| -@gerrit:opendev.org- Ian Wienand proposed: [zuul/zuul-jobs] 798969: Add introduction to the container jobs documentation https://review.opendev.org/c/zuul/zuul-jobs/+/798969 | 00:06 | |
| @iwienand:matrix.org | clarkb: ^ as discussed | 00:06 |
|---|---|---|
| -@gerrit:opendev.org- Felix Edel proposed: | 06:42 | |
| - [zuul/zuul] 827299: Fix hanging MySQL database tests https://review.opendev.org/c/zuul/zuul/+/827299 | ||
| - [zuul/zuul] 827215: Report NODE_FAILURES caused by node request failures to SQL database https://review.opendev.org/c/zuul/zuul/+/827215 | ||
| @felixedel:matrix.org | Clark: Thanks, I've updated the patches | 06:44 |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828224: Fix leaked semaphore cleanup https://review.opendev.org/c/zuul/zuul/+/828224 | 07:16 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828224: Fix leaked semaphore cleanup https://review.opendev.org/c/zuul/zuul/+/828224 | 07:18 | |
| -@gerrit:opendev.org- Dominik Jaeger proposed: [zuul/zuul] 828125: Add support for filesets in job configuration https://review.opendev.org/c/zuul/zuul/+/828125 | 08:23 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Alfredo Moralejo: [zuul/zuul-jobs] 827529: Make CentOS9 jobs voting https://review.opendev.org/c/zuul/zuul-jobs/+/827529 | 09:19 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828098: Don't add changes already in superceding pipeline https://review.opendev.org/c/zuul/zuul/+/828098 | 09:24 | |
| -@gerrit:opendev.org- Felix Edel proposed: [zuul/zuul] 828264: Ignore events without data in node cache update https://review.opendev.org/c/zuul/zuul/+/828264 | 09:50 | |
| -@gerrit:opendev.org- Felix Edel proposed: [zuul/zuul] 828268: Ignore KeyErrors in node cache listener deleted events https://review.opendev.org/c/zuul/zuul/+/828268 | 10:10 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828288: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828288 | 12:04 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828288: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828288 | 12:07 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828288: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828288 | 12:14 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828288: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828288 | 12:15 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828292: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828292 | 12:18 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828224: Fix leaked semaphore cleanup https://review.opendev.org/c/zuul/zuul/+/828224 | 13:17 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828224: Fix leaked semaphore cleanup https://review.opendev.org/c/zuul/zuul/+/828224 | 13:22 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | 14:18 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 828356: Serialize update of changes in timer driver https://review.opendev.org/c/zuul/zuul/+/828356 | 14:53 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: | 15:13 | |
| - [zuul/zuul] 828224: Fix leaked semaphore cleanup https://review.opendev.org/c/zuul/zuul/+/828224 | ||
| - [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | ||
| -@gerrit:opendev.org- Simon Westphahl proposed: | 17:02 | |
| - [zuul/zuul] 828292: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828292 | ||
| - [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | ||
| @clarkb:matrix.org | Looks like there are a number of good bug fixes I'll need to try and review this afternoon | 17:28 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 828292: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828292 | 17:34 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | 17:35 | |
| @jim:acmegating.com | swest: ^ made one tiny update to https://review.opendev.org/828292 | 17:35 |
| @jim:acmegating.com | swest: can you take a look at https://review.opendev.org/827841 and make sure that your PS3 -> PS4 transition is correct? it lost a lot of the commit message as well as other stuff... just want to make sure that was intentional | 17:40 |
| @westphahl:matrix.org | corvus: grr, no it wasn't :/ | 17:49 |
| @westphahl:matrix.org | push the wrong branch | 17:49 |
| @westphahl:matrix.org | * pushed the wrong branch | 17:49 |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | 17:50 | |
| @westphahl:matrix.org | corvus: ^ should be good now | 17:51 |
| @westphahl:matrix.org | corvus: wait, I think it's the wrong parent. too many changes at the end of the day ... | 17:53 |
| @jim:acmegating.com | i know the feeling... | 17:54 |
| -@gerrit:opendev.org- Simon Westphahl proposed: | 17:55 | |
| - [zuul/zuul] 828224: Fix leaked semaphore cleanup https://review.opendev.org/c/zuul/zuul/+/828224 | ||
| - [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | ||
| @westphahl:matrix.org | corvus: ^ that's how it should be (hopefully) | 17:56 |
| @jim:acmegating.com | swest: lgtm -- i left 2 comments on PS6 which are still relevant to PS7; i'll wait to hear your thoughts on them | 18:08 |
| @jim:acmegating.com | swest: and an ack on my change to https://review.opendev.org/828292 would be good when you have a sec | 18:09 |
| @westphahl:matrix.org | corvus: done | 18:45 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | 19:40 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 827841: Don't lock pipelines during local layout update https://review.opendev.org/c/zuul/zuul/+/827841 | 19:41 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 828224: Fix leaked semaphore cleanup https://review.opendev.org/c/zuul/zuul/+/828224 | 19:43 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 828292: Avoid data races when dequeueing superceded items https://review.opendev.org/c/zuul/zuul/+/828292 | 19:49 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul] 828175: Add a merger graceful command https://review.opendev.org/c/zuul/zuul/+/828175 | 20:06 | |
| @clarkb:matrix.org | was https://review.opendev.org/c/zuul/zuul/+/827841 rebased off of the 828224 <- 828292 stack intentionally? just making sure I'm understanding the stuff involved here as I start reviewing | 21:03 |
| @jim:acmegating.com | Clark: yes, i left a comment about that? | 21:09 |
| @jim:acmegating.com | i can elaborate if you want (but not sure if you saw it) | 21:10 |
| @clarkb:matrix.org | ah yup there it is | 21:10 |
| @jim:acmegating.com | but basically the only things stacked now are the two model changes; i think other than that everything is independent | 21:11 |
| @jim:acmegating.com | so i've rearranged things for hopefully easiest review + merging | 21:11 |
| @clarkb:matrix.org | got it | 21:12 |
| @clarkb:matrix.org | corvus: swest left a thought (or maybe a few thoughts) on https://review.opendev.org/c/zuul/zuul/+/828224 | 21:25 |
| @clarkb:matrix.org | can you take a look and let me know if there has been consideration for that or if maybe it is a non issue for some reason I've missed | 21:25 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 828429: Make a global component registry https://review.opendev.org/c/zuul/zuul/+/828429 | 21:37 | |
| @jim:acmegating.com | lookin | 21:37 |
| @clarkb:matrix.org | I think the next change in the stack has a similar concern which I noted on review there | 21:40 |
| @y2kenny:matrix.org | Is changing the keystore password possible? If so, what's the best way to do it? (I tried delete-state but I guess delete-state doesn't touch the keystore...) | 21:43 |
| @clarkb:matrix.org | Kenny Ho: I believe it will render all of your secrets invalid as they won't be decryptable. If you do this you'll be rotating all the secrets essentially. corvus should be able to confirm | 21:43 |
| @jim:acmegating.com | correct; i don't think we have a facility for that. | 21:46 |
| @y2kenny:matrix.org | ok. thanks. | 21:48 |
| @y2kenny:matrix.org | @corvus I just replied you elsewhere. I didn't see your message until now. | 21:48 |
| @jim:acmegating.com | Kenny Ho: thanks, and no worries, glad it was still there to reply to :) | 21:49 |
| @y2kenny:matrix.org | A separate question. I have been trying to add OIDC auth to zuul-web. The provider I am using is dex but seems like I am missing something (there's no error, but there's no OIDC sign in either.) Other than [auth dex] for example, is there anything else I need to configure? I feel like I need to have something that refers to the authenticator but I don't know what... | 21:56 |
| @jim:acmegating.com | Kenny Ho: you made a tenant access rule definition? https://zuul-ci.org/docs/zuul/latest/tenants.html#admin-rule-definition | 21:59 |
| @y2kenny:matrix.org | ok... I saw that... but I wasn't sure if there was something default/automatic. I will read into it | 22:00 |
| @y2kenny:matrix.org | thanks. | 22:00 |
| @jim:acmegating.com | np. that'll establish a requirement that the authenticator can fulfill | 22:00 |
| @y2kenny:matrix.org | (my head was filled with all the LDAP stuff at the time and didn't want to dive into another set of things :)) | 22:00 |
| @y2kenny:matrix.org | looking at the doc, is zuul_uid the most basic type of authorization (I am not familiar with JWT or OIDC internals.) | 22:04 |
| @jim:acmegating.com | Clark: replied. i think it's a matter of how much effort we want to exert to make it bulletproof vs "should mostly work". | 22:04 |
| @clarkb:matrix.org | corvus: makes sense. I think I'm happy as long as we call out in the docs that stepping through feature releases wtih some period of time between them is a good idea now | 22:07 |
| @clarkb:matrix.org | or doing disruptive updates with a zk cleanup step | 22:07 |
| @clarkb:matrix.org | OpenDev shouldn't have any issues since we already step through all the things. | 22:07 |
| @clarkb:matrix.org | corvus: https://review.opendev.org/c/zuul/zuul/+/826400 and https://review.opendev.org/c/zuul/zuul/+/826789 appear to have merge conflicted. | 22:07 |
| @clarkb:matrix.org | corvus: any reason to wait for https://review.opendev.org/c/zuul/zuul/+/828224 and child to pass check before approving? If not I can approve otherwise I can take a look after the school run and approve then if check is happy | 22:08 |
| @jim:acmegating.com | Clark: yeah, we'll have to take care with release notes once we remove backwards compat. | 22:10 |
| @jim:acmegating.com | Clark: it was the 'graceful' change :) | 22:12 |
| @clarkb:matrix.org | oh whoops | 22:12 |
| @jim:acmegating.com | i'll fix that up now with the assumption it won't conflict with in progress stuff, we'll see :) | 22:12 |
| -@gerrit:opendev.org- Zuul merged on behalf of Felix Edel: | 22:14 | |
| - [zuul/zuul] 828264: Ignore events without data in node cache update https://review.opendev.org/c/zuul/zuul/+/828264 | ||
| - [zuul/zuul] 828268: Ignore KeyErrors in node cache listener deleted events https://review.opendev.org/c/zuul/zuul/+/828268 | ||
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 826400: Add zuul-scheduler tenant-reconfigure https://review.opendev.org/c/zuul/zuul/+/826400 | 22:14 | |
| @jim:acmegating.com | lol | 22:14 |
| @clarkb:matrix.org | I +2'd https://review.opendev.org/c/zuul/zuul/+/828224 and its child if you want to approve now. But I need to step out for a school run and can approve after checking if we want to wait on that | 22:14 |
| @jim:acmegating.com | Clark: i don't see any conflicts between 826400 and master or any in-flight stuff so we should be good | 22:15 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 826789: Add some ZK data size stats https://review.opendev.org/c/zuul/zuul/+/826789 | 22:23 | |
| @jim:acmegating.com | and that one was the cache stat fix | 22:23 |
| @y2kenny:matrix.org | Another unrelated question: for pipeline.trigger.<connection> event comment-added, has there been any thoughts about making it more... "complex" (beyond regex.) i.e., perhaps extracting values from the comment to pass into the job? Alternatively, is the comment passed into the context of a zuul job? | 22:28 |
| @jim:acmegating.com | Kenny Ho: the comment is not passed in; that's at least partially by design, in that trigger events are not intended to alter what's run. the builds are intended to be idempotent based only on the change content. this way a build that's triggered by a comment, or an approval or a command line or a web ui or an mqtt event, etc, all behave the same way. | 23:07 |
| @y2kenny:matrix.org | corvus: Ah ok. I was thinking along the line of giving devs the ability to selectively trigger jobs by comment. | 23:09 |
| @clarkb:matrix.org | I rechecked https://review.opendev.org/c/zuul/zuul/+/827841 for more data but both py38 and py36 failed in sos's test_semaphore so I expect there may be a regression there | 23:13 |
| @clarkb:matrix.org | going to start reviewing it now. maybe it will pop out at me | 23:14 |
| @jim:acmegating.com | Clark: i understand the issue, i'm currently halfway through writing a novel about it. | 23:14 |
| @clarkb:matrix.org | ha ok. Maybe I'll wait then | 23:15 |
| @clarkb:matrix.org | or at least wait before leaving comments. I'll try to understand the change in genreal | 23:16 |
| @jim:acmegating.com | Clark: novel complete | 23:19 |
| @fungicide:matrix.org | > <@y2kenny:matrix.org> corvus: Ah ok. I was thinking along the line of giving devs the ability to selectively trigger jobs by comment. | 23:22 |
| it comes up somewhat often, and one of the biggest problems is that it's unclear what the result of that should be. would zuul replace the build result of the independently triggered job from a previous buildset? or would that singular build become a separate buildset? should it replace the verified vote from the prior buildset? also, i'm sure you can already imagine the pathological incentives that option could open up for developers who "just want their change merged" | ||
| @fungicide:matrix.org | also, job dependencies throw in a fairly giant wrench there, presumably you'd expect any dependency of the rerun job to also be rerun? | 23:23 |
| @jim:acmegating.com | i think it is possible to incorporate it into zuul's model, but it's far from trivial. | 23:26 |
| @fungicide:matrix.org | if the rerun build required an atrifact provided by another build from that buildset, you may otherwise end up using an old artifact which wasn't built from the same repository state as the new build | 23:26 |
| @fungicide:matrix.org | or which could even no longer be available | 23:27 |
| @clarkb:matrix.org | corvus: wow there is a lot going on in that change (827841). I'm still trying to make sense of it so that I can give the update a quicker review | 23:28 |
| @fungicide:matrix.org | and yes, these aren't intractable problems, but we'd need to answer those questions in some way, i think, before a design could be sketched out | 23:28 |
| @jim:acmegating.com | fungi: yeah, we answered about half of them when we sketched out a design for that for zuulv2. the other half didn't exist then. :) | 23:31 |
| @clarkb:matrix.org | corvus: can you check my questions on that change to make sure I'm following along? | 23:44 |
| @y2kenny:matrix.org | fungi: I see. Thanks for the explaination | 23:46 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!