 2022 07:04:56 GMT):
@grace.hartley just remembered you posted this:
> Next Contributor Call: I am going to schedule an off-cycle Contributor call for either Tuesday or Wednesday of next week. Please let me know if you have feedback on timing. Looking at 12:30 PM ET on either day. I would like to propose this meeting be focused on reviewing the Near-Term Proposal I shared above around the Incentive Program.
Is that still on? I'm super busy and might not be able to make it. Please let us know, hope you're feeling better.
macfarla (Tue, 11 Jan 2022 08:07:47 GMT):
We have been asked by a customer to release besu 21.10.7 with further dependency upgrades (thanks to log4j-related scrutiny). I have cherry-picked dependency upgrades onto this branch. I still want to have another look tomorrow but I think it's what we need for 21.10.7 (might have to exclude a transitive dep for okhttp) https://github.com/hyperledger/besu/pull/3265
macfarla (Tue, 11 Jan 2022 08:09:22 GMT):
I have this in my cal for 5am tomorrow morning my time ~11h from now. https://lists.hyperledger.org/g/besu/viewevent?eventid=1386059
diegoll (Tue, 11 Jan 2022 12:15:59 GMT):
Hello Sally, would it be possible to include [b964e90](https://github.com/hyperledger/besu/commit/b964e9098f2c95faca7a65f0a3447d7cd2825f73) too? I don't know if this fits the criteria for a minor release
RoboCopsGoneMad (Tue, 11 Jan 2022 13:45:32 GMT):
gm besunians
atoulme (Tue, 11 Jan 2022 14:24:33 GMT):
Thanks Sally looking
grace.hartley (Tue, 11 Jan 2022 16:46:52 GMT):
I have scheduled the next Contributor Call to be at 2 PM ET today. Here is the wiki agenda. See ya'll there! https://wiki.hyperledger.org/display/BESU/2022-01-11+Besu+Contributor+Call+-+Off-Cycle
macfarla (Tue, 11 Jan 2022 22:45:15 GMT):
My first thought is that belongs in the RC - but who is it that needs it?
shemnon (Tue, 11 Jan 2022 22:47:45 GMT):
When is Mystique supposed to activate?
macfarla (Tue, 11 Jan 2022 23:22:02 GMT):
https://github.com/ethereumclassic/ECIPs/blob/master/_specs/ecip-1104.md says
`This document proposes the following blocks at which to implement these changes in the Classic networks:
5_520_000 on Mordor Classic PoW-testnet (Jan 13th 2022)
5_578_000 on Kotti Classic PoA-testnet (Jan 23th 2022)
14_525_000 on Ethereum Classic PoW-mainnet (Feb 13th 2022)`
diegoll (Wed, 12 Jan 2022 01:46:03 GMT):
Exactly. Core-geth already released their Mystique enabled version and I built a custom Besu from `master` to use as a canary for Mordor. It would be great if we can have the release before any activation but I made the PR too close to the block of Mordor (it took us a while the inter client private test) and given that the great great majority of Mordor is still core-geth, it would be probably fine if we wait for the upcoming RC (is there a tentative date for it?)
macfarla (Wed, 12 Jan 2022 02:56:53 GMT):
the timing of the next RC (or will we be ready for 22.1.0?) that's dependent on feature work the mainnet team are doing - @GarySchulte or @SajidaZ can you shed any light on this?
macfarla (Wed, 12 Jan 2022 08:22:29 GMT):
FYI 21.10.7 build is all green except for sonar. https://github.com/hyperledger/besu/pull/3265
I'm waiting on more info from the customer (emailed) before I do the release - they have requested earlier versions of some dependencies and I want more context before we go rolling things back. I'm looking at output from snyk and the owasp gradle plugin and can't see the vulnerabilities they are highlighting eg they have asked for Final commons-codec: commons-codec:20041127.091804 which is a really old version https://mvnrepository.com/artifact/commons-codec/commons-codec
I also highlighted the CVE with tuweni which doesn't have an updated release yet - and the customer's report did not flag.
macfarla (Wed, 12 Jan 2022 08:22:56 GMT):
@atoulme any update on the tuweni release front?
atoulme (Wed, 12 Jan 2022 08:23:19 GMT):
No, did you send the email Sally?
macfarla (Wed, 12 Jan 2022 08:24:28 GMT):
sure did https://www.mail-archive.com/dev@tuweni.apache.org/msg02269.html
macfarla (Wed, 12 Jan 2022 08:24:48 GMT):
anything else I can do?
atoulme (Wed, 12 Jan 2022 08:26:54 GMT):
completely missed that email, sorry
atoulme (Wed, 12 Jan 2022 08:27:10 GMT):
Well reminding me should help
atoulme (Wed, 12 Jan 2022 08:27:18 GMT):
I'll get a RC out
atoulme (Wed, 12 Jan 2022 08:27:26 GMT):
sorry about that
macfarla (Wed, 12 Jan 2022 08:27:55 GMT):
thanks!
sajz (Wed, 12 Jan 2022 11:39:25 GMT):
Yes looking into it
diegoll (Wed, 12 Jan 2022 11:42:12 GMT):
thanks!
shemnon (Wed, 12 Jan 2022 13:42:44 GMT):
Can the reasons for the rollback be made public? It feels weird doing a special release for "a customer" with opaque requirements when the ETC fork is being delayed into the RC branch.
shemnon (Wed, 12 Jan 2022 13:42:44 GMT):
Can the reasons for the version rollbacks be made public? It feels weird doing a special release for "a customer" with opaque requirements when the ETC fork is being delayed into the RC branch.
macfarla (Wed, 12 Jan 2022 16:54:34 GMT):
agree. I've reverted the rollback so 21.10.7 is back on 1.39.0 and requesting scan report from customer - they are using Nexus IQ - because I can't see any serious vulnerabilities with it on snyk or owasp
macfarla (Wed, 12 Jan 2022 16:54:58 GMT):
they can scan the snapshot version so I've changed version back to snapshot and set auto-merge
macfarla (Wed, 12 Jan 2022 17:39:19 GMT):
FYI https://hyperledger.jfrog.io/ui/native/besu-binaries/besu/21.10.7-SNAPSHOT/
macfarla (Thu, 13 Jan 2022 00:21:18 GMT):
Emeritus change is sitting at 8 approvals and 8 days https://github.com/hyperledger/besu/pull/3233
lucassaldanha (Thu, 13 Jan 2022 00:42:39 GMT):
Has anyone encountered this issue when building Besu before?
> Source file requires different compiler version (current compiler is 0.6.12+commit.27d51765.Darwin.appleclang)
What is the specific version that we require? I didn't find anything mentioning that our build depends on Solidity in our build from source page (https://wiki.hyperledger.org/display/BESU/Building+from+source).
macfarla (Thu, 13 Jan 2022 01:58:57 GMT):
we recently changed to
pragma solidity ^0.6.0;
but your version satisfies that
macfarla (Thu, 13 Jan 2022 01:59:42 GMT):
actually
macfarla (Thu, 13 Jan 2022 01:59:44 GMT):
.//acceptance-tests/tests/contracts/SimpleStorage.sol:pragma solidity >=0.7.0 <0.9.0;
macfarla (Thu, 13 Jan 2022 02:00:45 GMT):
and that commit has my name on it...
macfarla (Thu, 13 Jan 2022 02:03:44 GMT):
DCO broken on non-main branch prevents main branch from building. @rjones is there anything we can do about this? eg https://app.circleci.com/pipelines/github/hyperledger/besu/12679/workflows/b324a5e5-88d9-437a-86de-f68762230f1b/jobs/69954
shemnon (Thu, 13 Jan 2022 02:19:00 GMT):
I think anyone can force-push to the merge line.
shemnon (Thu, 13 Jan 2022 02:30:26 GMT):
I rolled the merge branch back 1 commit. The commit can be seen here - https://github.com/shemnon/besu/commit/11b5f95dd5ef51af398f8b343b266debadd6f0b9
shemnon (Thu, 13 Jan 2022 02:32:00 GMT):
@RoboCopsGoneMad Please amend the commit to add proper dco and re-push.
macfarla (Thu, 13 Jan 2022 03:59:40 GMT):
right. refreshing my memory on this https://github.com/hyperledger/besu/pull/3203 - in order to support later solc versions (0.8), we had to make updates to the contracts which mean they need at least 0.7 to compile.
macfarla (Thu, 13 Jan 2022 04:00:04 GMT):
can you raise a docs ticket to add required min solc version @lucassaldanha ?
macfarla (Thu, 13 Jan 2022 04:00:59 GMT):
thanks @shemnon - should we add a note to the DCO wiki page for the next time this happens?
macfarla (Thu, 13 Jan 2022 04:02:49 GMT):
or - we can go back to depending on the gradle plugin bundled solc, assuming that they fix their dependency on an internal server to get the version info
atoulme (Thu, 13 Jan 2022 06:58:18 GMT):
The Tuweni RC is out. This Maven repo has 2.1.0: https://repository.apache.org/content/repositories/orgapachetuweni-1022
atoulme (Thu, 13 Jan 2022 06:58:43 GMT):
I'm going to test the RC with Besu.
macfarla (Thu, 13 Jan 2022 07:00:06 GMT):
LFR - getting rid of the last of the orion dependencies https://github.com/hyperledger/besu/pull/3275
macfarla (Thu, 13 Jan 2022 07:19:07 GMT):
added some extra info here about reverting commit on non-main branch in case it helps anyone https://wiki.hyperledger.org/display/BESU/DCO
RoboCopsGoneMad (Thu, 13 Jan 2022 13:35:36 GMT):
ah, thanks @shemnon and @macfarla. that makes more sense now.
RoboCopsGoneMad (Thu, 13 Jan 2022 13:41:57 GMT):
well this is a new and exciting way to break main. can't wait to learn why.
RoboCopsGoneMad (Thu, 13 Jan 2022 13:42:02 GMT):
gm besu
RoboCopsGoneMad (Thu, 13 Jan 2022 15:48:35 GMT):
welp, that'd do it: https://github.com/hyperledger/besu/blob/2a72a3b51c5ebbb977f269a2cc6ef6f8d5af7bef/scripts/dco_check.sh#L7 seems pretty easy to fix, anyone got a reason this loop should be in here instead of just checking the current branch/pr being built?
shemnon (Thu, 13 Jan 2022 15:51:23 GMT):
(a) we would want `main` and `.*release.*` to be in the list
(b) long lived development branches are problematic. The merge is a special case making the pain worth it, but we should take actions to minimize the impact.
RoboCopsGoneMad (Thu, 13 Jan 2022 18:02:51 GMT):
I think I may not be understanding the value of scanning other branches than the one being built, when looking for DCO violations.
shemnon (Thu, 13 Jan 2022 18:16:28 GMT):
The only advantage I see is when a release or main branch has a DCO failure the world stops until it is fixed. Otherwise other lines may sit DCO broken without notice.
shemnon (Thu, 13 Jan 2022 18:16:50 GMT):
LinuxFoundation is crazy-committed to DCO at the moment.
RoboCopsGoneMad (Thu, 13 Jan 2022 18:17:43 GMT):
huh. ok well this did accomplish a worldstop, let me think if there is a more proper way to do that.
RoboCopsGoneMad (Thu, 13 Jan 2022 18:18:01 GMT):
Do I dare open a can of worms re:DCO?
RoboCopsGoneMad (Thu, 13 Jan 2022 18:18:26 GMT):
why not use crypto signed commits?
GarySchulte (Thu, 13 Jan 2022 18:56:57 GMT):
seems to me we already have the DCO check gating the PR, can we apply the same requirement for the merge action to succeed?
GarySchulte (Thu, 13 Jan 2022 18:57:24 GMT):
I would wager that it is vanishingly rare that a committer would not intend to sign off or fix it if they didn't know it was an issue
GarySchulte (Thu, 13 Jan 2022 18:58:40 GMT):
ideally we don't stop the world, we just stop the committer
shemnon (Thu, 13 Jan 2022 19:11:25 GMT):
> Do I dare open a can of worms re:DCO?
> why not use crypto signed commits?
You would need to go to the Hyperledger Governing Council, and they would need to go to battle with the LF. There's better battles to be waged.
shemnon (Thu, 13 Jan 2022 19:11:46 GMT):
DCO is more than an identity, you are certifying it's "open source" - https://developercertificate.org/
shemnon (Thu, 13 Jan 2022 19:12:16 GMT):
I blame the SCO lawsuit.
RoboCopsGoneMad (Thu, 13 Jan 2022 19:16:38 GMT):
ok, so we need to basically attest to that for each contribution
RoboCopsGoneMad (Thu, 13 Jan 2022 19:18:35 GMT):
i'm kind of amazed that this single line in my commit message manages to confer that. Ok, so different issue: should it stop the world for everyone, or just block the offender from continuing? Is that a matter that the powers that be will also have a strong opinion on?
GarySchulte (Thu, 13 Jan 2022 19:41:20 GMT):
https://github.com/marketplace/actions/merge-me
GarySchulte (Thu, 13 Jan 2022 19:43:20 GMT):
I am not sure what we could use as contributors to trigger the workflow, but if there was an auto-merge bot that ensured DCO pass and merged with a sign-off friendly commit message, I'd use it
GarySchulte (Thu, 13 Jan 2022 19:45:43 GMT):
labels would be an obvious choice, but IDK if we can or want to restrict the use of labels to contributors
RoboCopsGoneMad (Thu, 13 Jan 2022 20:02:22 GMT):
Another Reminder: could use some more input on a better release process. specifically a description of the bare minimum needed. https://wiki.hyperledger.org/display/BESU/Release+Process+Improvement
RoboCopsGoneMad (Thu, 13 Jan 2022 20:03:41 GMT):
If anyone who has executed a release successfully without using the Gar-E tool from ConsenSys ( @shemnon ? @macfarla ?) could provide input on the simplest possible thing that would work, that would be perfect.
diegoll (Thu, 13 Jan 2022 20:37:45 GMT):
Hello Sajida, do you have an estimate date for the new major release which would include ETC hardfork?
lucassaldanha (Thu, 13 Jan 2022 20:57:14 GMT):
If we could go back to using the plugin that bundles solc we would be at a better place. It is one less step for someone to be able to build Besu.
macfarla (Thu, 13 Jan 2022 22:01:08 GMT):
it did do a worldstop, but in retrospect it was a pretty easy manual fix to unstop the world - and if we let long-running branches continue on with DCO broken, that's going to be a much bigger pain if we then want to merge back into main (which I can attest to from cherry-picking commits for 21.10.7) - rather than someone just having to fix up their most recent commit which is relatively easy. Also an auto-merge bot sounds great!
macfarla (Thu, 13 Jan 2022 22:01:49 GMT):
ok I will create a ticket :)
macfarla (Thu, 13 Jan 2022 22:05:18 GMT):
I have this on my list to do today
macfarla (Thu, 13 Jan 2022 22:14:02 GMT):
https://github.com/hyperledger/besu/issues/3277
shemnon (Thu, 13 Jan 2022 22:29:47 GMT):
> i'm kind of amazed that this single line in my commit message manages to confer that.
What is and is not sufficiently legal also amazes me.
shemnon (Thu, 13 Jan 2022 22:34:22 GMT):
For stuff like merge me, where do the commit messages come from? I don't want to just take the PR description since we have some pro-forma template for first-time filers. I also don't want a smash of the commit chain, since a long-lived PR can have some really crazy commit messages. I would want a squash commit where I can summarize anew what the PR is.
RoboCopsGoneMad (Thu, 13 Jan 2022 22:35:39 GMT):
me too
GarySchulte (Thu, 13 Jan 2022 22:36:59 GMT):
for commit message, seems that a PR title and link would provide sufficient detail
shemnon (Thu, 13 Jan 2022 22:39:03 GMT):
I rarely use github as my UI for doing change archeology, so I still think we need more than the title.
shemnon (Thu, 13 Jan 2022 22:40:22 GMT):
Hyperledger once used Gerrit and self hosted git repos as the main workflow, so there is precedent to change providers, so I would want the data still in the git stream.
shemnon (Thu, 13 Jan 2022 22:40:22 GMT):
Hyperledger once used Gerrit and self hosted git repos as the main workflow, so there is precedent to change providers, so I would want the history self-contained in the git stream.
RoboCopsGoneMad (Thu, 13 Jan 2022 22:54:32 GMT):
i'm pretty sure we can just delete the script. the DCO rules are already applied as a repo check
macfarla (Thu, 13 Jan 2022 23:48:16 GMT):
21.10.7 - customer has scanned the updated SNAPSHOT and are happy. No versions of any dependencies were downgraded. PR to release the final version: https://github.com/hyperledger/besu/pull/3278
We may need to circle back to this one when CPE is provided: Protobuf has a recent vulnerability CVE-2021-22569 released on 10 Jan so CPE is not provided yet
macfarla (Fri, 14 Jan 2022 03:05:32 GMT):
LFR - 21.10.8-SNAPSHOT version https://github.com/hyperledger/besu/pull/3279
macfarla (Fri, 14 Jan 2022 03:21:12 GMT):
Besu 21.10.7 release https://github.com/hyperledger/besu/releases/tag/21.10.7
macfarla (Fri, 14 Jan 2022 03:21:14 GMT):
LFR https://github.com/hyperledger/homebrew-besu/pull/67
macfarla (Fri, 14 Jan 2022 03:21:14 GMT):
LFR homebrew update https://github.com/hyperledger/homebrew-besu/pull/67
macfarla (Fri, 14 Jan 2022 03:21:57 GMT):
LFR adding 21.10.7 to main changelog https://github.com/hyperledger/besu/pull/3280
macfarla (Fri, 14 Jan 2022 03:23:33 GMT):
docs release https://github.com/hyperledger/besu-docs/releases/tag/21.10.7
sajz (Fri, 14 Jan 2022 11:33:55 GMT):
Hi Diego, we can do a dedicated release for ETC hardfork before we do RC3.
I think it is good to have separation of concerns here. We don't want to rush RC3 as RC2 is still pretty recent and we might want to test some improvements in RC3 that are not yet ready (we don't have a date). We recently had 21.10.7 minor release to answer a specific set of concerns for a customer and I don't see why we couldn't do a specific release for ETC hardfork too, so if you want to roll with it, that is fine by me at least.
diegoll (Fri, 14 Jan 2022 12:46:54 GMT):
Awesome, I'll take a look on how a release is performed (probably following the notes from @RoboCopsGoneMad ) and I'll propose a 21.10.8 only adding that commit
RoboCopsGoneMad (Fri, 14 Jan 2022 16:12:54 GMT):
thank you both for your comments
macfarla (Sun, 16 Jan 2022 21:37:24 GMT):
@diegoll you will want to do your PR against the release-21.10.x branch https://github.com/hyperledger/besu/tree/release-21.10.x
diegoll (Sun, 16 Jan 2022 21:40:44 GMT):
@macfarla sure! so I cherry-pick [964e90](https://github.com/hyperledger/besu/commit/b964e9098f2c95faca7a65f0a3447d7cd2825f73) and make a PR against that branch?
diegoll (Sun, 16 Jan 2022 21:40:44 GMT):
@macfarla sure! so I cherry-pick [b964e9098](https://github.com/hyperledger/besu/commit/b964e9098f2c95faca7a65f0a3447d7cd2825f73) and make a PR against that branch?
macfarla (Sun, 16 Jan 2022 21:48:49 GMT):
Yes. You will also need to add a changelog entry for 21.10.8 and update the version in gradle.properties
diegoll (Sun, 16 Jan 2022 21:50:46 GMT):
excellent, I'll
diegoll (Sun, 16 Jan 2022 21:50:46 GMT):
excellent, I'll do it
macfarla (Sun, 16 Jan 2022 21:51:07 GMT):
Re https://wiki.hyperledger.org/display/BESU/Refactor+EVM+into+a+stand+alone+library - The status of this doc is "Proposed" but some of this work is already done. @shemnon has everything on this list been done? Can you update this page to make the status clear? Thanks
diegoll (Sun, 16 Jan 2022 22:07:24 GMT):
https://github.com/hyperledger/besu/pull/3283
macfarla (Sun, 16 Jan 2022 22:58:16 GMT):
LFR https://github.com/hyperledger/besu/pull/3284
macfarla (Sun, 16 Jan 2022 22:58:16 GMT):
LFR - reverting back to bundled solc for gradle plugin - https://github.com/hyperledger/besu/pull/3284
macfarla (Sun, 16 Jan 2022 23:09:29 GMT):
@diegoll my only hesitation with approving this is that it's still Sunday in most parts of the world - I'd like to leave it open for feedback here for 24 hrs, does that work for you?
diegoll (Sun, 16 Jan 2022 23:10:07 GMT):
sure!
macfarla (Sun, 16 Jan 2022 23:12:32 GMT):
LFR Release 21.10.8 will include the ETC Mystique hardfork - I'd like to leave this open for feedback until it's Monday everywhere :) https://github.com/hyperledger/besu/pull/3283
macfarla (Mon, 17 Jan 2022 02:18:00 GMT):
MLK day means US folks won't be online for another 24h.
shemnon (Mon, 17 Jan 2022 03:32:29 GMT):
Reviewed. Approved.
macfarla (Mon, 17 Jan 2022 03:34:35 GMT):
good enough for me
macfarla (Mon, 17 Jan 2022 03:38:28 GMT):
@diegoll once this pipeline is complete, you will have 21.10.8 artefacts https://app.circleci.com/pipelines/github/hyperledger/besu/12724/workflows/fc1b074b-5aeb-4591-ba49-17fbb9fd63c3
diegoll (Mon, 17 Jan 2022 04:21:48 GMT):
That's awesome Sally! Thanks a ton
diegoll (Mon, 17 Jan 2022 04:23:44 GMT):
And thanks @shemnon for the review as well!
macfarla (Mon, 17 Jan 2022 04:42:07 GMT):
LFR adding 21.10.8 to main changelog https://github.com/hyperledger/besu/pull/3287
macfarla (Mon, 17 Jan 2022 04:43:38 GMT):
LFR 21.10.9-SNAPSHOT version https://github.com/hyperledger/besu/pull/3286
macfarla (Mon, 17 Jan 2022 04:47:03 GMT):
Besu 21.10.8 https://github.com/hyperledger/besu/releases/tag/21.10.8
macfarla (Mon, 17 Jan 2022 04:50:25 GMT):
LFR homebrew https://github.com/hyperledger/homebrew-besu/pull/68
macfarla (Mon, 17 Jan 2022 04:50:59 GMT):
docs release https://github.com/hyperledger/besu-docs/releases/tag/21.10.8
macfarla (Mon, 17 Jan 2022 05:12:05 GMT):
Also have added this page to the wiki - prob useful to track upcoming releases and who's doing them as well as how long the release process is taking https://wiki.hyperledger.org/display/BESU/Release+Rotations+2022
shemnon (Mon, 17 Jan 2022 05:41:00 GMT):
A second proposal for the incentive program (very rough draft) https://wiki.hyperledger.org/display/BESU/Second+Incentive+Proposal
andreolf1 (Mon, 17 Jan 2022 19:21:39 GMT):
Re Besu contributor call tomorrow: https://wiki.hyperledger.org/display/BESU/2022-01-18+Besu+Contributor+Call
Feel free to add your topics in the agenda that you want to discuss.
macfarla (Mon, 17 Jan 2022 22:06:56 GMT):
added some info
macfarla (Mon, 17 Jan 2022 22:07:26 GMT):
People are using discussions! https://github.com/hyperledger/besu/discussions
grace.hartley (Tue, 18 Jan 2022 00:30:31 GMT):
For tomorrow's contributor call, the ConsenSys team won't be ready to discuss @shemnon's most recent proposal around the Incentive Program. The US team has been out of office today so we haven't had a chance to give it a detailed review.
Would the group we prefer we go ahead with tomorrow's contributor call and discuss other topics (roadmap, LF metrics, other bodies of work) or would the group prefer we push back the call to later in the week when we will be prepared to discuss?
macfarla (Tue, 18 Jan 2022 02:08:26 GMT):
Nothing urgent from me but also this call is 1am my time so I won't vote either way.
macfarla (Tue, 18 Jan 2022 03:09:33 GMT):
draft PR for a nightly scan of docker images with trivy - one problem is that a notification to a consensys slack channel isn't visible to other hyperledger folks - is there an option to send alerts to RC? Or are we happy to have the report visible in CI - https://github.com/hyperledger/besu/pull/3295
diegoll (Tue, 18 Jan 2022 03:43:34 GMT):
LFR https://github.com/hyperledger/besu/pull/3296
diegoll (Tue, 18 Jan 2022 03:49:09 GMT):
given this seems to be a regression, I'm thinking if it wouldn't make sense to make a `21.10.9` release out of it. This is particularly affecting public nodes of ETC which are running on Besu. Kotti was downgraded b/c it hasn't upgraded yet, but Metamask is unusable for Mordor (already upgraded)
diegoll (Tue, 18 Jan 2022 04:11:46 GMT):
I built a custom version with this patch applied on top of `release-21.10.x` and Metamask is working again
macfarla (Tue, 18 Jan 2022 04:37:58 GMT):
Alternative idea - would enabling dependabot alerts help? I've only used it on JS projects https://docs.github.com/en/code-security/supply-chain-security/managing-vulnerabilities-in-your-projects-dependencies/about-alerts-for-vulnerable-dependencies
macfarla (Tue, 18 Jan 2022 05:00:31 GMT):
That makes sense to me. I've approved the PR. You should be able to draft the 21.10.9 release
RoboCopsGoneMad (Tue, 18 Jan 2022 13:17:29 GMT):
gm besu :coffee:
andreolf1 (Tue, 18 Jan 2022 14:45:09 GMT):
GM
grace.hartley (Tue, 18 Jan 2022 15:01:45 GMT):
Any strong feelings here? Otherwise, I think I will reschedule to Thursday or Friday.
RoboCopsGoneMad (Tue, 18 Jan 2022 15:14:25 GMT):
i've added an outline of our current manual release process: https://wiki.hyperledger.org/display/BESU/Release+Process+Improvement
RoboCopsGoneMad (Tue, 18 Jan 2022 15:15:01 GMT):
could someone who understands homebrew add an explanation of what happens after we update the version and hash in the tap?
RoboCopsGoneMad (Tue, 18 Jan 2022 15:15:10 GMT):
seems like magic to me
RoboCopsGoneMad (Tue, 18 Jan 2022 15:15:30 GMT):
Satus Quo, section 9c.
grace.hartley (Tue, 18 Jan 2022 15:22:20 GMT):
Since there were no strong feelings, I am going to* cancel today's Besu Contributor call *and push to later this week.
grace.hartley (Tue, 18 Jan 2022 15:22:20 GMT):
Since there were no strong feelings, I am going to *cancel today's Besu Contributor call* and push to later this week.
grace.hartley (Tue, 18 Jan 2022 15:26:16 GMT):
I have moved the Contributor call to Thursday at 11 AM ET. Thanks all!
shemnon (Tue, 18 Jan 2022 15:33:14 GMT):
Nothing happens, there is no server, it's all pull based.
When someone who has besu in their tap runs the update command the brew software goes out to that github url and looks for updates. If it sees any then brew does the tasks described in the homebrew repo.
shemnon (Tue, 18 Jan 2022 15:33:54 GMT):
So the update step updates the repo so that when brew users update or upgrade the current release is reflected.
RoboCopsGoneMad (Tue, 18 Jan 2022 15:34:31 GMT):
ty, will add to doc
shemnon (Tue, 18 Jan 2022 15:37:21 GMT):
Build process, 3c and 4b - "PROBLEM: so far, release-like artifacts with any version number are created by the act of opening PRs, not approving them!"
This is not a problem. While the artifacts are generated they only exist on the CI server. The build automation does not push these artifacts out to the servers (jfrog, docker) unless it is (a) building on `main` or `release-*` _and_ (b) building a non`-SNAPSHOT` version.
You need to have CI server access to get to these artifacts so they are not exposed to the world unless published.
shemnon (Tue, 18 Jan 2022 15:40:32 GMT):
When there are bugs that are tripped by version numbers (for example debug code that only activates in `-SNAPSHOT` versions) I've seen such speculative version builds used by QA to run full test external test suites. So it's not something that surprises me.
RoboCopsGoneMad (Tue, 18 Jan 2022 15:58:42 GMT):
but can't anyone create a `release-*` and set the version in gradle properties to whatever they want?
shemnon (Tue, 18 Jan 2022 16:01:23 GMT):
The PR has to complete into the release-* branch, you can't just push a randomly named branch. That and github has branch protection rules that are meant to prevent all but admins from pushing to that branch. (if not we need to update them)
shemnon (Tue, 18 Jan 2022 16:01:44 GMT):
So the webhook triggers on a PR merge. Which still has the must be reviewed requirement
RoboCopsGoneMad (Tue, 18 Jan 2022 16:28:19 GMT):
thx @shemnon i've updated it.
timbeiko (Tue, 18 Jan 2022 20:07:49 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=ajar58onvYCK7yLDq) :wave: a few thoughts about this!
Overall, I like the idea of having equally sized shares to both cover node operations and a "Besu Community Fund". I also like the idea of using maintainers contributing to mainnet over vesting periods in order to split the bulk of the fees+rewards.
The one challenge I can see with this is withdrawal key custody and liquidation of validators. Assuming ConsenSys remains the employer of the majority of Besu mainnet maintainers, the majority of the funds each period will the flow back to them. I think for security + accounting reasons, that probably means you'd want to send withdrawal keys proportionally to CS for each period. For example, if in a vesting period, 8/12 of maintainers of Besu who worked on mainnet are at CS, they should get 2/3rds of withdrawal credentials from that vesting period. If it changes in the next one, the share is updated.
This means that, at that point, CS would have full control over that share of the validators, and hence the ability to liquidate. I think it's probably right/fair/simplest for that right to be fully theirs. For validators whose withdrawal keys have been sent to HLF, a separate process can be used to decide on liquidation (which obviously takes into account that CS may have a large share of 'extra' validators and should avoid double-dipping).
Talked with Grace and Daniela today about all of this, and we will be looking into how this is feasible legally + technically. I won't make the contributor call on Thursday, but Grace should have some updates there.
shemnon (Tue, 18 Jan 2022 20:12:01 GMT):
If validator keys are going to CS, then the other employers who have maintainer should also be able to get the keys by the same process CS receives them. What I want to avoid is a set of rules for CS and a set of rules for everyone else.
shemnon (Tue, 18 Jan 2022 20:14:05 GMT):
> What I want to avoid is a set of rules for CS and a set of rules for everyone else.
If anything is my north star in this process, that would be it.
macfarla (Tue, 18 Jan 2022 21:02:08 GMT):
gm
macfarla (Tue, 18 Jan 2022 21:03:19 GMT):
@diegoll is this release urgent for you guys or are you using your patched version
macfarla (Tue, 18 Jan 2022 21:09:55 GMT):
Another release anyone? Seems the vertx upgrade has changed the behaviour of --rpc-http-cors-origins="*" which previously allowed any protocol but in the new vertx only allows http/https https://github.com/hyperledger/besu/pull/3296 so @diegoll is proposing 21.10.9 with that patch.
RoboCopsGoneMad (Tue, 18 Jan 2022 21:51:33 GMT):
I'm gonna cut a release for 21.10.9 tomorrow to include the CORS fix. Does anyone want to pair with me when I do?
diegoll (Tue, 18 Jan 2022 23:00:03 GMT):
We have patched our servers with the fix, so it's not so urgent for us. I don't know how it could affect private networks though
macfarla (Wed, 19 Jan 2022 03:47:10 GMT):
14 days, no vetos - I am going to hit merge on this https://github.com/hyperledger/besu/pull/3233
macfarla (Wed, 19 Jan 2022 03:47:23 GMT):
is it gonna be a party?
RoboCopsGoneMad (Wed, 19 Jan 2022 16:21:06 GMT):
release 21.10.9 is building here: https://github.com/hyperledger/besu/pull/3305/files
RoboCopsGoneMad (Wed, 19 Jan 2022 16:52:10 GMT):
last call for any bugfixes to be cherrypicked to the point release branch!
RoboCopsGoneMad (Wed, 19 Jan 2022 17:59:15 GMT):
once i have an approver for this I will start the release
RoboCopsGoneMad (Wed, 19 Jan 2022 17:59:16 GMT):
https://github.com/hyperledger/besu/pull/3305
RoboCopsGoneMad (Wed, 19 Jan 2022 18:26:25 GMT):
release 21.10.9 begins now, please refrain from making any commits into branch `release-21.10.x`
RoboCopsGoneMad (Wed, 19 Jan 2022 19:30:11 GMT):
release is complete, you are weapons free
RoboCopsGoneMad (Wed, 19 Jan 2022 19:31:06 GMT):
pr for updating homebrew
RoboCopsGoneMad (Wed, 19 Jan 2022 19:31:08 GMT):
https://github.com/hyperledger/homebrew-besu/pull/69
macfarla (Wed, 19 Jan 2022 22:49:58 GMT):
LFR adds 21.10.9 to main changelog https://github.com/hyperledger/besu/pull/3307
macfarla (Wed, 19 Jan 2022 23:17:55 GMT):
Any feedback on this https://github.com/hyperledger/besu/pull/3295 - if not I'm going to merge it in and monitor what happens :)
RoboCopsGoneMad (Wed, 19 Jan 2022 23:23:37 GMT):
eh, the curling from github and piping results into sh gives me bad feels
RoboCopsGoneMad (Wed, 19 Jan 2022 23:24:26 GMT):
maybe thats ok cuz its a throwaway container? idk
RoboCopsGoneMad (Wed, 19 Jan 2022 23:28:54 GMT):
has this task been run yet? i'm only seeing the default workflow running linked to this pr. i was curious what the output looks like
macfarla (Wed, 19 Jan 2022 23:33:02 GMT):
it hasn't been run yet... I guess I could change it to run on main | release* and push it to the release snapshot branch? merge it over there and see what happens?
RoboCopsGoneMad (Wed, 19 Jan 2022 23:34:10 GMT):
idk how non-pr webhook triggers work in CircleCI, might want @rjones input on this
macfarla (Wed, 19 Jan 2022 23:36:13 GMT):
here's a link to what the trivy scan output for teku looks like - I think it's publicly accessible - https://app.circleci.com/jobs/github/ConsenSys/teku/104672
RoboCopsGoneMad (Wed, 19 Jan 2022 23:37:44 GMT):
cool thanks. so not really a fancy report, which i like. still enough info to be actionable and halt a build.
macfarla (Wed, 19 Jan 2022 23:57:29 GMT):
https://github.com/hyperledger/besu/pull/3308
macfarla (Thu, 20 Jan 2022 01:23:45 GMT):
Re security scanning - another option is to enable "vulnerability scanning" in DockerHub @rjones
macfarla (Thu, 20 Jan 2022 01:23:45 GMT):
Re security scanning - another option is to enable "vulnerability scanning" in DockerHub - @rjones are you able to see what that does for us?
shemnon (Thu, 20 Jan 2022 02:59:19 GMT):
I'm triple booked at that slot. I'll do my best to make it.
shemnon (Thu, 20 Jan 2022 03:03:01 GMT):
an hour or two either side would make a world of difference.
grace.hartley (Thu, 20 Jan 2022 14:46:02 GMT):
Sorry about that. We will send the recording and notes out if you can't make it. We won't be spending too much time at all discussing the Incentive Program proposal because we're still reviewing and next steps are on EF/HLF, but we'll give a brief update.
rjones (Thu, 20 Jan 2022 15:00:01 GMT):
@RoboCopsGoneMad not sure.
rjones (Thu, 20 Jan 2022 19:27:03 GMT):
Hi, @grace.hartley . There is a public place to ask questions about LFX: https://community.lfx.dev/
rjones (Thu, 20 Jan 2022 19:28:00 GMT):
I only just found out about it. I think your questions around "how are these insights graphs calculated" can get answered there
GarySchulte (Thu, 20 Jan 2022 20:17:26 GMT):
As discussed on the HL contrib call earlier, I am planning to release 22.10.0-RC3 `from main` today. I intend to start the release in an hour, let me know if there are concerns with using RC3 as a release from main.
diegoll (Thu, 20 Jan 2022 20:54:31 GMT):
Hello folks, it seems that the `acceptanceTestsQuorum` [are failing](https://app.circleci.com/pipelines/github/hyperledger/besu/12792/workflows/a832e434-f9e7-446a-99bc-88c613a14192/jobs/70750) on `main`. I was trying to spot the problem for [PR#3285](https://github.com/hyperledger/besu/pull/3285) but then I re-run it on main and it failed in the same way. It looks like a problem on the _ethsigner_ image regarding file system permissions
diegoll (Thu, 20 Jan 2022 21:22:58 GMT):
Btw, did anyone tried to run Circle locally through their [CLI](https://circleci.com/docs/2.0/local-cli/)? I tried with `$ circleci local execute --job acceptanceTestsQuorum` but it fails b/c it cannot download the `ubuntu-2004:202101-01` image. Maybe this one is intended to be used only within the cloud environment but I couldn't find anything about it in [the docs](https://circleci.com/docs/2.0/configuration-reference/#available-machine-images)
GarySchulte (Thu, 20 Jan 2022 21:56:30 GMT):
We are going to delay the RC3 release until Friday Morning Pacific while we sort out the ethsigner permissions. @diegoll Usman is working on the ethsigner problem now and will have it sorted one way or the other shortly
diegoll (Thu, 20 Jan 2022 22:08:17 GMT):
awesome! thanks for taking care of it
GarySchulte (Thu, 20 Jan 2022 23:50:51 GMT):
for now the pipeline is changed to use ethsigner:latest rather than ethsigner:develop. I restarted your build from the failed test, 🤞
macfarla (Fri, 21 Jan 2022 00:17:01 GMT):
success!
GarySchulte (Fri, 21 Jan 2022 00:18:44 GMT):
:men_with_bunny_ears_partying: :women_with_bunny_ears_partying:
macfarla (Fri, 21 Jan 2022 06:47:17 GMT):
Github discussions - I propose disabling them. We enabled it as an experiment, but it doesn't seem to be adding a lot of value and it is one more thing to monitor aka drop the ball on. https://github.com/hyperledger/besu/discussions
macfarla (Fri, 21 Jan 2022 06:58:53 GMT):
cc @lucassaldanha
macfarla (Fri, 21 Jan 2022 07:00:51 GMT):
There are a total of 4 discussions since this was enabled in October. One was created by me (converted from an issue), one was a test, and all were answered by me.
macfarla (Fri, 21 Jan 2022 07:01:42 GMT):
I would rather the slight inconvenience of users opening an issue to ask a question and answering it there, rather than have effectively another channel to monitor
macfarla (Fri, 21 Jan 2022 07:02:25 GMT):
Welcome feedback from others though!
macfarla (Fri, 21 Jan 2022 08:44:19 GMT):
LFR adding MockitoJunitRunner to more tests https://github.com/hyperledger/besu/pull/3311
atoulme (Sat, 22 Jan 2022 07:40:11 GMT):
Tuweni 2.1 is out, and this PR upgrades Besu to it: https://github.com/hyperledger/besu/pull/3281
macfarla (Sun, 23 Jan 2022 23:22:09 GMT):
Following up on a couple of things from last week's contributor call...
macfarla (Sun, 23 Jan 2022 23:22:23 GMT):
* trace API work - a number of folks are working on the different methods - we are tracking this one via this (zenhub) epic https://github.com/hyperledger/besu/issues/2884 but it's not totally transparent via github - I just linked all the sub-issues as a workaround. @daniellehrner has a PR for trace_call in draft and we're aiming to reuse his approach there on the others
macfarla (Sun, 23 Jan 2022 23:24:27 GMT):
* approach to updating dependencies - wonder if we should doc this in a wiki page eg after the quarterly release, we raise a PR to update all available dependencies so that they can have maximum time to bake in. Any updates that require significant code changes - raise a separate issue in github and prioritise accordingly
macfarla (Sun, 23 Jan 2022 23:49:57 GMT):
* security scan options - I've looked at a few different things here as well as what other Consensys projects are doing - scanning the docker image seems like a good option. we can use trivy via GitHub actions to scan code and the docker image. this is an example of what the output looks like for ethsigner https://github.com/ConsenSys/ethsigner/security/code-scanning?query=is%3Aopen+autofilter%3Atrue - I'm still trying to make sense of what will be useful for Besu - @rjones are we able to experiment with github actions?
rjones (Mon, 24 Jan 2022 00:10:23 GMT):
@macfarla sure thing. You don’t need my permission to do that 👍
diegoll (Mon, 24 Jan 2022 00:11:44 GMT):
Page Not Found. Probably security outputs are protected
diegoll (Mon, 24 Jan 2022 00:35:18 GMT):
LFR https://github.com/hyperledger/besu/pull/3285
diegoll (Mon, 24 Jan 2022 00:35:18 GMT):
LFR https://github.com/hyperledger/besu/pull/3285 - some maintainers have already been looking at it (thanks!), but it wasn't properly announced here
macfarla (Mon, 24 Jan 2022 00:35:19 GMT):
here's a PR which has a few things in it https://github.com/ConsenSys/ethsigner/pull/427 - you can see github actions report
macfarla (Mon, 24 Jan 2022 00:35:44 GMT):
I guess if we set code-scanning up in besu all maintainers would be able to see the report?
macfarla (Mon, 24 Jan 2022 00:37:20 GMT):
And this ethsigner PR adds a github action for trivy scanning https://github.com/ConsenSys/ethsigner/pull/426
macfarla (Mon, 24 Jan 2022 00:40:07 GMT):
ok cool. What about enabling scanning in dockerhub? Can we turn that on and see what the output is?
rjones (Mon, 24 Jan 2022 00:53:30 GMT):
Sure, but if you used GHCR, it would be easier
macfarla (Mon, 24 Jan 2022 00:55:47 GMT):
GHCR?
rjones (Mon, 24 Jan 2022 01:03:41 GMT):
https://github.com/hyperledger-labs/firefly/pkgs/container/firefly For instance
rjones (Mon, 24 Jan 2022 02:13:32 GMT):
I've enabled Dockerhub's image scanning, regardless.
macfarla (Mon, 24 Jan 2022 05:41:46 GMT):
LFR https://github.com/hyperledger/besu/pull/3317 to add trivy scan of docker file - output looks like this: https://github.com/hyperledger/besu/pull/3317/checks?check_run_id=4917443542
macfarla (Mon, 24 Jan 2022 05:48:16 GMT):
I can't see anything different in dockerhub - https://hub.docker.com/r/hyperledger/besu/tags - but I'm not in the hyperledger org - is that what would do it? @rjones
shemnon (Mon, 24 Jan 2022 15:59:23 GMT):
It looks very noisy and I didn't see (out of the 49) an actual vulnerability to besu or a library besu includes. They are all mostly about the docker image chosen or the secondary docker stuff auto-loaded by the docker image chosen. While useful for docker security I wouldn't want to add this to every commit nor would I want it to gate as it provides too many non-actionable errors.
shemnon (Mon, 24 Jan 2022 17:18:13 GMT):
Roadmap issue for next week's contributor call: Move to Java 17 - https://github.com/hyperledger/besu/issues/3320
RoboCopsGoneMad (Mon, 24 Jan 2022 19:59:02 GMT):
I was thinking of applying for a free IntelliJ Ultimate license on the basis of working on OSS. Anyone tried this yet?
RoboCopsGoneMad (Mon, 24 Jan 2022 19:59:03 GMT):
https://www.jetbrains.com/community/opensource/#support
RoboCopsGoneMad (Mon, 24 Jan 2022 19:59:16 GMT):
community edition has no Go support :(
shemnon (Mon, 24 Jan 2022 20:16:28 GMT):
The two last terms will present problems:
* Are not funded by commercial companies or organizations (NGO, educational, research, or governmental).
* Do not pay their core project developers.
shemnon (Mon, 24 Jan 2022 20:17:14 GMT):
ConsenSys paying your salary and the incentive program each independently disqualify. Ask CS to pay for ultimate.
RoboCopsGoneMad (Mon, 24 Jan 2022 20:18:58 GMT):
I wasn't thinking about me, I already own it.
RoboCopsGoneMad (Mon, 24 Jan 2022 20:19:30 GMT):
was more considering if anyone else applying would get denied since we are a Hyperledger project
shemnon (Mon, 24 Jan 2022 20:20:19 GMT):
It would be a mixed bag for sure. It would have better luck Hyperledger wide, but IBM's role in their own blockchain hits at another blocker.
RoboCopsGoneMad (Mon, 24 Jan 2022 20:20:39 GMT):
k
shemnon (Mon, 24 Jan 2022 20:21:15 GMT):
They had issues years ago with the groovy and grails developers expecting free licenses when g2one and later pivotal was profiting. I haven't asked because the besu model aligns with the issues they had there.
GarySchulte (Mon, 24 Jan 2022 21:57:28 GMT):
Shall we hold RC3 for @diegoll 's SLF4J PR ( https://github.com/hyperledger/besu/pull/3285 ) or just plan to do an RC4 after? It seems to me that a quarterly release with this PR would have good optics after all of the log4j drama
GarySchulte (Mon, 24 Jan 2022 21:58:14 GMT):
Either is good by me with a slight preference for getting an RC3 out there sooner rather than later.
shemnon (Mon, 24 Jan 2022 22:22:00 GMT):
Putting it in a RC/Beta first however it happens is good optics. It's worth holding it a few days IMHO unless someone else needs RC3.
shemnon (Mon, 24 Jan 2022 22:22:28 GMT):
Follow-on, who are the engineers running the release? I dropped off the contributor call before we could have release schedule dicussions.
macfarla (Mon, 24 Jan 2022 22:24:59 GMT):
Agree I'm inclined to hold RC3 to get SLF4J in
diegoll (Mon, 24 Jan 2022 22:31:16 GMT):
I'll try to fix most of the Sonar issues today
GarySchulte (Mon, 24 Jan 2022 23:31:21 GMT):
I was planning to be primary for the RC3 release, but it makes no difference to me. My main interest is in breaking the logjam for main
(and sort out the RCx / Beta nomenclature at next contrib call
macfarla (Tue, 25 Jan 2022 01:12:13 GMT):
agree we don't want PRs gated on this. What's there now isn't useful because we've just updated all the dependencies :) I think if we'd had this in place it may have helped us proactively with some of the CVEs
frankliawesome (Tue, 25 Jan 2022 01:55:39 GMT):
I found a test yesterday that fails on Java 16 https://github.com/hyperledger/besu/issues/3318, since we're also talking about going to Java 17, does it make sense to run all our tests on Java 16 and iron out errors first? @shemnon
shemnon (Tue, 25 Jan 2022 01:56:28 GMT):
Java 16 is never getting another security update. Target 17.
shemnon (Tue, 25 Jan 2022 01:57:17 GMT):
How would we know what to look for with 49 non-actionable warnings? Devs are not going to go through it with a fine tooth comb each time.
shemnon (Tue, 25 Jan 2022 01:57:56 GMT):
Dependabot may be the only viable alternative. It's bad, but better than the other alternatives.
frankliawesome (Tue, 25 Jan 2022 02:00:16 GMT):
And the idea would be if we're passing everything on java 17 we're naturally backwards compatible with 16? Not sure if that's how it works
macfarla (Tue, 25 Jan 2022 02:47:16 GMT):
have you tried your issue on Java 17 @frankliawesome ?
macfarla (Tue, 25 Jan 2022 02:48:33 GMT):
I would expect the things we need to action would be critical/high priority ie they would show above the 49 things and no need for fine tooth comb.
macfarla (Tue, 25 Jan 2022 02:48:38 GMT):
Dependabot is next on my list!
shemnon (Tue, 25 Jan 2022 04:24:02 GMT):
If we compile on 17 we have to use a Java 17 vm or newer to run. But if we set the minimum to 17 it will never compile on 16.
diegoll (Tue, 25 Jan 2022 04:24:23 GMT):
Who has permissions to review Security Hotspots in Sonar? https://sonarcloud.io/project/security_hotspots?id=hyperledger_besu&pullRequest=3285&resolved=false&types=SECURITY_HOTSPOT
shemnon (Tue, 25 Jan 2022 04:24:25 GMT):
Even if zero features are used, the gradle script will fail
macfarla (Tue, 25 Jan 2022 04:42:03 GMT):
I can view it
diegoll (Tue, 25 Jan 2022 04:44:01 GMT):
Thanks. I put a little more context here https://github.com/hyperledger/besu/pull/3285#issuecomment-1020797286
macfarla (Tue, 25 Jan 2022 04:46:02 GMT):
I can view it, I think it looks ok, but I can't change the status for the security hotspot ie mark it as safe - and would want a second opinion on that anyway @shemnon ?
diegoll (Tue, 25 Jan 2022 04:50:32 GMT):
sure, no problem.
I would be interesting to have some opinions on [1bcae13](https://github.com/hyperledger/besu/pull/3285/commits/1bcae13b9ac6c01a6313d6091ea1de57b96f51ae). Sonar has this rule saying that _"Exceptions should be either logged or rethrown but not both"_, so I had to remove few log calls and I mostly added that information to the exception message when it wasn't the same
diegoll (Tue, 25 Jan 2022 04:50:32 GMT):
sure, no problem.
I would be interesting to have some opinions on [1bcae13](https://github.com/hyperledger/besu/pull/3285/commits/1bcae13b9ac6c01a6313d6091ea1de57b96f51ae) too. Sonar has this rule saying that _"Exceptions should be either logged or rethrown but not both"_, so I had to remove few log calls and I mostly added that information to the exception message when it wasn't the same
diegoll (Tue, 25 Jan 2022 04:50:32 GMT):
sure, no problem.
I would be interesting to have some opinions on [1bcae13](https://github.com/hyperledger/besu/pull/3285/commits/1bcae13b9ac6c01a6313d6091ea1de57b96f51ae) too. Sonar has this rule saying that _"Exceptions should be either logged or rethrown but not both"_, so I had to remove a few log calls and I mostly added that information to the exception message when it wasn't the same
diegoll (Tue, 25 Jan 2022 04:50:32 GMT):
sure, no problem.
It would be interesting to have some opinions on [1bcae13](https://github.com/hyperledger/besu/pull/3285/commits/1bcae13b9ac6c01a6313d6091ea1de57b96f51ae) too. Sonar has this rule saying that _"Exceptions should be either logged or rethrown but not both"_, so I had to remove a few log calls and I mostly added that information to the exception message when it wasn't the same
diegoll (Tue, 25 Jan 2022 04:50:32 GMT):
sure, no problem.
It would be interesting to have some opinions on [1bcae13](https://github.com/hyperledger/besu/pull/3285/commits/1bcae13b9ac6c01a6313d6091ea1de57b96f51ae) too. Sonar has this rule saying that _"Exceptions should be either logged or rethrown but not both"_, so I had to remove a few log calls and I mostly added that information to the exception message when this wasn't the same
frankliawesome (Tue, 25 Jan 2022 05:00:12 GMT):
makes sense! @macfarla also fails on java 17 so we definitely should fix it
diegoll (Tue, 25 Jan 2022 05:01:05 GMT):
FYI, https://github.com/hyperledger/besu/pull/3285#issuecomment-1020797286
diegoll (Tue, 25 Jan 2022 05:09:16 GMT):
btw, it's worth to mention that through this PR we would not end the log4j2 drama on our side b/c we're still depending on it, but it'll be certainly easier to change the underlying logging backend in the near future (if there is interest I can make a follow up PR migrating log4j2 to, let's say, logback)
macfarla (Tue, 25 Jan 2022 06:35:17 GMT):
I am on board with 1bcae13 - it might change what ends up in the logs slightly so maybe worth calling out in the changelog
macfarla (Tue, 25 Jan 2022 06:35:45 GMT):
:+1:
macfarla (Tue, 25 Jan 2022 06:42:54 GMT):
dependabot shows zero open/closed alerts. I don't know how long it's been turned on for (wasn't me) https://github.com/hyperledger/besu/security/dependabot
macfarla (Tue, 25 Jan 2022 07:37:44 GMT):
Can someone from the mainnet team review this Clique block choice rule PR from @shemnon - it's been waiting a while - I added some comments but it's a little out of my comfort zone - https://github.com/hyperledger/besu/pull/2084
sbohanlf (Tue, 25 Jan 2022 15:16:32 GMT):
Has joined the channel.
diegoll (Tue, 25 Jan 2022 19:01:42 GMT):
https://github.com/hyperledger/besu/pull/3285 is all :white_check_mark: and up-to-date with `main`
diegoll (Tue, 25 Jan 2022 19:01:42 GMT):
https://github.com/hyperledger/besu/pull/3285 is all :white_check_mark: and up-to-date with `main` (I added a changelog entry)
shemnon (Tue, 25 Jan 2022 19:59:12 GMT):
I've approved, but for a change this large it would be nice to have another maintainer approve as well. @GarySchulte or @macfarla ?
GarySchulte (Tue, 25 Jan 2022 20:04:18 GMT):
will do.
GarySchulte (Tue, 25 Jan 2022 20:04:40 GMT):
this must have been the "alert" noice I heard rather than the whitelist question from `besu`
GarySchulte (Tue, 25 Jan 2022 20:13:37 GMT):
what was the problem with printf format ? I can't find the sonar output that was complaining
GarySchulte (Tue, 25 Jan 2022 20:16:45 GMT):
Not going to nitpick here, but if slf4j doesn't support printf style we could wrap String.format
GarySchulte (Tue, 25 Jan 2022 20:20:54 GMT):
:cruise_ship:
macfarla (Tue, 25 Jan 2022 21:23:09 GMT):
the only thing for 3285 is the changelog entry should be in RC3 not RC2 - but @GarySchulte would you be ok with fixing this up as part of RC3 - there's no RC3 section in the changelog yet
shemnon (Tue, 25 Jan 2022 21:40:18 GMT):
IMHO printf style tricks is what burned log4j.
GarySchulte (Tue, 25 Jan 2022 21:40:37 GMT):
can do
shemnon (Tue, 25 Jan 2022 21:40:50 GMT):
message and a series of key/values from our logger library would be a dream setup, but not this year,
macfarla (Tue, 25 Jan 2022 21:41:21 GMT):
Can we still use log4j2.xml config file eg https://besu.hyperledger.org/en/stable/HowTo/Monitor/Logging/#advanced-logging
GarySchulte (Tue, 25 Jan 2022 22:01:23 GMT):
I will defer to @diegoll on that @macfarla. If we can't I presume it would be a docs update we need to get out post haste
GarySchulte (Tue, 25 Jan 2022 22:04:19 GMT):
I am about to head out and get some vitamin D for a bit. Happy to start an RC3 this afternoon or tomorrow morning. I can merge Diego's SLF4J PR now and start the CI ball rolling before I go
GarySchulte (Tue, 25 Jan 2022 22:05:21 GMT):
lmk
diegoll (Tue, 25 Jan 2022 22:48:42 GMT):
Yes, that should still work. The backend is still Log4J2 so the usage of SLF4J is transparent from the outside
diegoll (Tue, 25 Jan 2022 22:59:52 GMT):
would you prefer if I just remove that commit from the PR and then @GarySchulte can set that up once building the RC3?
diegoll (Tue, 25 Jan 2022 23:12:45 GMT):
The rule that Sonar was complaining regarding the string format was [java:S3457](https://sonarcloud.io/organizations/hyperledger/rules?open=java%3AS3457&rule_key=java%3AS3457)
GarySchulte (Tue, 25 Jan 2022 23:13:40 GMT):
nah, just merge it as is and I will tweak with the deployment PR.
diegoll (Tue, 25 Jan 2022 23:14:02 GMT):
awesome. Merging it then
GarySchulte (Tue, 25 Jan 2022 23:14:15 GMT):
:+1:
diegoll (Tue, 25 Jan 2022 23:15:07 GMT):
done!
GarySchulte (Tue, 25 Jan 2022 23:15:24 GMT):
cool. starting an RC3 deployment now
macfarla (Tue, 25 Jan 2022 23:19:40 GMT):
so this is approved now but do we have to coordinate with Geth / ACD as to when to merge it?
macfarla (Tue, 25 Jan 2022 23:19:40 GMT):
so this is approved now but do we have to coordinate with Geth / ACD as to when to merge it @shemnon ?
GarySchulte (Tue, 25 Jan 2022 23:29:34 GMT):
deployment thread
GarySchulte (Tue, 25 Jan 2022 23:29:52 GMT):
```
What is the project to release? (e.g. ConsenSys/plus)
hyperledger/besu
What is the full version number to release?
22.10.0-RC3
What is the branch to release?
main
The plan for this release is:
1. Announce in #pegasys-prod-dev-eth1, #RocketChat
2. Checkout project
a. Check out hyperledger/besu
b. Switch to branch main
3. Prepare branch for release of 22.10.0-RC3
a. Create branch release-22.10.0-RC3
b. Set version to 22.10.0-RC3
c. Commit changes
d. Push to GitHub
e. Create PR: Release 22.10.0-RC3
f. Wait for approvals on PR
GarySchulte (Tue, 25 Jan 2022 23:30:31 GMT):
releasing RC3 from main
GarySchulte (Tue, 25 Jan 2022 23:31:55 GMT):
`@here` (no perms for notifying) - We're beginning the release process for hyperledger/besu, please don't merge anything to main until further notice
GarySchulte (Tue, 25 Jan 2022 23:36:56 GMT):
https://github.com/hyperledger/besu/pull/3327/files
can I get :eyes: , double check the changelog :pray:
GarySchulte (Tue, 25 Jan 2022 23:39:05 GMT):
worth mentioning again that this is a release from `main`, not the cherry-picked RC series
diegoll (Tue, 25 Jan 2022 23:39:54 GMT):
I think the changelog entry should be `22.10.0-RC3` instead of `22.1.0-RC3`
diegoll (Tue, 25 Jan 2022 23:40:41 GMT):
I mean, for the title of the section
GarySchulte (Tue, 25 Jan 2022 23:42:05 GMT):
it should be 22.1.0-RC3. I goofed on the release branch name. I am going to fix the change log and gradle.properties and just proceed with the existing branch name
macfarla (Tue, 25 Jan 2022 23:45:13 GMT):
approved
GarySchulte (Tue, 25 Jan 2022 23:45:22 GMT):
TY
GarySchulte (Wed, 26 Jan 2022 00:13:05 GMT):
https://github.com/hyperledger/besu/pull/3328
macfarla (Wed, 26 Jan 2022 00:14:20 GMT):
:white_check_mark:
grace.hartley (Wed, 26 Jan 2022 00:37:36 GMT):
Hey folks! The ConsenSys team reviewed Danno's Incentive Program proposal. I worked with @timbeiko to draft a new proposal. This proposal included Guiding Principles, Funds Distributions, and Processes in an attempt to break down the different functions of the program. If folks can review, we can set up a call later this week to discuss as a community. Thanks for everyone's patience while we worked through this. https://wiki.hyperledger.org/pages/viewpage.action?pageId=62236988
GarySchulte (Wed, 26 Jan 2022 01:00:00 GMT):
`@here` - It's now safe to merge into main for hyperledger/besu again. Still finalizing the release, but no need to block merges
GarySchulte (Wed, 26 Jan 2022 01:06:01 GMT):
https://github.com/hyperledger/besu/releases/tag/22.1.0-RC3
GarySchulte (Wed, 26 Jan 2022 01:10:27 GMT):
whoops wrong thread
GarySchulte (Wed, 26 Jan 2022 01:11:11 GMT):
https://github.com/hyperledger/besu/releases/tag/22.1.0-RC3
NicolasMassart (Wed, 26 Jan 2022 10:45:11 GMT):
Hi team, could you review this contributor PR please? https://github.com/hyperledger/besu-native/pull/52
macfarla (Thu, 27 Jan 2022 02:29:29 GMT):
Hey @diegoll could you update this page in the wiki to reflect the change from Log4j? https://wiki.hyperledger.org/display/BESU/Logging
macfarla (Thu, 27 Jan 2022 04:30:54 GMT):
More release fun - LFR https://github.com/hyperledger/besu/pull/3330
macfarla (Thu, 27 Jan 2022 04:53:43 GMT):
FYI I've just merged this PR which will do a nightly scan of the develop docker image so we can see what it looks like. https://github.com/hyperledger/besu/pull/3295 - if it causes any problems at all, or we don't like the report, it's easy to revert it.
grace.hartley (Thu, 27 Jan 2022 14:41:07 GMT):
Hey All - Here is the agenda for Monday evening ET/Tuesday morning AEST contributor call. The main topic of discussion will be Incentive Program's Proposal #3. We will record and take notes per usual for folks who aren't able to join. Thanks! https://wiki.hyperledger.org/display/BESU/2022-01-31+Besu+Contributor+Call
GarySchulte (Thu, 27 Jan 2022 19:39:04 GMT):
Hey M1 fans... https://github.com/facebook/rocksdb/pull/9386/files#diff-648afe3d986261d8f2015b2b131b0e4a448d4dc6946cfde1a7a836876cee255eR23-R25
so we are just waiting on the next release of rocksdbjni. I am not sure if that is 6.28.1 and/or when that would be. Maven still only shows 6.27.3 as latest. @shemnon do you have any more detail on the problem encountered with 6.27.3 ?
https://github.com/hyperledger/besu/pull/3221
shemnon (Thu, 27 Jan 2022 19:41:29 GMT):
rollback promt - https://github.com/hyperledger/besu/issues/3220
GarySchulte (Thu, 27 Jan 2022 19:42:43 GMT):
thanks - I either hadn't seen that issue or didn't remember it. Now to figure out what is up with aws
shemnon (Thu, 27 Jan 2022 19:43:16 GMT):
If needed we could host a rocksdbjni build on hyperledger's artifactory. Not sure who will make the binary bits however...
shemnon (Thu, 27 Jan 2022 19:43:38 GMT):
we would want to put a tag on the version to prevent maven central collision
GarySchulte (Thu, 27 Jan 2022 19:43:57 GMT):
I would be down for that. m1 has been a persistent thorn in my side
GarySchulte (Thu, 27 Jan 2022 22:14:42 GMT):
not able to reproduce the RocksDbException on ec2 instance :thinking:
diegoll (Fri, 28 Jan 2022 00:31:35 GMT):
Hello Sally. I just slightly rephrased it and removed the `FATAL` level. Let me know if it makes sense
shemnon (Fri, 28 Jan 2022 03:59:57 GMT):
Proposal 3.5 - https://wiki.hyperledger.org/pages/viewpage.action?pageId=62237321
Mostly using Patron as a generic word and adding my employer. Diffs can be seen on this page - https://wiki.hyperledger.org/pages/viewpage.action?pageId=62237303
diegoll (Fri, 28 Jan 2022 18:32:51 GMT):
Would the ETC Coop be considered a _patron_ in this case?
shemnon (Fri, 28 Jan 2022 19:19:54 GMT):
EF is only considering Ethereum Mainnet work, so no.
macfarla (Fri, 28 Jan 2022 19:47:06 GMT):
Looks good, thanks! Did you verify that the custom logging config file still works as documented?
timbeiko (Fri, 28 Jan 2022 22:34:47 GMT):
No - the program is targeted to incentivise client teams building on the Ethereum mainnet, not ETC or other networks (e.g. private chains).
diegoll (Fri, 28 Jan 2022 22:55:28 GMT):
Hello Tim. The majority of my contributions were not ETC-specific and benefits ETH Mainnet too
GarySchulte (Sat, 29 Jan 2022 14:36:09 GMT):
This is a really good point @diegoll - it is hard to differentiate between mainnet ethereum, etc and enterprise contributions, since they all share the same codebase.
Just because you would contribute to ETC specific items does not mean you would not be contributing to eth mainnet functionality
GarySchulte (Sat, 29 Jan 2022 14:38:52 GMT):
I think it is going to be essentially impossible to make an 'ethereum mainnet' distinction on the basis of code commits.
GarySchulte (Sat, 29 Jan 2022 14:42:43 GMT):
IMO it should be either all "besu code contributors"
GarySchulte (Sat, 29 Jan 2022 14:42:43 GMT):
IMO we should include all besu code contributors if the goal is to incentivize besu development for mainnet
GarySchulte (Sat, 29 Jan 2022 14:49:17 GMT):
your contributions have definitely helped ethereum mainnet. the sjlf4j replacement is a great example. If Hedera can be a patron, why not ETC? They are functionally idential
loziniak (Sun, 30 Jan 2022 13:05:35 GMT):
Has joined the channel.
macfarla (Sun, 30 Jan 2022 22:10:29 GMT):
I tried with my custom log4j config and it works for me
diegoll (Sun, 30 Jan 2022 23:40:59 GMT):
Thanks Sally for checking it. I was AFK on Friday when you asked me and I was going to check it tomorrow first time in the morning
macfarla (Mon, 31 Jan 2022 00:39:34 GMT):
all good! I also raised a ticket with docs to remove the FATAL log level https://github.com/hyperledger/besu-docs/issues/935
frankliawesome (Mon, 31 Jan 2022 02:07:02 GMT):
Doing 22.1.0-RC4 release with a besu-native bump right now
frankliawesome (Mon, 31 Jan 2022 03:38:13 GMT):
Release is now complete
RoboCopsGoneMad (Mon, 31 Jan 2022 18:19:19 GMT):
I noticed we have dagger2 in evmtool, that is awesom. I watched this video on it over the weekend, https://youtu.be/oK_XtfXPkqw and I think it is really good fit for us. I think I'd like to try it for the merge configuration stuff. There area lot of `isMergeEnabled()` conditionals popping up in various places, and I'd like to use Dagger to inject the class that is Merge aware into any client code that needs it.
timbeiko (Mon, 31 Jan 2022 18:20:19 GMT):
Apologies, I didn't realize that, @diegoll.
Happy to elaborate on the call, but my general view is that to separate transaction/validator fees, using a broader approach might make sense, although you still probably want to have some "check" about contributing to things that benefit mainnet. IMO the examples that @GarySchulte states would be sufficient to be considered for a certain period.
That said, access to the principal/withdrawal credential should go towards organizations who have shown long term, significant support towards Ethereum mainnet. A rough list of things that contribute towards that are developing major features (e.g. sync/DB improvements), implementing consensus changes (e.g. hard forks), leading/contributing to R&D or testing efforts beyond the single client, employing a dedicated team to work on this full time, etc.
I'm pretty sure the EF will want to maintain discretion about who would be included there, with CS and HLF being two organizations we are comfortable with today, but we can definitely open the door for others to be considered.
timbeiko (Mon, 31 Jan 2022 18:20:19 GMT):
Apologies, I didn't realize that, @diegoll
Happy to elaborate on the call, but my general view is that to separate transaction/validator fees, using a broader approach might make sense, although you still probably want to have some "check" about contributing to things that benefit mainnet. IMO the examples that @GarySchulte states would be sufficient to be considered for a certain period.
That said, access to the principal/withdrawal credential should go towards organizations who have shown long term, significant support towards Ethereum mainnet. A rough list of things that contribute towards that are developing major features (e.g. sync/DB improvements), implementing consensus changes (e.g. hard forks), leading/contributing to R&D or testing efforts beyond the single client, employing a dedicated team to work on this full time, etc.
I'm pretty sure the EF will want to maintain discretion about who would be included there, with CS and HLF being two organizations we are comfortable with today, but we can definitely open the door for others to be considered.
timbeiko (Mon, 31 Jan 2022 18:20:19 GMT):
Apologies, I didn't realize that, @diegoll
Happy to elaborate on the call, but my general view is that to separate transaction/validator fees, using a broader approach might make sense, although you still probably want to have some "check" about contributing to things that benefit mainnet. IMO the examples that @GarySchulte states would be sufficient to be considered for a certain period.
That said, access to the principal/withdrawal credential should go towards organizations who have shown long term, significant support towards Ethereum mainnet. A rough list of things that contribute towards that are developing major features (e.g. sync/DB improvements), implementing consensus changes (e.g. hard forks), leading/contributing to R&D or testing efforts beyond the single client, employing a dedicated team to work on this full time, regularly engaging in cross-client conversations/collaborations, etc.
I'm pretty sure the EF will want to maintain discretion about who would be included there, with CS and HLF being two organizations we are comfortable with today, but we can definitely open the door for others to be considered.
RoboCopsGoneMad (Mon, 31 Jan 2022 19:59:00 GMT):
i have a test (unrelated to my pr) that is now failing consistently on CI, but not locally. I see the test reports go to:
RoboCopsGoneMad (Mon, 31 Jan 2022 19:59:03 GMT):
```file:///home/circleci/project/besu/build/reports/tests/test/index.html```
RoboCopsGoneMad (Mon, 31 Jan 2022 19:59:25 GMT):
do we retain that anywhere or is that trashed when the container that runs the test terminates?
shemnon (Mon, 31 Jan 2022 20:41:43 GMT):
Review Request, expose a RLP parsing field: https://github.com/hyperledger/besu/pull/3354
RoboCopsGoneMad (Mon, 31 Jan 2022 21:00:41 GMT):
lgtm, what's it for?
shemnon (Mon, 31 Jan 2022 21:09:07 GMT):
Some work at Hedera. We want the RLP string and also want to point _into_ that RLP string instead of duplicating it.
shemnon (Mon, 31 Jan 2022 21:09:57 GMT):
If we can use the bytes directly we would just use the sliced bytes, but it will go through a protobuf encoding.
rjones (Mon, 31 Jan 2022 21:20:33 GMT):
@macfarla @grace.hartley I've updated the invite for the APAC call to use a Hyperledger zoom. https://lists.hyperledger.org/g/besu/viewevent?repeatid=22224&eventid=1413443&calstart=2022-01-31
macfarla (Mon, 31 Jan 2022 22:17:21 GMT):
you can look at "artifacts" on the CI page but idk if that one would be there
macfarla (Mon, 31 Jan 2022 22:21:21 GMT):
Nightly scan of docker images - this is what the report looks like. It's showing the protobuf CVE which we know about. https://app.circleci.com/pipelines/github/hyperledger/besu/12961/workflows/dde97a21-0eb3-4345-8767-0d4490a2ee44/jobs/71864
shemnon (Tue, 01 Feb 2022 01:00:41 GMT):
which zoom is the contributor call on? I'm on https://zoom.us/j/91405289771?pwd=QzA3a0ozTjVSZ0hsYzgwTm9lVDRPQT09
macfarla (Tue, 01 Feb 2022 01:19:06 GMT):
@Daniela_Barbosa
Daniela_Barbosa (Tue, 01 Feb 2022 01:28:41 GMT):
The EF person i mentioned met with me, Tim and our LF legal team last week was Danny Ryan for the record.
timbeiko (Tue, 01 Feb 2022 01:34:36 GMT):
Is the call over already?
timbeiko (Tue, 01 Feb 2022 01:34:55 GMT):
Wiki had 6pm LA, which is 25 minutes from now
timbeiko (Tue, 01 Feb 2022 01:34:58 GMT):
Screen Shot 2022-01-31 at 5.34.45 PM.png
macfarla (Tue, 01 Feb 2022 02:58:24 GMT):
^ is the time for LA / NY both wrong there?
diegoll (Tue, 01 Feb 2022 02:59:15 GMT):
Folks, what do you think of creating an open space for talking about technical stuff? We can do something in between technical and the chill, off the record, space that was mentioned some Contributors Calls back. There have been really interesting points mentioned today, like the JVM optimizations stuff, or using dagger2 in this same chat. I personally wanted to talk about moving Log4j2 to logback and I'm not sure if the Contributor Call is the better place to came up with that conversation given the current structure they have
shemnon (Tue, 01 Feb 2022 02:59:54 GMT):
Discord will make this sooo much easier.
shemnon (Tue, 01 Feb 2022 03:00:17 GMT):
I think we should book a regular non-recorded tech chat call on the weeks without contributor calls.
shemnon (Tue, 01 Feb 2022 03:00:34 GMT):
And I think contributor calls are totally a good place to do it. Agendas rarely fill up the time.
shemnon (Tue, 01 Feb 2022 03:00:50 GMT):
(and this tech chat should be on discord)
diegoll (Tue, 01 Feb 2022 03:02:40 GMT):
I've never used the voice channels that Discord offer, maybe that would be useful
macfarla (Tue, 01 Feb 2022 03:08:51 GMT):
I've received zero feedback on this proposal - if no objections I will turn off GitHub discussions in 48 hrs.
diegoll (Tue, 01 Feb 2022 03:09:49 GMT):
these past Contributor Calls have been taken mostly by the incentive program, which is fine and the right place to be, with an heterogeneous audience, which is great for those topics, not so sure about hard technical stuff, at least until this Incentive Program situation is resolved (which hopefully would be soon). Anyway, we can try to make something like Besu Dev Calls, non recorded, every other week. Most likely most of us would have something to share, discuss or whatever
RoboCopsGoneMad (Tue, 01 Feb 2022 13:54:36 GMT):
love all of this. I'm down for whatever.
grace.hartley (Tue, 01 Feb 2022 14:03:27 GMT):
yep. it seems with daylight savings, things got mixed up.
grace.hartley (Tue, 01 Feb 2022 14:03:42 GMT):
Hey all - Here are notes from last night's contributor call. https://wiki.hyperledger.org/display/BESU/2022-1-31+Contributor+Call+notes
RoboCopsGoneMad (Tue, 01 Feb 2022 14:43:23 GMT):
is there somewhere I can see our CI costs?
RoboCopsGoneMad (Tue, 01 Feb 2022 14:44:36 GMT):
Last night @shemnon mentioned they were the most expensive among hyperledger projects, I should probably keep that in mind about that while thinking on the release process
RoboCopsGoneMad (Tue, 01 Feb 2022 14:44:36 GMT):
Last night @shemnon mentioned they were the most expensive among hyperledger projects, I should probably keep that in mind while thinking on the release process
rjones (Tue, 01 Feb 2022 19:03:55 GMT):
@RoboCopsGoneMad let me look
rjones (Tue, 01 Feb 2022 19:04:28 GMT):
Meanwhile, take a look at this: https://ryjones.github.io/stuffandthings/besu.html
rjones (Tue, 01 Feb 2022 19:08:39 GMT):
@RoboCopsGoneMad Hyperledger paid $18600 on 30-July-2021 and $22,200 on 10-January-2022
rjones (Tue, 01 Feb 2022 19:10:37 GMT):
Over the last 90 days, Besu has used 12,442,540 credits. There are 38,987,026 credits remaining from the 10-Jan purchase.
rjones (Tue, 01 Feb 2022 19:11:30 GMT):
Besu used 4,248,445 credits from 30-Dec to 29-Jan
shemnon (Tue, 01 Feb 2022 19:14:26 GMT):
That report is awesome. what generates it? I'de love to poke and prod at time horizons.
rjones (Tue, 01 Feb 2022 19:27:51 GMT):
this is the action I used to generate it: https://github.com/hyperledger-labs/hyperledger-community-management-tools/blob/main/.github/workflows/publish-reports.yml
rjones (Tue, 01 Feb 2022 19:28:16 GMT):
the code is here: https://github.com/hyperledger-labs/hyperledger-community-management-tools/tree/main/project-reports
rjones (Tue, 01 Feb 2022 19:48:56 GMT):
you might not have enough permissions to run the reports - feel free to submit a PR over here: https://github.com/ryjones/hyperledger-community-management-tools and I'll run the report
lucassaldanha (Tue, 01 Feb 2022 20:59:17 GMT):
Do it!
RoboCopsGoneMad (Tue, 01 Feb 2022 22:54:19 GMT):
got pretty stuck today, this is a rubber duck thread :duck:
RoboCopsGoneMad (Tue, 01 Feb 2022 23:02:14 GMT):
My PR for adding a new http service to dedicate engine apis to their own port, is failing to pass CI.
RoboCopsGoneMad (Tue, 01 Feb 2022 23:02:27 GMT):
https://github.com/hyperledger/besu/pull/3350
RoboCopsGoneMad (Tue, 01 Feb 2022 23:03:15 GMT):
The test that fails, is totally unrelated to my changes so far as I can tell now.
RoboCopsGoneMad (Tue, 01 Feb 2022 23:03:16 GMT):
org.hyperledger.besu.services.BesuEventsImplTest
RoboCopsGoneMad (Tue, 01 Feb 2022 23:04:52 GMT):
The test pass locally in my ide, from the command line with gradle in isolation, from the command line with gradle in a full build, and even if I pull down the executor docker image, and run the test on that, just like CI should be doing.
RoboCopsGoneMad (Tue, 01 Feb 2022 23:05:09 GMT):
that lest test, Gary also ran, and he had the test fail for him.
RoboCopsGoneMad (Tue, 01 Feb 2022 23:08:15 GMT):
when it fails in CI, it fails consistently.
RoboCopsGoneMad (Tue, 01 Feb 2022 23:20:51 GMT):
what would help is if i could bisect each commit and run it in CI, maybe i'll duplicate my branch and cherry pick them over one at a time to see what commit started this debacle
lucassaldanha (Wed, 02 Feb 2022 00:55:44 GMT):
Proposal to add Ryan Jones as a non-code maintainer: https://github.com/hyperledger/besu/pull/3359
macfarla (Wed, 02 Feb 2022 02:23:17 GMT):
could it be to do with clique fork choice changes? something different in remote vs local env?
macfarla (Wed, 02 Feb 2022 02:49:15 GMT):
sounds great.
macfarla (Wed, 02 Feb 2022 02:49:30 GMT):
is everyone on the discord?
shemnon (Wed, 02 Feb 2022 02:58:30 GMT):
The part where it always succeeds when run alone makes me think of static fields being changed.
shemnon (Wed, 02 Feb 2022 02:58:42 GMT):
I don't think I'm using static in the fork choice parts?
shemnon (Wed, 02 Feb 2022 02:58:42 GMT):
I don't think I'm using statics in the fork choice parts?
macfarla (Wed, 02 Feb 2022 03:49:46 GMT):
I saw forkBlock in the test but I think it's unrelated
shemnon (Wed, 02 Feb 2022 03:51:55 GMT):
can't be fork choice, last merge to that branch was #3303, last week.
shemnon (Wed, 02 Feb 2022 04:35:26 GMT):
I can reproduce on my local machine and git bisect came to the same conclusion as https://github.com/hyperledger/besu/pull/3360
RoboCopsGoneMad (Wed, 02 Feb 2022 14:04:29 GMT):
thanks for the help frens. i'm gonna try to break that commit down into smaller pieces to further isolate it
RoboCopsGoneMad (Wed, 02 Feb 2022 14:05:17 GMT):
when you reproduced it, was it failing on the entire test suite or did you run just the problem test?
RoboCopsGoneMad (Wed, 02 Feb 2022 14:05:50 GMT):
maybe there is something i'm git ignoring, will try a fresh checkout...
RoboCopsGoneMad (Wed, 02 Feb 2022 14:06:34 GMT):
i am, just say the word and i'm there
shemnon (Wed, 02 Feb 2022 14:19:18 GMT):
Just that one test
RoboCopsGoneMad (Wed, 02 Feb 2022 14:23:42 GMT):
:thumbsup:
shemnon (Wed, 02 Feb 2022 14:49:19 GMT):
Is anyone still using YoutKit? - https://github.com/hyperledger/besu/pull/341
RoboCopsGoneMad (Wed, 02 Feb 2022 15:39:46 GMT):
fresh checkout passes. :(
RoboCopsGoneMad (Wed, 02 Feb 2022 15:39:57 GMT):
what jvm did you use @shemnon
RoboCopsGoneMad (Wed, 02 Feb 2022 15:41:54 GMT):
i am trying with temurin 11.0.13 since that was what the executing image should be using
RoboCopsGoneMad (Wed, 02 Feb 2022 15:42:11 GMT):
tried other recent adopt 11 variants, test passes on all of 'em
RoboCopsGoneMad (Wed, 02 Feb 2022 15:47:44 GMT):
I last used it back in July, fruitlessly. would not miss it if it disappeared.
diegoll (Wed, 02 Feb 2022 16:11:43 GMT):
Hello Ry, would you have the recording of this? so it can be added into [the wiki](https://wiki.hyperledger.org/display/BESU/2022-01-31+Besu+Contributor+Call)
RoboCopsGoneMad (Wed, 02 Feb 2022 16:11:50 GMT):
i would think running it on the same executor image would've controlled for jdk version
shemnon (Wed, 02 Feb 2022 16:11:54 GMT):
```
dannoferrin@hederas-mbp besu % java --version
openjdk 17 2021-09-14
OpenJDK Runtime Environment Temurin-17+35 (build 17+35)
OpenJDK 64-Bit Server VM Temurin-17+35 (build 17+35, mixed mode, sharing)
dannoferrin@hederas-mbp besu %
```
RoboCopsGoneMad (Wed, 02 Feb 2022 16:12:12 GMT):
thx
shemnon (Wed, 02 Feb 2022 16:16:30 GMT):
I think it's because BesuEventsImplTest depends on random numbers to generate some data. More tests means different random numbers.
shemnon (Wed, 02 Feb 2022 16:17:25 GMT):
or maybe not. But random numbers make the tester in me itchy.
RoboCopsGoneMad (Wed, 02 Feb 2022 16:21:45 GMT):
maybe? nothing so far has felt nondeterministic though. I have never gotten a failure locally, despite trying all the vms
RoboCopsGoneMad (Wed, 02 Feb 2022 16:44:52 GMT):
looks like you are on OSX, what version? i'm on 12.1
RoboCopsGoneMad (Wed, 02 Feb 2022 16:45:02 GMT):
i'm totally grasping at straws here....
RoboCopsGoneMad (Wed, 02 Feb 2022 16:57:19 GMT):
ok, so what makes Danno's machine just like CI, but mine, Jiri and Gary's aren't.
shemnon (Wed, 02 Feb 2022 17:15:56 GMT):
It's a static field issue.
RoboCopsGoneMad (Wed, 02 Feb 2022 17:16:50 GMT):
:thinking:
shemnon (Wed, 02 Feb 2022 17:17:15 GMT):
`org.hyperledger.besu.config.experimental.MergeOptions` has a static field `mergeEnabled`.
shemnon (Wed, 02 Feb 2022 17:17:44 GMT):
`org.hyperledger.besu.ethereum.chain.DefaultBlockchain` sets the `blockChoiceRule` with code that checks that field.
shemnon (Wed, 02 Feb 2022 17:18:00 GMT):
Even if the other tests properly un-set the static field this presents a test problem.
shemnon (Wed, 02 Feb 2022 17:18:12 GMT):
Because tests cannot be run in parallel then.
shemnon (Wed, 02 Feb 2022 17:18:32 GMT):
We had this issue with goQuorum compatability. We cannot depend on static fields to drive feature flags.
RoboCopsGoneMad (Wed, 02 Feb 2022 17:18:54 GMT):
i was planning on redoing this with dagger
shemnon (Wed, 02 Feb 2022 17:18:58 GMT):
When set the reorg is not detected because the fork choice rule says it's never a reorg.
shemnon (Wed, 02 Feb 2022 17:19:09 GMT):
Dagger is a good solution to this.
RoboCopsGoneMad (Wed, 02 Feb 2022 17:19:54 GMT):
so this breaks differently on different machine due to thread scheduling of the tests?
shemnon (Wed, 02 Feb 2022 17:20:20 GMT):
exactly. Which is why I have a generally allergic reaction to mutable static fields.
shemnon (Wed, 02 Feb 2022 17:20:24 GMT):
statics should be static.
RoboCopsGoneMad (Wed, 02 Feb 2022 17:26:40 GMT):
did it ever pass for you?
RoboCopsGoneMad (Wed, 02 Feb 2022 17:27:12 GMT):
and how many cores do you have? gradle runs a test on each core...
RoboCopsGoneMad (Wed, 02 Feb 2022 17:36:17 GMT):
thanks for the time danno, really appreciate it. tough to fix bugs when you can't see em.
shemnon (Wed, 02 Feb 2022 17:42:49 GMT):
It passes for me if I run just those tests or do an incremental test that doesn't run any merge tests.
shemnon (Wed, 02 Feb 2022 17:43:16 GMT):
The reason bisect triggered it is that there were so many broad based changes the incremental tests ran the tests in question
shemnon (Wed, 02 Feb 2022 17:43:16 GMT):
The reason bisect triggered it is that there were so many broad based changes the incremental tests ran the problematic tests in question
RoboCopsGoneMad (Wed, 02 Feb 2022 17:46:46 GMT):
ok, thats the part i'm interested in. how did you do that once bisect led you to the same place my bisecting pr got to?
rjones (Wed, 02 Feb 2022 18:32:17 GMT):
done
diegoll (Wed, 02 Feb 2022 19:01:31 GMT):
thank!
diegoll (Wed, 02 Feb 2022 19:01:31 GMT):
thanks!
macfarla (Wed, 02 Feb 2022 22:54:59 GMT):
argh now that I've decided to disable it I don't have permissions. @rjones or @shemnon can you turn off discussions please? https://github.com/hyperledger/besu/discussions
rjones (Wed, 02 Feb 2022 22:55:52 GMT):
done
macfarla (Wed, 02 Feb 2022 22:57:42 GMT):
:thanks:
macfarla (Wed, 02 Feb 2022 23:49:14 GMT):
license checking - I just created an issue here - https://github.com/hyperledger/besu/issues/3366 - any objections to changing the license checking plugin to match what we use for other consensys projects?
rjones (Thu, 03 Feb 2022 18:04:11 GMT):
Hi, Besu People. Let's talk about ACLs and stuff. Would here or Discord be better?
rjones (Thu, 03 Feb 2022 18:07:45 GMT):
I invited the confusingly-named @RyanJJones10 to join [this GitHub group](https://github.com/orgs/hyperledger/teams/besu-triage/members), which has `Triage` permission on all **Besu** repos.
RyanJJones10 (Thu, 03 Feb 2022 18:07:45 GMT):
Has joined the channel.
rjones (Thu, 03 Feb 2022 18:09:20 GMT):
I noticed that the [overall parent group](https://github.com/orgs/hyperledger/teams/besu-contributors/repositories) had _very broad_ permissions. I changed those all to `Triage`, since the child groups all have more appropriate powers.
shemnon (Thu, 03 Feb 2022 18:14:51 GMT):
Triage rights are not enough for how the PMs at ConesnSys want to use Zenhub.
rjones (Thu, 03 Feb 2022 18:23:34 GMT):
ok, maintain? write?
rjones (Thu, 03 Feb 2022 18:24:18 GMT):
Should there be a side group of Besu PMs with broader powers that are not maintainers?
rjones (Thu, 03 Feb 2022 18:25:05 GMT):
Besu Contributors has 40 members. Only 29 people are in the maintainers file.
RoboCopsGoneMad (Thu, 03 Feb 2022 21:29:48 GMT):
triage seems ok to me.
RoboCopsGoneMad (Thu, 03 Feb 2022 21:29:58 GMT):
not sure what is missing there
shemnon (Thu, 03 Feb 2022 21:36:02 GMT):
Zenhub requires write, it checks for it and refuses to work unless it has it. It's in their FAQ. So without write access zenhub is a read-only too.
shemnon (Thu, 03 Feb 2022 21:36:02 GMT):
Zenhub requires write, it checks for it and refuses to work unless it has it. It's in their FAQ. So without write access zenhub is a read-only tool.
shemnon (Thu, 03 Feb 2022 21:36:23 GMT):
They call out triage as not enough.
rjones (Thu, 03 Feb 2022 21:46:27 GMT):
I've set it to Maintain for now.
rjones (Thu, 03 Feb 2022 21:47:00 GMT):
@shemnon https://github.com/orgs/hyperledger/teams/besu-triage/repositories take a look
shemnon (Thu, 03 Feb 2022 21:51:57 GMT):
@RyanJJones10 can you confirm you have the access you need now?
RyanJJones10 (Thu, 03 Feb 2022 21:57:35 GMT):
It doesn't seem to be the case in Zenhub. I have the access in GH but not ZH - https://help.zenhub.com/support/solutions/articles/43000037261-github-permissions-and-access-in-zenhub
RyanJJones10 (Thu, 03 Feb 2022 21:57:46 GMT):
I could be doing something wrong so will do some more testing
shemnon (Thu, 03 Feb 2022 22:01:19 GMT):
I blame zenhub. I think you need plain old write as a github permission.
RyanJJones10 (Thu, 03 Feb 2022 22:01:42 GMT):
That is correct
RoboCopsGoneMad (Thu, 03 Feb 2022 22:04:36 GMT):
ahhhh its a zenhub thing
RyanJJones10 (Thu, 03 Feb 2022 22:27:15 GMT):
Definitely a ZH thing. Any ideas on how to get me write access in ZH then?
rjones (Fri, 04 Feb 2022 01:35:04 GMT):
I set the access from `Maintain` to `Write`
macfarla (Fri, 04 Feb 2022 03:02:15 GMT):
I remember this is why we ended up making @SajidaZ and @VijayMichalik "non-code" maintainers, because there didn't seem to be a way around it at the time. To move tickets between pipelines in zenhub, you need access to change the status of the ticket.
macfarla (Fri, 04 Feb 2022 03:11:37 GMT):
I remember this is why we ended up making SajidaZ and VijayMichalik "non-code" maintainers, because there didn't seem to be a way around it at the time. To move tickets between pipelines in zenhub, you need access to change the status of the ticket. But if there is a way around it, that's even better!
macfarla (Fri, 04 Feb 2022 03:11:37 GMT):
I remember this is why we ended up making @SajidaZ and @VijayMichalik "non-code" maintainers, because there didn't seem to be a way around it at the time. To move tickets between pipelines in zenhub, you need access to change the status of the ticket. But if there is a way around it, that's even better!
RoboCopsGoneMad (Fri, 04 Feb 2022 16:33:00 GMT):
fyi, unit testing with dagger is kinda complicated. i haven't formed an opinion on it yet
RoboCopsGoneMad (Fri, 04 Feb 2022 16:33:01 GMT):
https://dagger.dev/dev-guide/testing.html
shemnon (Fri, 04 Feb 2022 16:39:54 GMT):
Yea, we will still need unit, integration, and acceptance tests. Hard to shake that. It also will complicate our unit and integration story. That figures into why I never had time to take a serious swing at it.
RoboCopsGoneMad (Fri, 04 Feb 2022 16:57:34 GMT):
i think it's ok so far. I'm still unclear how much cross-dependency happens when wiring up components. I think you have to specify both how something is created, and what classes it can be injected into, which gives me pause
RoboCopsGoneMad (Fri, 04 Feb 2022 16:57:56 GMT):
on the one hand, i like being forced to draw an explicit dependency graph
RoboCopsGoneMad (Fri, 04 Feb 2022 16:58:45 GMT):
on the other hand, then the wiring diagram itself becomes rigid as you decouple dependencies. my gut says I'm ok with that tradeoff.
RoboCopsGoneMad (Fri, 04 Feb 2022 17:00:01 GMT):
the documentation is kinda meh, and you're gonna get a lot of android hits when searching for stuff
RoboCopsGoneMad (Fri, 04 Feb 2022 17:00:47 GMT):
but I am making progress, so far so good, 7/10 would buy again
RoboCopsGoneMad (Fri, 04 Feb 2022 23:41:43 GMT):
re: dagger. how do we feel about having a context like component that needs to be passed around to get dependencies from.
RoboCopsGoneMad (Fri, 04 Feb 2022 23:43:02 GMT):
There are static accessors that can be used to get dependencies, but those are isolated dep graphs, that are recomposed at the point the are staticly accessed. Ok for somethings, not ok for stateful things
atoulme (Fri, 04 Feb 2022 23:47:59 GMT):
is this to avoid the huge spaghetti bowl of BesuCommand?
atoulme (Fri, 04 Feb 2022 23:48:28 GMT):
dumb question, sorry - is Dagger working with the generic DI annotations
atoulme (Fri, 04 Feb 2022 23:48:28 GMT):
dumb question, sorry - is Dagger working with the generic DI annotations? I know Java built some to quell the Spring folks
RoboCopsGoneMad (Fri, 04 Feb 2022 23:48:37 GMT):
for instance, configuration state. Lets say DefaultBlockchain needs to know if `--Xmerge-enabled` was flipped. If you get the MergeConfiguration using the static factories that Dagger produces. the singleton scope of the configuration object is relative to that newly created component (whole new object graph). It will be totally disconnected from one created by BesuControllerBuilder. I think the way to solve this is making sure every new one is produced by MergeOptions, but for cases where you can't produce the new thing from scratch, you have to pass around some sort of UrComponent.
RoboCopsGoneMad (Fri, 04 Feb 2022 23:48:58 GMT):
yes, this is to address things like besu command
RoboCopsGoneMad (Fri, 04 Feb 2022 23:49:48 GMT):
it uses some JSR 330 annotations, but those are not sufficient. Realistically you'll need a mix of those and Daggers oen
atoulme (Fri, 04 Feb 2022 23:50:04 GMT):
got it
atoulme (Fri, 04 Feb 2022 23:50:25 GMT):
I mean, it doesn't add business value, it's just tech debt - so it's been kinda hard to get at it
shemnon (Fri, 04 Feb 2022 23:50:31 GMT):
If dagger is "done right" after the assembly there should be no need for static accessors. I think the idea is if you need a component you ask for a `Provider
shemnon (Fri, 04 Feb 2022 23:50:31 GMT):
If dagger is "done right" after the assembly there should be no need for context accessors. I think the idea is if you need a component you ask for a `Provider
RoboCopsGoneMad (Fri, 04 Feb 2022 23:50:53 GMT):
totally right, i've made a singleton with nothing static
shemnon (Fri, 04 Feb 2022 23:51:31 GMT):
so you don't pass around a single context like component, you inject the specific providers needed.
RoboCopsGoneMad (Fri, 04 Feb 2022 23:51:32 GMT):
however, we need an incremental path, I don't think Dagger All The Things is gonna be realistic on a long lived branch
RoboCopsGoneMad (Fri, 04 Feb 2022 23:52:12 GMT):
right, that would work fine if an object was fully managed
shemnon (Fri, 04 Feb 2022 23:52:14 GMT):
Where would we need the context like component? Is this from "outside" the dagger injected systems
shemnon (Fri, 04 Feb 2022 23:52:14 GMT):
Where would we need the context like component? Is this from "outside" the dagger injected systems?
RoboCopsGoneMad (Fri, 04 Feb 2022 23:53:16 GMT):
we'd need it in cases where an object is not created by dagger
RoboCopsGoneMad (Fri, 04 Feb 2022 23:53:39 GMT):
If i I have dependencies, and I am not managed by dagger, I have 2 ways to get them
RoboCopsGoneMad (Fri, 04 Feb 2022 23:54:09 GMT):
1) have a component (basically any graph of dependent objects) passed to me, where I can ask it for the things I need.
shemnon (Fri, 04 Feb 2022 23:54:36 GMT):
I think the component generated by dagger would be that context component.
RoboCopsGoneMad (Fri, 04 Feb 2022 23:54:46 GMT):
2) Statically create my own new graph. That will use the configured providers. If all of those are sufficient, great
RoboCopsGoneMad (Fri, 04 Feb 2022 23:54:51 GMT):
yes, thats right
shemnon (Fri, 04 Feb 2022 23:55:52 GMT):
So that component could be long lived, but we can get fresh pieces as the system goes on. It will hide the tracking of the singletons and create fresh dependenceis as we need them
shemnon (Fri, 04 Feb 2022 23:56:00 GMT):
What classes did you have in mind?
shemnon (Fri, 04 Feb 2022 23:56:15 GMT):
that would be created throught the life of the system
RoboCopsGoneMad (Fri, 04 Feb 2022 23:57:32 GMT):
one sec, let me try to avoid having to do that first, listen to this
RoboCopsGoneMad (Fri, 04 Feb 2022 23:58:21 GMT):
I wanna try it with the Configuration objects, because I think I can make a Module/Provider that lives in the CLI package, and creates a new Configuration object for anything that needs it. These wouldn't be singletons, but that wouldn't matter, because the Options don't change, and can just be re-read from the provider.
RoboCopsGoneMad (Fri, 04 Feb 2022 23:58:59 GMT):
so in that case, it would be fine to say "gimme a config graph" and it's brand new, but generated from the respective options.
shemnon (Fri, 04 Feb 2022 23:59:13 GMT):
Then we can move the CLI flags into modules too...
RoboCopsGoneMad (Fri, 04 Feb 2022 23:59:42 GMT):
exactly. the first impl is just annotating an option, and making the `toDomainObject` the providing method
shemnon (Sat, 05 Feb 2022 00:00:24 GMT):
sounds reasonable.
RoboCopsGoneMad (Sat, 05 Feb 2022 00:02:36 GMT):
right, it'll be cool if it works. now wrt incrementally applying this to stuff... I'm imagining a BesuComponent that gets passed into constructors, and added to as dependent classes are updated. Eventually a constructor shrinks down to that being the only parameter, at which point you should be able to treat it as fully managed, and then let Dagger inject each one back into ... a new constructor that looks just like the old one.
RoboCopsGoneMad (Sat, 05 Feb 2022 00:04:53 GMT):
realistically BesuComponent is just a root of other useful dependency graphs
GarySchulte (Sat, 05 Feb 2022 00:09:36 GMT):
is it fair to say that this strategy would not work for any component that has mutable state? Thinking about MergeContext of course
shemnon (Sat, 05 Feb 2022 00:09:56 GMT):
Rather than pass in the compoent why not have the component populate the constructor?
shemnon (Sat, 05 Feb 2022 00:10:41 GMT):
Eventually as higher level objects become injected we will see unneeded args in the constructor, and then trim those out./
RoboCopsGoneMad (Sat, 05 Feb 2022 00:10:55 GMT):
thats possible, as long as the thing it is populating from is managed
RoboCopsGoneMad (Sat, 05 Feb 2022 00:11:37 GMT):
so mutable state is ok.... as long as you are mindful of how the thing is scoped within the component
RoboCopsGoneMad (Sat, 05 Feb 2022 00:12:14 GMT):
dagger will manage singletons vs prototypes for you
RoboCopsGoneMad (Sat, 05 Feb 2022 00:13:38 GMT):
having dagger fill the constructor is fine, but everything that you expect to be passed in has to be managed
RoboCopsGoneMad (Sat, 05 Feb 2022 00:14:32 GMT):
i think thats really an end state, not a starting point, which is what i'm talking through here. at some point I have to commit something, and it will look halfassed no matter what
GarySchulte (Sat, 05 Feb 2022 00:19:20 GMT):
looking forward to seeing your PR. will it be separate / ahead of the merge execution endpoint PR or embedded in it?
RoboCopsGoneMad (Sat, 05 Feb 2022 00:20:11 GMT):
I'm gonna collect this convo into a doc to go with a different PR into main. After I do this for the merge options, i'm gonna see if I can apply the lessons learned to the Jumpdest cache config I did this summer. I think that could be a good demo of the technique.
RoboCopsGoneMad (Sat, 05 Feb 2022 00:20:39 GMT):
this one will be in 2914-bisect
shemnon (Sat, 05 Feb 2022 00:21:23 GMT):
Random request, can this doc be on the hyperledger wiki?
shemnon (Sat, 05 Feb 2022 00:21:45 GMT):
Start in your personal space and then once "pretty" it can go into the main section
shemnon (Sat, 05 Feb 2022 00:23:12 GMT):
or a permissive hackmd. Hyperledger has made a request that google docs not be the means since they have lost "public" fabric design docs when corporate accounts were closed.
shemnon (Sat, 05 Feb 2022 00:23:12 GMT):
or a permissive hackmd. Hyperledger has made a request that google docs not be the doc of record since they have lost "public" fabric design docs when corporate accounts were closed.
RoboCopsGoneMad (Sat, 05 Feb 2022 00:24:57 GMT):
was gonna do it in HL wiki
RoboCopsGoneMad (Sat, 05 Feb 2022 00:25:18 GMT):
in the Design section
RoboCopsGoneMad (Sat, 05 Feb 2022 00:26:14 GMT):
thanks for the chat, back to work and catching up on Boba Fett with the kiddo
rjones (Sat, 05 Feb 2022 03:51:44 GMT):
[ ](https://chat.hyperledger.org/channel/besu-contributors?msg=Gwb9Wsgr8Scg9xsmA) thank you! Google Docs are also not available in some geos, and data loss has hit as you said from accounts closing
RoboCopsGoneMad (Sat, 05 Feb 2022 13:46:08 GMT):
Thought about this some more, and it won't work as long as Configurations are decoupled from Options. That is a one way dependency so the Module that @Provides the Configuration to the Component would need access to the Option. We either have to pass around a BesuComponent or move the PicoCLI annotations on to the Configuration.
RoboCopsGoneMad (Sat, 05 Feb 2022 13:53:18 GMT):
Maybe we can introduce a new root level gradle module that depends on both Options and Configurations, and let all app code depend on that, but wouldn't that be circular?
shemnon (Sat, 05 Feb 2022 19:31:48 GMT):
we probably need to re-architect it. CLI Flags are a module, configurations depend on the flags. (using `@Named` annotations)
shemnon (Sat, 05 Feb 2022 19:32:09 GMT):
CLI flags are multiple modules.
RoboCopsGoneMad (Sat, 05 Feb 2022 23:40:13 GMT):
ok, where do we put them to avoid the circular dependency?
shemnon (Sun, 06 Feb 2022 14:11:48 GMT):
Make CLI f;ags their own module - https://github.com/hyperledger/besu/blob/39ee371556749a7123630566ae22981454c6df65/ethereum/evmtool/src/main/java/org/hyperledger/besu/evmtool/EvmToolCommandOptionsModule.java#L36
shemnon (Sun, 06 Feb 2022 14:11:48 GMT):
Make CLI flags their own module - https://github.com/hyperledger/besu/blob/39ee371556749a7123630566ae22981454c6df65/ethereum/evmtool/src/main/java/org/hyperledger/besu/evmtool/EvmToolCommandOptionsModule.java#L36
shemnon (Sun, 06 Feb 2022 14:12:14 GMT):
Then add @Inject constructors to the Configuration objects.
shemnon (Sun, 06 Feb 2022 14:12:47 GMT):
Make multiple CLI flag modules, one per configuration object at least.
RoboCopsGoneMad (Sun, 06 Feb 2022 14:50:35 GMT):
k, thats what I had in mind. Would put the modules right in the same package with the Configuration
RoboCopsGoneMad (Mon, 07 Feb 2022 15:00:51 GMT):
gm besu
RoboCopsGoneMad (Mon, 07 Feb 2022 20:01:48 GMT):
anyone know why BesuController has a static inner Builder class that returns BesuControllerBuilders ? According to usage it seems like those are convenience methods used by tests, but I don't see why this isn't at least on BesuControllerBuilder
RoboCopsGoneMad (Mon, 07 Feb 2022 20:38:07 GMT):
it seems like it is being used as a factory to construct the subclasses of BesuController.
shemnon (Tue, 08 Feb 2022 15:32:58 GMT):
My take is that most of what the besu controller builder does will be subsumed by a dagger component. The logic would be inverted and we would used flags to pick modules rather than letting the builder pick the implementations. Possibly even in-module, have a super module that the provider returns what the main if blocks `fromGenesisConfig` spits out.
RoboCopsGoneMad (Tue, 08 Feb 2022 18:49:31 GMT):
same thinking here. I like the idea of having the genesis config help drive the module layout
atoulme (Tue, 08 Feb 2022 21:19:04 GMT):
I never really ascribed to the builder pattern. I find it rather annoying whenever a refactor is needed.
atoulme (Tue, 08 Feb 2022 21:19:26 GMT):
Just moving off that pattern would reduce some issues imo.
RoboCopsGoneMad (Tue, 08 Feb 2022 23:39:31 GMT):
hoo buddy are you gonna like @Provides in dagger
RoboCopsGoneMad (Tue, 08 Feb 2022 23:40:20 GMT):
WebSocketRequestHandlerTest is flaky. It's got some kind of concurrency issue
RoboCopsGoneMad (Tue, 08 Feb 2022 23:47:40 GMT):
I think there is a way to get some reporting out of CircleCI on which tests are flaky, which might be candidates to have dagger applied to them.
macfarla (Wed, 09 Feb 2022 03:37:28 GMT):
Flaky tests are reported here https://app.circleci.com/insights/github/hyperledger/besu/workflows/default/tests - the quorum ATs have been problematic lately and I don't know why
loziniak (Wed, 09 Feb 2022 14:35:24 GMT):
Hello. Is `create-contract-permission` branch of _permissioning-smart-contracts_ ready to use in production? I'll happily work on it under some guidance. What has to be done there? I can see no tests – I could write some, is that ok?
macfarla (Wed, 09 Feb 2022 18:20:16 GMT):
gm besu!
macfarla (Wed, 09 Feb 2022 18:20:26 GMT):
Do we have a planned date for 22.1.0 release yet?
macfarla (Wed, 09 Feb 2022 18:23:54 GMT):
It is not production ready. It was done as a spike for granular account permissioning. However in the end we decided that using OpenZeppelin access control and roles was a better solution for the use case we were looking at. We kept the code as a branch just in case but I would recommend looking into the OpenZeppelin stuff before considering going back to that branch.
macfarla (Wed, 09 Feb 2022 18:58:15 GMT):
@loziniak do you have a use case in mind for the permissioning, or are you just looking for something to work on?
macfarla (Wed, 09 Feb 2022 19:46:32 GMT):
Anyone know what is this lgtm alerts badge we have on besu repo? when I click on the link I get "page not found" https://lgtm.com/projects/g/hyperledger/besu/alerts/
shemnon (Wed, 09 Feb 2022 19:54:27 GMT):
It appears LGTM has data problems for besu in particular - https://lgtm.com/search?q=besu
loziniak (Wed, 09 Feb 2022 20:05:31 GMT):
Yes, I have a usecase. I'm running a private chain and want to disallow contract creation by users, just allow some admin account(s) to do this. Exactly the same as this person on StackExchange: https://ethereum.stackexchange.com/questions/97196/how-to-disable-deployment-of-smart-contracts-on-besu-private-network
loziniak (Wed, 09 Feb 2022 20:07:48 GMT):
I'll also check OZ access, thanks.
macfarla (Wed, 09 Feb 2022 20:09:11 GMT):
Yep I think OZ might make your life easier here rather than reinventing the wheel
rjones (Wed, 09 Feb 2022 21:02:23 GMT):
yeah, it works for the other besu repos
macfarla (Wed, 09 Feb 2022 21:55:24 GMT):
any ideas why it doesn't work for besu?
GarySchulte (Wed, 09 Feb 2022 22:03:36 GMT):
can we push a tag to hyperledger/besu repo outside of the pipeline? We'd like to tag the last `:merge` image that supports kintsugi as `:kintsugi`, and then drop support for the test net
GarySchulte (Wed, 09 Feb 2022 22:04:14 GMT):
the test net is not going to be up much longer, and building out shims for it are not a good use of time
shemnon (Wed, 09 Feb 2022 22:07:24 GMT):
I have never tried it, but I would imagine the permissions required for a tag are the same as for pushing to main.
shemnon (Wed, 09 Feb 2022 22:07:41 GMT):
since merge has no commit restrictions it should work.
macfarla (Wed, 09 Feb 2022 23:17:25 GMT):
would it make sense to create a pre-release in github using that new tag?
macfarla (Wed, 09 Feb 2022 23:17:25 GMT):
would it make sense to create a pre-release in github using that new tag? @GarySchulte
m.terry (Thu, 10 Feb 2022 03:20:25 GMT):
I've got a PR to fix an NPE in legacy mode: https://github.com/hyperledger/besu/pull/3390
There's a bit of existing code in this function that I'm not following: https://github.com/hyperledger/besu/pull/3390/files#diff-301e1731fa376a0f0177dbe3d4f5da63c86c9cb0adbc788687b3cc41340d49ecL69
Is there a reason why an empty ForkId list returns null in legacy mode (and causes an NPE), or a new ForkId in non-legacy mode? Can they both return new ForkIds? @abdelhamidbakhta @rai
rjones (Thu, 10 Feb 2022 06:46:33 GMT):
nope. I can't even load it in the admin UI to see what's wrong. File a support ticket with LGTM?
shemnon (Thu, 10 Feb 2022 15:55:24 GMT):
Let's do this! https://discord.gg/hyperledger
shemnon (Thu, 10 Feb 2022 15:56:11 GMT):
(Grace's task force recommendation passed the TSC)
grace.hartley (Thu, 10 Feb 2022 15:56:35 GMT):
:woo:
GarySchulte (Thu, 10 Feb 2022 22:41:24 GMT):
See you all on discord :beers: I am uninstalling rocketchat. full commit :sign_of_the_horns:
rjones (Fri, 11 Feb 2022 20:43:50 GMT):
this chat has moved to https://discord.gg/hyperledger
