| -@gerrit:opendev.org- Ian Wienand proposed: [zuul/zuul-jobs] 861924: stage-output: use links in docs https://review.opendev.org/c/zuul/zuul-jobs/+/861924 | 00:07 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/nodepool] 861947: Serve all paused handlers before unpausing https://review.opendev.org/c/zuul/nodepool/+/861947 | 08:02 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/nodepool] 861947: Serve all paused handlers before unpausing https://review.opendev.org/c/zuul/nodepool/+/861947 | 08:13 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Benedikt Löffler: [zuul/nodepool] 860470: Cleanup local builds without .d folder https://review.opendev.org/c/zuul/nodepool/+/860470 | 08:35 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/nodepool] 861947: Serve all paused handlers before unpausing https://review.opendev.org/c/zuul/nodepool/+/861947 | 09:55 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/nodepool] 861947: Serve all paused handlers before unpausing https://review.opendev.org/c/zuul/nodepool/+/861947 | 10:07 | |
| -@gerrit:opendev.org- Thomas Cardonne proposed wip: [zuul/zuul] 862134: [DNM] TEST zuul_stream: close streamer when task is skipped https://review.opendev.org/c/zuul/zuul/+/862134 | 12:42 | |
| -@gerrit:opendev.org- Thomas Cardonne proposed wip: [zuul/zuul] 862134: [DNM] TEST zuul_stream: close streamer when task is skipped https://review.opendev.org/c/zuul/zuul/+/862134 | 13:03 | |
| @jim:acmegating.com | Anyone know if Thomas Cardonne is on matrix? | 13:58 |
|---|---|---|
| -@gerrit:opendev.org- Thomas Cardonne proposed wip: [zuul/zuul] 862134: [DNM] TEST zuul_stream: close streamer when task is skipped https://review.opendev.org/c/zuul/zuul/+/862134 | 14:24 | |
| @jim:acmegating.com | I emailed them to try to understand what's going on with ^ | 14:26 |
| -@gerrit:opendev.org- Thomas Cardonne proposed wip: [zuul/zuul] 862134: [DNM] TEST zuul_stream: close streamer when task is skipped https://review.opendev.org/c/zuul/zuul/+/862134 | 14:46 | |
| -@gerrit:opendev.org- Thomas Cardonne proposed wip: [zuul/zuul] 862134: [DNM] TEST zuul_stream: close streamer when task is skipped https://review.opendev.org/c/zuul/zuul/+/862134 | 15:20 | |
| @newbie2024:matrix.org | Hi guys, | 16:38 |
| here is a newbie question. | ||
| I would like to trigger some actions when "a pipeline starts" (i.e. when the first job for a certain pipeline is about to start) and when "a pipeline ends" (the last job of a certain pipeline finished). | ||
| I could think of the following: | ||
| - add some pre-pipeline and post-pipeline job, make all the job depend from the pre-pipeline (is that even possible without adding a lot of dependencies: directives?), make post-pipeline depend on all jobs (sounds like a nightmare to maintain even if it works | ||
| - configure some reporter like MQTT to report when a pipeline start/end | ||
| @newbie2024:matrix.org | Any suggestions on how to approach this? | 16:38 |
| @avass:vassast.org | newbie2024: i think i would likely use mqtt to make sure it doesn't affect jobs. But of course it depends on what you're trying to achieve | 16:46 |
| @avass:vassast.org | It also decouples zuul from the service carrying out the actions so it's easier to hook up other services in the future if they need to listen to events from zuul | 16:49 |
| @newbie2024:matrix.org | > <@avass:vassast.org> It also decouples zuul from the service carrying out the actions so it's easier to hook up other services in the future if they need to listen to events from zuul | 17:00 |
| The idea is to collect some metrics about pipelines. | ||
| I read latest and greatest version of Zuul come with nice Prometheus and Statd metrics, but 1. in the short term a Zuul upgrade is not an option and 2. not totally sure you get that granularity level that allow you to monitor pipeline latency, etc. | ||
| @avass:vassast.org | newbie2024: yup we used mqtt to collect metrics as well | 17:02 |
| @newbie2024:matrix.org | > <@avass:vassast.org> newbie2024: yup we used mqtt to collect metrics as well | 17:14 |
| Got it, thanks! | ||
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/nodepool] 861797: DNM checking docker image builds https://review.opendev.org/c/zuul/nodepool/+/861797 | 17:24 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/nodepool] 861797: DNM checking docker image builds https://review.opendev.org/c/zuul/nodepool/+/861797 | 18:34 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [zuul/nodepool] 861797: DNM checking docker image builds https://review.opendev.org/c/zuul/nodepool/+/861797 | 21:10 | |
| @jim:acmegating.com | i believe https://bugs.launchpad.net/ubuntu/+source/git/+bug/1993586 is affecting us; a local test run just failed with that | 21:20 |
| @jim:acmegating.com | i think the workaround described in that bug report will probably work for us in the unit test, so i'll work on adding that | 21:21 |
| @jim:acmegating.com | (it fails tests.unit.test_merger_repo.TestMergerRepo.test_ensure_cloned but i suspect the issue is related strictly to our unit test framework) | 21:22 |
| @clarkb:matrix.org | fwiw I think that may be affecting the pip tests too | 21:26 |
| @jim:acmegating.com | i just need to figure out how to get a "-c" into the git command :) | 21:28 |
| @clarkb:matrix.org | I wonder what symlinks we've got? | 21:34 |
| @clarkb:matrix.org | seems like not having symlinks might be a good way to avoid this, but maybe they are there by default? I think HEAD for example is a symlink? so maybe the distinction in the git update isn't really helpful | 21:34 |
| @iwienand:matrix.org | after more than a bit of head-scratching, i think i've narrowed down our zuul-jobs documentation issues to probably a docutils bug with the way it detects recursive includes | 21:38 |
| @iwienand:matrix.org | https://paste.opendev.org/show/brTxHYllHebBIs1wHbfM/ | 21:38 |
| @iwienand:matrix.org | ^ before i send it ... do we feel like that is understandable as a replication? | 21:38 |
| @clarkb:matrix.org | ianw: I haevn't followed that as I'ev been distracted so my input isn't super helpful. But you might want to have an assert or similar in that script to show what you expect the behavior to be and call out how it is buggy? | 21:41 |
| @iwienand:matrix.org | oh sorry, yeah the bug is that this fails with a warning that there is a recursive include, which i don't believe there is | 21:41 |
| @jim:acmegating.com | Clark: we don't have symlinks, we have a git submodule with a file:// prefix which is no longer allowed by default | 21:45 |
| @jim:acmegating.com | ianw: looks comprehensible :) | 21:46 |
| @clarkb:matrix.org | oh! I see | 21:46 |
| @clarkb:matrix.org | symlinks in objects is another way to trip this I guess | 21:46 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: | 21:49 | |
| - [zuul/zuul] 862192: Fix test_ensure_cloned after git CVE-2022-39253 https://review.opendev.org/c/zuul/zuul/+/862192 | ||
| - [zuul/zuul] 862193: Fix implied branch matchers and override-checkout https://review.opendev.org/c/zuul/zuul/+/862193 | ||
| @jim:acmegating.com | Clark: yeah, i think either way would do it. | 21:49 |
| @jim:acmegating.com | there's the magic incantation to gitpython to fix that. | 21:50 |
| @jim:acmegating.com | zuul-maint: https://review.opendev.org/862192 is necessary to unblock zuul tests due to a git CVE | 21:50 |
| @jim:acmegating.com | zuul-maint: https://review.opendev.org/860989 is the next change needed for 8.0.0 | 21:51 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 862194: Defensively handle null JobData hashes https://review.opendev.org/c/zuul/zuul/+/862194 | 21:57 | |
| @iwienand:matrix.org | > <@iwienand:matrix.org> ^ before i send it ... do we feel like that is understandable as a replication? | 22:31 |
| https://sourceforge.net/p/docutils/bugs/459/ | ||
| @jim:acmegating.com | when it rains it pours... now there's something with jwt | 22:52 |
| @jim:acmegating.com | i'm trying to track down what changed | 22:52 |
| @jim:acmegating.com | pyjwt 2.5.0 .. 2.6.0 | 22:56 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: | 23:01 | |
| - [zuul/zuul] 862192: Fix test_ensure_cloned and JWT tests https://review.opendev.org/c/zuul/zuul/+/862192 | ||
| - [zuul/zuul] 862193: Fix implied branch matchers and override-checkout https://review.opendev.org/c/zuul/zuul/+/862193 | ||
| @jim:acmegating.com | Clark: tristanC ^ 862192 is updated with a second urgent fix | 23:02 |
| @jim:acmegating.com | (if anyone has time to look into/fix the pyjwt issue and unpin, feel free!) | 23:02 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!