| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] Don't add node resources to nonexistent tenant https://review.opendev.org/c/zuul/zuul/+/811800 | 00:28 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Joshua Watt: [zuul/zuul] gitlabconnection: Use keyword argument for retries https://review.opendev.org/c/zuul/zuul/+/811204 | 00:28 | |
| @icey:matrix.org | my 3.19.1 Zuul is still doing strange things with how long things are taking, I have two jobs that showed up in the Zuul status page around an hour ago, but Zuul still hasn't populated any actual jobs for them | 07:32 |
|---|---|---|
| -@gerrit:opendev.org- Zuul merged on behalf of Matthieu Huin https://matrix.to/#/@mhuin:matrix.org: [zuul/zuul] Web UI: add duration info on buildset page https://review.opendev.org/c/zuul/zuul/+/811752 | 07:55 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] wip: show different role use on change vs. tag https://review.opendev.org/c/zuul/zuul/+/811887 | 08:12 | |
| @avass:vassast.org | icey: that sounds like an event reached the scheduler but it's still waiting for a merge job to complete | 08:34 |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] wip: show different role use on change vs. tag https://review.opendev.org/c/zuul/zuul/+/811887 | 09:45 | |
| @icey:matrix.org | > <@avass:vassast.org> icey: that sounds like an event reached the scheduler but it's still waiting for a merge job to complete | 09:46 |
| there are no merge jobs? that job is now up to 3.5 hours, and there are like 7 jobs after it with no progress on any of them :-/ | ||
| @icey:matrix.org | I'm restarting my zuul :-/ | 09:49 |
| @j.campos:matrix.org | Hi, is there a way to make a shallow clone on Zuul's checkout step? I couldn't find anything in the docs related to this. | 11:05 |
| -@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] Web UI: add builds timeline on buildset page https://review.opendev.org/c/zuul/zuul/+/811753 | 11:20 | |
| @mhuin:matrix.org | > <@jim:acmegating.com> mhu: 811752+3 with note; suggestion on 811753 but i love it. | 11:34 |
| thanks! I addressed your suggestions on https://review.opendev.org/c/zuul/zuul/+/811753 and added some questions too | ||
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] Pass ZK context to deserialize method of ZKObjects https://review.opendev.org/c/zuul/zuul/+/811955 | 12:33 | |
| -@gerrit:opendev.org- Felix Edel proposed: [zuul/zuul] Flatten SourceContext data structure https://review.opendev.org/c/zuul/zuul/+/811967 | 13:33 | |
| -@gerrit:opendev.org- Felix Edel proposed: [zuul/zuul] WIP Put frozen jobs in ZooKeeper https://review.opendev.org/c/zuul/zuul/+/811969 | 13:35 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley: [zuul/zuul-jobs] Stop testing playbooks/roles on Ubuntu Xenial https://review.opendev.org/c/zuul/zuul-jobs/+/809899 | 14:51 | |
| @clarkb:matrix.org | Jose Campos: Zuul relies on being able to do merge operations to build speculative future states. For this reason using shallow clones would be difficult. Instead the Zuul executors and mergers cache the repos to ensure their operations are quick. Then on the test nodes you could theoretically push a shallow state to them maybe? In OpenDev we maintain a git repo cache on our test node images that we can update to reduce time spent pushing code | 14:51 |
| @clarkb:matrix.org | A non zero number of opendev jobs do things like merges and tags too which I think shallow clones would also complicate | 14:52 |
| -@gerrit:opendev.org- Artem Goncharov proposed: [zuul/zuul] Fix failed merge detection in gitlab https://review.opendev.org/c/zuul/zuul/+/806230 | 15:10 | |
| -@gerrit:opendev.org- Artem Goncharov proposed: [zuul/zuul] Add support for gitlab squash merge https://review.opendev.org/c/zuul/zuul/+/806231 | 15:10 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley: [zuul/zuul-jobs] Pin to funcparserlib prerelease for new SetupTools https://review.opendev.org/c/zuul/zuul-jobs/+/809885 | 15:35 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley: [zuul/zuul-jobs] Add tox_config_file rolevar to tox https://review.opendev.org/c/zuul/zuul-jobs/+/806613 | 15:37 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley: | 15:51 | |
| - [zuul/zuul-jobs] Support verbose showconfig in tox siblings https://review.opendev.org/c/zuul/zuul-jobs/+/806621 | ||
| - [zuul/zuul-jobs] Include tox_extra_args in tox siblings tasks https://review.opendev.org/c/zuul/zuul-jobs/+/806612 | ||
| - [zuul/zuul-jobs] Explicit tox_extra_args in zuul-jobs-test-tox https://review.opendev.org/c/zuul/zuul-jobs/+/809456 | ||
| @fungicide:matrix.org | the dependency for https://review.opendev.org/810299 if anybody has time to review some policy documentation additions for zuul-jobs | 16:28 |
| @fungicide:matrix.org | * the dependency for https://review.opendev.org/810299 has merged if anybody has time to review some policy documentation additions for zuul-jobs | 16:30 |
| @jim:acmegating.com | fungi: thanks! lgtm | 16:31 |
| @fungicide:matrix.org | thanks for looking! | 16:51 |
| @avass:vassast.org | fungi: lgtm | 16:54 |
| @avass:vassast.org | I suppose we want more eyes on that? or should I +3 and let 810299 gate? | 16:56 |
| @jim:acmegating.com | i think we're probably good to go since so many people reviewed the original change to remove support. | 16:57 |
| @avass:vassast.org | that's what I thought, but wasn't sure | 16:57 |
| @avass:vassast.org | gave it a +w | 16:58 |
| @homeski:matrix.org | Hi there | 17:14 |
| @homeski:matrix.org | I have a job that calls a playbook that makes use of a collection from galaxy `import_role: name.collection.role` | 17:15 |
| @homeski:matrix.org | I couldn't find a way to get zuul-executor to install the collection | 17:16 |
| @homeski:matrix.org | I was able to install the collection locally, into the playbook repository, and that allows the zuul job to run | 17:18 |
| @homeski:matrix.org | Are there any other options? | 17:18 |
| @homeski:matrix.org | Configuring `job.roles.`galaxy` did work: https://zuul-ci.org/docs/zuul/reference/job_def.html#attr-job.roles.galaxy | 17:19 |
| @homeski:matrix.org | I also tried configuring a trusted pre-run job that manually ran `ansible-galaxy collection install` but fell a bit short | 17:20 |
| @homeski:matrix.org | * Configuring `job.roles.`galaxy` did not work: https://zuul-ci.org/docs/zuul/reference/job_def.html#attr-job.roles.galaxy | 17:20 |
| @homeski:matrix.org | * I was able to install and commit the collection locally, into the playbook repository, and that allows the zuul job to run | 17:21 |
| @homeski:matrix.org | I think this is what I'm looking for: https://review.opendev.org/c/zuul/zuul/+/723071 but by now it is a 1+ year old WIP | 17:22 |
| @homeski:matrix.org | * Configuring `job.roles.galaxy` did not work: https://zuul-ci.org/docs/zuul/reference/job_def.html#attr-job.roles.galaxy | 17:24 |
| @jim:acmegating.com | homeski: that feature is not yet implemented. the change you linked would install the collection system-wide, but would not do so for an individual job. the only extant option is to do it within the job and run a nested ansible. | 17:25 |
| @jim:acmegating.com | homeski: or of course the vendor option that you have used. | 17:27 |
| @homeski:matrix.org | ok thanks, I took a look at zuul-discuss and I saw some discussion around running nested ansible like you mention | 17:27 |
| @homeski:matrix.org | is the general idea to run that nested ansible from the executor or from a node provided by nodepool? | 17:28 |
| @homeski:matrix.org | * is the general idea to run that nested ansible from the executor (trusted jobs) or from a node provided by nodepool (untrusted)? | 17:29 |
| @homeski:matrix.org | * is the general idea to run that nested ansible from the executor (trusted project) or from a node provided by nodepool (untrusted)? | 17:29 |
| @jim:acmegating.com | homeski: either can work, but running it on the executor requires a trusted context (config project). so running it from another node may be easier. | 17:30 |
| @homeski:matrix.org | ok, thanks. And finally is there a plan to add support for galaxy collections natively? | 17:32 |
| @homeski:matrix.org | The commit I referenced makes an attempt at it: https://review.opendev.org/c/zuul/zuul/+/723071 | 17:32 |
| @homeski:matrix.org | It doesn't **seem** like a huge amount of diff | 17:32 |
| @avass:vassast.org | homeski: I don't think I left that in a finished state however :) | 17:33 |
| @homeski:matrix.org | So I'm just curious collection support is planned, or not really needed due to the above work arounds | 17:33 |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley: [zuul/zuul-jobs] Deprecate EOL Python releases and OS versions https://review.opendev.org/c/zuul/zuul-jobs/+/810299 | 17:33 | |
| @homeski:matrix.org | * So I'm just curious if collection support is planned, or not really needed due to the above work arounds | 17:34 |
| @homeski:matrix.org | * So I'm just curious if collection support is planned, or not really needed due to the above work arounds, and thus not a priority | 17:34 |
| @homeski:matrix.org | * So I'm just curious if collection support is planned, or not really needed due to the above work arounds, and thus not a priority. The current state of ansible really has a huge dependency on being able to consume collections. | 17:36 |
| @homeski:matrix.org | Anyways, appreciate the help :) | 17:53 |
| @homeski:matrix.org | Been having fun trying to use zuul to test our playbooks | 17:54 |
| -@gerrit:opendev.org- Jeremy Stanley proposed: [zuul/zuul-jobs] Revert "Include tox_extra_args in tox siblings tasks" https://review.opendev.org/c/zuul/zuul-jobs/+/812001 | 17:57 | |
| @fungicide:matrix.org | expediting that ^ would be appreciated, i'll work on a way to test the corner case it uncovered | 18:00 |
| @jim:acmegating.com | homeski: that commit is not an attempt to implement the feature you're referencing; that would only install the collection system-wide, not for a job | 18:01 |
| @jim:acmegating.com | fungi: double single approved. :) | 18:07 |
| @homeski:matrix.org | thanks corvus | 18:07 |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley: [zuul/zuul-jobs] Revert "Include tox_extra_args in tox siblings tasks" https://review.opendev.org/c/zuul/zuul-jobs/+/812001 | 18:17 | |
| -@gerrit:opendev.org- Jeremy Stanley proposed: | 19:17 | |
| - [zuul/zuul-jobs] Test tox.requires doesn't break sibling installs https://review.opendev.org/c/zuul/zuul-jobs/+/812004 | ||
| - [zuul/zuul-jobs] Revert "Revert "Include tox_extra_args in tox siblings tasks"" https://review.opendev.org/c/zuul/zuul-jobs/+/812005 | ||
| @fungicide:matrix.org | those demonstrate the problem, and admitting you have a problem is the first step | 19:28 |
| @fungicide:matrix.org | unfortunately, we test the tox role by using the tox config for the zuul-jobs project itself, which means testing tox settings zuul-jobs doesn't need gets hairy | 19:30 |
| @fungicide:matrix.org | fortunately, my change to be able to specify an alternate tox config file path also merged today! so maybe i can leverage that to easily test with a non-default tox.ini | 19:30 |
| @fungicide:matrix.org | kill two birds with one test | 19:31 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] Handle long strings in build/buildsets tables https://review.opendev.org/c/zuul/zuul/+/812008 | 21:25 | |
| -@gerrit:opendev.org- Jeremy Stanley proposed: [zuul/zuul-jobs] Test tox.requires doesn't break sibling installs https://review.opendev.org/c/zuul/zuul-jobs/+/812004 | 21:47 | |
| -@gerrit:opendev.org- Jeremy Stanley proposed: [zuul/zuul-jobs] Revert "Revert "Include tox_extra_args in tox siblings tasks"" https://review.opendev.org/c/zuul/zuul-jobs/+/812005 | 21:48 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] Fix config error detection on same project branch https://review.opendev.org/c/zuul/zuul/+/812049 | 22:49 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] Fix config error detection on same project branch https://review.opendev.org/c/zuul/zuul/+/812049 | 22:51 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Felix Edel: [zuul/zuul] Flatten SourceContext data structure https://review.opendev.org/c/zuul/zuul/+/811967 | 22:52 | |
| @jim:acmegating.com | Clark, tobiash: https://review.opendev.org/812049 fixes a latent bug i introduced; you may remember some of the context there so i'd particularly welcome your reviews and whether you think this is the right fix. | 22:57 |
| @jim:acmegating.com | i'm slightly confused because it almost looks like we intended the other thing (based on the unit test, which had to be constructed in just that way to let the bug through), but all the words in the commit message and the code itself says we didn't intend it. so i'm giving more weight to that interpretation. | 22:58 |
| @clarkb:matrix.org | hrm my only half concern is that people could end up in a situation like the one we have with gerrit where you are foced to correct many hundreds of errors in order to correct any one | 23:26 |
| @clarkb:matrix.org | However, in the zuul case I think that is a lot simpler to do | 23:27 |
| @jim:acmegating.com | Clark: yeah - you should be able to do it in one change. | 23:27 |
| @jim:acmegating.com | (and zuul will actually tell you if you got it right, vs the other way where it might not) | 23:28 |
| @clarkb:matrix.org | And ya since this is only looking at a project branch you don't have to edit 3 different places to make it go away (gerrit could do this) | 23:29 |
| -@gerrit:opendev.org- Niklas Borg proposed: [zuul/zuul] Add GitLab timeout configuration https://review.opendev.org/c/zuul/zuul/+/812093 | 23:31 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!