01:11:11  * jaywonquit (Remote host closed the connection)
01:25:09  * jaywonjoined
01:53:43  * gabrielschulhofjoined
05:00:12  * gabrielschulhofquit (Ping timeout: 245 seconds)
06:09:47  * jaywonquit (Remote host closed the connection)
06:18:20  * jaywonjoined
06:22:25  * jaywonquit (Ping timeout: 248 seconds)
06:56:51  * jaywonjoined
07:01:45  * jaywonquit (Ping timeout: 265 seconds)
07:43:31  * seishunjoined
07:59:02  * jaywonjoined
08:03:53  * jaywonquit (Ping timeout: 276 seconds)
09:00:26  * jaywonjoined
09:05:05  * jaywonquit (Ping timeout: 255 seconds)
10:13:01  * jaywonjoined
10:17:30  * jaywonquit (Ping timeout: 265 seconds)
10:25:09  * mylesborinsquit (Quit: farewell for now)
10:25:40  * mylesborinsjoined
11:00:21  * juggernaut451joined
11:01:25  * juggernaut451quit (Client Quit)
12:09:16  * gabrielschulhofjoined
12:17:39  * jaywonjoined
12:22:35  * jaywonquit (Ping timeout: 276 seconds)
12:44:12  <lucalanziani>Do you know why we are using the type-provider-os-arch-uid format to define hosts in the inventory and set hostvars instead of trying to get those info from the hosts with ansible setup?
12:46:15  * gabrielschulhofquit (Ping timeout: 260 seconds)
12:56:05  * node-ghjoined
12:56:05  * node-ghpart
13:20:26  * jaywonjoined
13:25:05  * jaywonquit (Ping timeout: 276 seconds)
13:40:23  * node-ghjoined
13:40:23  * node-ghpart
13:57:52  * gabrielschulhofjoined
14:21:05  <maclover7>lucalanziani: I think it was chosen at the time to simplify the data setup
14:21:19  <maclover7>since all of our infra is donations, it's easier to keep track of who donated what
14:22:24  * jaywonjoined
14:25:52  * 14WAAEAXPjoined
14:25:52  * 14WAAEAXPpart
14:25:52  * 07IAC6ILAjoined
14:25:52  * 07IAC6ILApart
14:26:25  * node-ghjoined
14:26:25  * node-ghpart
14:27:02  * jaywonquit (Ping timeout: 260 seconds)
14:27:52  * node-ghjoined
14:27:52  * node-ghpart
14:34:26  * node-ghjoined
14:34:26  * node-ghpart
14:34:40  * node-ghjoined
14:34:40  * node-ghpart
14:36:45  * node-ghjoined
14:36:45  * node-ghpart
14:39:15  * node-ghjoined
14:39:15  * node-ghpart
14:39:25  * node-ghjoined
14:39:25  * node-ghpart
14:41:37  * node-ghjoined
14:41:37  * node-ghpart
14:41:52  * node-ghjoined
14:41:52  * node-ghpart
14:41:59  <maclover7>refack rvagg do you know where the version selector groovy script is being called from in jenkins now?
14:42:04  <maclover7>I can't seem to find it on the subjobs
14:42:09  <maclover7>(of node-test-commit)
14:49:38  <maclover7>ahh nvm, found it
14:56:44  * node-ghjoined
14:56:44  * node-ghpart
15:01:08  * node-ghjoined
15:01:08  * node-ghpart
15:27:55  <refack>maclover7: I did some digging at the time and got this script for finding all jobs that use groovy scripts -https://github.com/nodejs/build/pull/1217#issuecomment-393688871
15:27:55  <refack>Based on https://github.com/samrocketman/jenkins-script-console-scripts/blob/master/find-all-freestyle-jobs-using-shell-command.groovy
15:28:28  <refack>It might be helpfull with future digging...
15:49:35  <maclover7>refack: ahh very cool
15:49:40  <maclover7>that's good to figure out what's running what
15:56:34  * jaywonjoined
16:01:38  * jaywonquit (Ping timeout: 276 seconds)
16:25:22  * node-ghjoined
16:25:22  * node-ghpart
16:30:42  * node-ghjoined
16:30:42  * node-ghpart
16:32:56  * node-ghjoined
16:32:57  * node-ghpart
17:19:46  * jaywonjoined
17:24:17  * jaywonquit (Ping timeout: 265 seconds)
18:03:01  * node-ghjoined
18:03:01  * node-ghpart
18:30:10  <maclover7>joyee: you around?
18:30:33  <joyee>maclover7: yep
18:30:59  <maclover7>joyee: just quickly re: https://github.com/nodejs/commit-queue/issues/1#issuecomment-396069666
18:31:02  <maclover7>and also the reliability repo
18:31:33  <joyee>Yes?
18:31:42  <maclover7>I know there's some frustration about how the build wg current functions and such, but things like tracking CI infra/commit queue/Jenkins/etc
18:31:55  <maclover7>should in theory come before the wg for at least discussion/vote before things are acted upon
18:32:11  <maclover7>like with the commit queue I had done some previous work along with refac
18:32:31  <maclover7>and (I don't think anyone's trying to be mean/rude to others) but the build wg is being left out of these discussions
18:32:46  <joyee>So the two repos came out of the collab summit this May
18:33:05  <joyee>I think it’s generally because people were not aware?
18:33:14  <maclover7>Yeah, I think most of the recent issues are just because only one or two people from build wg were at the summit
18:33:26  <refack>What I worry about is fragmentation of attention and missing out on help just because people are not aware of stuff
18:33:47  <maclover7>It might be good at some point to have a build-wg-moving-forward meeting in the next few weeks
18:33:57  <joyee>refack: exactly
18:33:58  <maclover7>maybe not this week because it's wg meeting week and a packed agenda
18:34:02  <maclover7>but the week after maybe?
18:34:10  <refack>:+1
18:34:19  <joyee>+1
18:34:20  <maclover7>so everybody knows what's going on with everyone else
18:34:48  <maclover7>maybe may 19?
18:34:55  <joyee>refack: I think part of the fragmentation comes out of the build WG working on top many things
18:35:04  <maclover7>so it's on the usual wg meeting day
18:35:31  <maclover7>joyee: refack: I'm going to open up an issue on nodejs/build
18:35:36  <refack>I know that for the months I wasn't active maclover7 was basically the only active member and things
18:35:44  <joyee>It’s kind of the same situation with people pinging build to review makefile PRs
18:35:54  <maclover7>Yeah, I try and not take this stuff personally
18:35:58  <refack>I think that was just an old habbit
18:36:04  <maclover7>but lately I've been the only one with rich keeping jenkins online
18:36:18  <maclover7>so it's tough with all of this stuff flying around
18:36:33  <refack>And nigher you or me were at the summit...
18:36:43  <refack>*nighter
18:37:06  <joyee>That’s why I think it’s still a good idea to keep a separate repo
18:37:23  <joyee>Because you sometimes get lost in the build issues
18:37:54  <joyee>Only people who know about the issues know where to find them
18:38:01  <maclover7>yeah that's why I've been trying to clean out the build tracker
18:38:03  <maclover7>it's overflowing
18:39:01  <joyee>But if we have a something focusing on that effort, and make it say a strategic initiative, it may not get lost that easily
18:39:31  <refack>If we're talking technical tools IMHO the second best thing is project boards.
18:39:31  <refack>But then only repo members can do stuff
18:40:09  <refack>May be create a team and have team discussions... I don't know...
18:40:29  <maclover7>joyee: what tz are you in?
18:40:51  <refack>(real TZ not the china general one ;)
18:40:53  <joyee>Team discussions are private so I think repo issues would be a bit better
18:41:04  <joyee>UTC+8
18:41:09  <refack>offff, github...
18:41:16  <joyee>But I am still awake at 2:40 now
18:41:19  <joyee>lol
18:42:15  <maclover7>lol
18:44:14  <maclover7>joyee: refack: https://github.com/nodejs/build/issues/1330
18:44:17  * node-ghjoined
18:44:17  * node-ghpart
18:44:22  <maclover7>working on the doodle now
18:45:57  <joyee>Maybe Trott will be interested since he created the commit queue repo
18:46:22  <maclover7>yeah rich has been doing a lot of jenkins stuff lately
18:48:05  <ryzokuken>parallel/test-eslint-require-buffer isn't a flake?
18:48:23  <joyee>I can stay up pretty late if need be, I work remotely with mostly European these days anyways. Just need to plan ahead because sometimes I do need to get up early to go out and do paperwork...like tomorrow (but then I am still awake, eh
18:48:24  * node-ghjoined
18:48:24  * node-ghpart
18:49:33  <ryzokuken>It just failed on me, it probably has become flaky because (A) No record of its flakiness (B) It just failed on my the way flakes do.
18:49:50  <refack>ryzokuken: link to job?
18:50:11  <ryzokuken>refack: take a look at https://ci.nodejs.org/job/node-test-commit-linuxone/2093/
18:50:17  <ryzokuken>LMK what you think.
18:50:25  <refack>it's very strange since it just tests a linting tool, so might be sensitive to tree changes...
18:53:22  <ryzokuken>but it didn't fail previously on the same PR.
18:53:38  <ryzokuken>maybe someone spoiled master?
18:53:50  <refack>looks like some sort of flakiness
18:54:01  <refack>since it passes on most other platforms
18:54:17  <refack>But it's very strange
18:54:51  <ryzokuken>indeed.
18:54:55  <ryzokuken>should I rebuild?
18:55:46  <maclover7>joyee: ok, np
18:55:55  <maclover7>tried to provide options that might work for utc+8
19:03:59  <joyee>maclover7: thanks, I've filled the doodle. If there are not enough good slots available, I can pick some earlier times later
19:05:35  <joyee>ryzokuken: can't see any similar failures in previous runs, so yeah maybe try rebuilding
19:05:53  <ryzokuken>joyee started, thanks.
19:09:17  <ryzokuken>joyee refack fails again on the same commit.
19:09:24  <ryzokuken>sorry, same test file
19:09:26  <ryzokuken>https://ci.nodejs.org/job/node-test-commit-linuxone/2095/label=rhel72-s390x/testReport/junit/(root)/test/parallel_test_eslint_require_buffer/
19:13:06  <ryzokuken>it's still not failing on any other platform though.
19:13:37  <joyee>I think it may has something to do with thePR
19:13:42  <joyee>It changes assert
19:13:54  <joyee>and is being backported to v9.x which may have a different eslint
19:14:10  <joyee>(wait why are we backporting to v9.x?
19:14:18  <ryzokuken>lemme check
19:15:04  <joyee>9.x is a few weeks to EOL anyway
19:15:24  <ryzokuken>the PR was created a week or two ago.
19:15:28  <ryzokuken>is it in maintainance?
19:15:59  <ryzokuken>it'll probably be a major bummer for a new backporter if their first backport wasn't merged, I guess.
19:16:03  <ryzokuken>idk, should we?
19:20:24  <maclover7>joyee: re https://github.com/orgs/nodejs/teams/collaborators/discussions/30?from_comment=5#discussion-30-comment-5, best way to view the permissions is via the secrets repo
19:20:35  <maclover7> / the ansible inventory
19:20:49  <maclover7>that's why each of the sections in the inventory are named infra/release/test
19:21:02  <joyee>That's what I thought, but then what's the point of that section...?
19:21:14  <maclover7>in the readme?
19:21:14  <joyee>in the readme
19:21:48  <joyee>Yes
19:21:54  <maclover7>for the readme, wg members == test, infra admins == infra, release == release
19:22:07  <maclover7>in theory release and infra should be the same people, but that's a legacy thing
19:22:33  <joyee>Also, having the access probably !== wanting to be pinged...I guess the same can be said about the README, but being in the README kind of implies that you want to be pinged
19:22:57  <maclover7>in theory if you have the access, it's because you want to help
19:23:11  <maclover7>I mean I hope someone wouldn't use the root access to mine bitcoin or whatever lol
19:23:39  <maclover7>nodejs/build is the right group to ping for issues with jenkins test cluster
19:23:43  <maclover7>just not many people respond to the ping
19:23:47  <maclover7>as in like one or two people
19:23:47  <joyee>I think quite a few people in the .gpg directory of test have not showed up in quite some time
19:23:59  <joyee>I could not recognize a lot of them
19:24:02  <maclover7>yeah I had been drafting emails to ask people if they still wanted to be involved
19:24:06  <maclover7>most of them I personally do not know
19:24:17  <maclover7>likely from the iojs days when all of the infra was first being setup
19:25:03  <joyee>But the build still has people from the github bot team
19:25:08  <joyee>which is kind of weird?
19:25:40  <maclover7>you mean the nodejs/build github team?
19:25:52  <maclover7>it's possible in the past they had helped with the jenkins test cluster
19:25:53  <maclover7>idk
19:26:10  <maclover7>if I was an org admin, I'd use the new github teams stuff and create proper subteams in the ui
19:26:18  <maclover7>to make suff easier to understand
19:26:27  <joyee>That's what I tried to do..
19:26:38  <joyee>But I am kind of confused
19:26:43  <maclover7>Yeah, wasn't trying to be critical
19:26:47  <maclover7>exactly, it's just super confusing
19:27:03  <maclover7>like I only know this because of old github threads
19:27:10  <maclover7>which is, uh, not great
19:27:23  <joyee>So what do you think about the subteams?
19:27:35  <joyee>I guess build-files make sense because they probably don't belong to build
19:27:45  <maclover7>yeah build-files is good to have separate
19:27:59  <maclover7>like I know zero about makefiles but can definitely help fixup linux boxes
19:28:08  <maclover7>the problem is that nodejs/build has many inactive members
19:28:13  <maclover7>that's why no one responds to the pings
19:28:22  <maclover7>not because pinging the team itself is a bad idea
19:29:09  <joyee>Yeah that's what I thought. If pings are not effective, then we need to redesign the structure
19:30:00  <joyee>Maybe we should discuss about this in this week's build meeting?
19:30:07  <joyee>Probably won't take long anyways
19:32:34  <maclover7>hmm maybe include as part of the separate meeting
19:32:38  <maclover7>this week is already pretty jammed
19:32:49  <maclover7>feel free to add to the agenda
19:48:27  * node-ghjoined
19:48:27  * node-ghpart
19:50:15  * node-ghjoined
19:50:15  * node-ghpart
19:50:27  * node-ghjoined
19:50:27  * node-ghpart
19:54:10  * node-ghjoined
19:54:10  * node-ghpart
19:54:34  * node-ghjoined
19:54:34  * node-ghpart
19:54:41  * node-ghjoined
19:54:41  * node-ghpart
19:55:00  * node-ghjoined
19:55:00  * node-ghpart
19:55:57  * node-ghjoined
19:55:57  * node-ghpart
19:56:19  * node-ghjoined
19:56:19  * node-ghpart
19:57:14  <joyee>maclover7: yeah, I've listed some reference in https://github.com/nodejs/build/issues/1331 and now I kind of start worrying about the access...
19:57:52  <joyee>This is more serious than the inactive collaborator issue, because there's actual access to machines and credentials involved
20:03:34  <joyee>Oh and then...there is also nodejs/jenkins-admins
20:05:36  * node-ghjoined
20:05:36  * node-ghpart
20:09:49  <maclover7>joyee: yeah I agree
20:10:04  <maclover7>part of the issue is that we have a script to recycle keys, but no one's used it in years
20:10:13  <maclover7>so worries about breaking machines and such
20:12:41  * node-ghjoined
20:12:41  * node-ghpart
20:16:58  * node-ghjoined
20:16:59  * node-ghpart
20:17:28  <joyee>maclover7: I've read the access.md and there is also a jenkins-admins, not sure if it's useful on its own anymore?
20:17:45  <joyee>It's not a subteam of build either
20:18:51  * node-ghjoined
20:18:51  * node-ghpart
20:19:35  * node-ghjoined
20:19:36  * node-ghpart
20:20:02  * node-ghjoined
20:20:02  * node-ghpart
20:35:58  * node-ghjoined
20:35:58  * node-ghpart
20:40:06  * node-ghjoined
20:40:06  * node-ghpart
20:44:29  * seishunquit (Ping timeout: 256 seconds)
21:23:23  <refack>maclover7: FYI when you create (or duplicate) a new Jenkins item uncheck "add this to this view" https://usercontent.irccloud-cdn.com/file/rOZAwx3s/image.png
21:24:02  <refack>For some reason Jenkins thinks it a good idea to add all those jobs to the main view... 🤷
21:29:38  * node-ghjoined
21:29:38  * node-ghpart
21:32:30  * node-ghjoined
21:32:30  * node-ghpart
21:34:38  * seishunjoined
21:39:30  * node-ghjoined
21:39:30  * node-ghpart
21:41:29  * node-ghjoined
21:41:29  * node-ghpart
21:43:33  * seishunquit (Ping timeout: 264 seconds)
22:15:35  * gabrielschulhofquit (Ping timeout: 260 seconds)
23:22:56  * gabrielschulhofjoined