| @g_gobi:matrix.org | > <@g_gobi:matrix.org> Is there any sample pipeline + job configuration for using timer option with gitlab? If yes, could you please share it here? | 02:01 |
|---|---|---|
| Any idea about this?🙁 | ||
| -@gerrit:opendev.org- daniel.pawlik https://matrix.to/#/@dpawlik:matrix.org proposed: [zuul/zuul] 889314: Change start_time and end_time field type to integer https://review.opendev.org/c/zuul/zuul/+/889314 | 09:19 | |
| -@gerrit:opendev.org- daniel.pawlik https://matrix.to/#/@dpawlik:matrix.org proposed: [zuul/zuul] 889314: Change start_time and end_time field type to integer; add new fields https://review.opendev.org/c/zuul/zuul/+/889314 | 09:35 | |
| -@gerrit:opendev.org- daniel.pawlik https://matrix.to/#/@dpawlik:matrix.org proposed: [zuul/zuul] 889314: Change start_time and end_time field type to integer; add new fields https://review.opendev.org/c/zuul/zuul/+/889314 | 09:35 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 889305: Add Zuul job variable to indicate a job will retry https://review.opendev.org/c/zuul/zuul/+/889305 | 11:13 | |
| @fungicide:matrix.org | > <@g_gobi:matrix.org> Any idea about this?🙁 | 12:53 |
| i'm not sure who's using the gitlab driver. i could give you examples of using a timer trigger with the gerrit driver, but i don't expect that would help you | ||
| @fungicide:matrix.org | hopefully another gitlab user will chime in with details | 12:53 |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 889858: Don't fail tenant validation for deprecations https://review.opendev.org/c/zuul/zuul/+/889858 | 13:05 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 889745: Refactor will_retry https://review.opendev.org/c/zuul/zuul/+/889745 | 14:08 | |
| @jpew:matrix.org | Ugh, Zuul web UI is pretty much unusable with how many branches we have :/ | 14:43 |
| @fungicide:matrix.org | jpew: do you have any design recommendations for addressing your case? | 15:08 |
| @fungicide:matrix.org | where does it break down? | 15:08 |
| @jpew:matrix.org | It's slow on the browser side, maybe the /status endpoint can have an option to only return say... 500 items for a given pipeline? | 15:20 |
| @jpew:matrix.org | It's the fetching and processing the gigantic /status JSON file that is slow on the browser side | 15:30 |
| @jpew:matrix.org | And it's huge because there are 1000's of unused branches in our periodic pipeline queue; those branches are in exclude-branches, but the scheduler still has to look at them (?) so they still show up in the queue | 15:32 |
| @mhuin:matrix.org | Hey quick question for folks who have deployed zuul: what does your database architecture look like? I'm especially interested in big deployments like opendev | 15:35 |
| @mhuin:matrix.org | Do you have a single instance or have you set up a cluster with replication? | 15:35 |
| @jim:acmegating.com | opendev uses a single mysql server; the operator deploys a cluster. the main question is your own fault tolerance requirements, not scaling/throughput. a single mysql/mariadb is sufficient for large scale. | 15:43 |
| @fungicide:matrix.org | > <@jpew:matrix.org> And it's huge because there are 1000's of unused branches in our periodic pipeline queue; those branches are in exclude-branches, but the scheduler still has to look at them (?) so they still show up in the queue | 18:01 |
| checking the docs and then tracing that option through the source code (because i don't have any projects using that option), it looks like it only limits what branches the scheduler will load configuration from, not what branches the source connection drivers are allowed to trigger on | ||
| @jpew:matrix.org | Ya, that appears to empirically be the case :) | 18:02 |
| @fungicide:matrix.org | so unless i'm misunderstanding the implementation, it looks like you could still have jobs triggered for branches that are excluded | 18:02 |
| @fungicide:matrix.org | if the desire is for source connection drivers to also filter their events to exclude any for those branches, then such a change could potentially also mean it's safe to filter down the status data to not cover them | 18:03 |
| @fungicide:matrix.org | i think that would be a backward-incompatible change in the drivers though, again if i'm looking at the code the right way up, so we'd probably have to make it an additional toggle or have a deprecation period or something | 18:04 |
| @fungicide:matrix.org | also, fair warning, i've only spent a few minutes thinking about it. entirely possible there are very good reasons that's not desirable, or not possible | 18:04 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 889858: Don't fail tenant validation for deprecations https://review.opendev.org/c/zuul/zuul/+/889858 | 19:07 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 886785: Add REPL https://review.opendev.org/c/zuul/nodepool/+/886785 | 19:16 | |
| @vlotorev:matrix.org | zuul maint, could you have a look on https://review.opendev.org/#/c/878677 ? | 19:39 |
| It's oneliner fix for ReST URL in the docs. The change already has one CR+2. | ||
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 888959: Add commit_id to zuul vars https://review.opendev.org/c/zuul/zuul/+/888959 | 19:41 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 889368: Add a release note about Python 3.11 https://review.opendev.org/c/zuul/zuul/+/889368 | 20:23 | |
| @fungicide:matrix.org | wow, the last zuul-nox-py311 build on https://review.opendev.org/878677 exploded. trying to load it in the web dashboard chewed up and spat out the mangled remains of my browser. fetching and decompressing the job-output.txt, it's over 8 million lines long... seems like mostly zk connection error tracebacks | 23:54 |
| @fungicide:matrix.org | `Unexpected issue in _run loop` | 23:56 |
| @fungicide:matrix.org | lots of `kazoo.exceptions.NoNodeError` i guess | 23:56 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!