| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/nodepool] 887356: dnm: Increase OpenStack driver cache TTL https://review.opendev.org/c/zuul/nodepool/+/887356 | 05:15 | |
| @jjbeckman:matrix.org | Hi folks. Would appreciate some advice. | 05:37 |
|---|---|---|
| I've just noticed that pipelines that take over 30 minutes to finish executing, get aborted, and end up as TIMED_OUT. | ||
| I've also noticed that in the `ansible.cfg` files generated in the executor appear to set `timeout` to 30 minutes, as the following example. | ||
| ``` | ||
| ansible/pre_playbook_1/ansible.cfg:14:timeout = 30 | ||
| ``` | ||
| Is there any way to change the value of the `ansible.cfg` files generated by the executor? | ||
| Thank you. | ||
| @g_gobi:matrix.org | Hi Team, | 06:15 |
| https://zuul-ci.org/docs/zuul/latest/config/project.html#attr-project.%3Cpipeline%3E.fail-fast | ||
| this fail-fast option aborts the current running builds that was expected but due to this, the cleanup tasks mentioned in post-run were not been executed. | ||
| Is there anyway to run post run jobs during fail-fast? | ||
| @clarkb:matrix.org | I think you need to use the cleanup phase not post-run | 06:15 |
| @g_gobi:matrix.org | Is there any cleanup-phase option available?😅 | 06:16 |
| @clarkb:matrix.org | Yes | 06:16 |
| @clarkb:matrix.org | https://zuul-ci.org/docs/zuul/latest/config/job.html#attr-job.cleanup-run | 06:17 |
| -@gerrit:opendev.org- Moritz Haase proposed: [zuul/zuul-jobs] 871822: roles/ensure-python: Fix 'python_use_stow' option https://review.opendev.org/c/zuul/zuul-jobs/+/871822 | 06:17 | |
| @g_gobi:matrix.org | Oh thank you. I'll take a look. | 06:18 |
| Can we able to show the logs for the aborted jobs? | ||
| @clarkb:matrix.org | Yes if you collect them in the cleanup I think | 06:37 |
| @flaper87:matrix.org | tristanC: 👋 I checked the preview and it seems to be ok. May I ask you to help me double check if you have a couple of mins? https://review.opendev.org/c/zuul/zuul/+/885455 | 08:13 |
| Then it'd only need the final +2 from corvus 💪 | ||
| @newbie23:matrix.org | > how does that work with merging, if you're proposing a pull request with the another pull request's topic branch as its target branch? you're not actually going to merge one pr into the other are you? | 08:22 |
| I was thinking to "fold" the PRs from the last one to the first one. Say we have PRs A, B and C. All of them are approved and ready for merge. Then C is merged into B branch. The resulting B branch is merged into A. Finally A is merged into "main". The reason for this would be to avoid conflicts arising from the squash&merge policy we have. | ||
| About GitHub, considering the already mentioned chained PRs A, B and C (where B add commits on A, and C add commits to B): | ||
| main <- A, easy for code review, you only have A commits | ||
| main <- B, bad for code review, you have B commits that includes A commits, when A is squashed&merged you get conflicts as the original A commits are present in B | ||
| main <- C, same | ||
| The alternative I was thinking | ||
| main <- A, easy for code review, you only have A commits | ||
| A <- B, easy for code review, you only have B commits | ||
| B <- C, easy for code review, you only have C commits | ||
| Once all PRs are code reviewed and checks pass (Zuul enabled on all unprotected branches), then gate&merge all of them in the reverse order, like this: | ||
| B <-merge-C, | ||
| A <-merge-B, | ||
| main <-merge-A | ||
| This should avoid conflicts even using squash&merge. | ||
| Does it make sense? | ||
| Also, thanks for the link to the circular deps, I will look into it. | ||
| @jim:acmegating.com | tdlaw: Clark i don't recommend collecting logs in the cleanup phase; they should only ever be done in the final post-run playbook. there is special handling in zuul for that. | 13:34 |
| @fungicide:matrix.org | i'm also curious why logs from aborted builds are useful, since allowing them to be aborted implies the builds themselves weren't useful (otherwise you'd set your pipelines up to not do that in the first place) | 13:56 |
| @fungicide:matrix.org | i don't use github at all, and have only a passing familiarity with it, so i can't really judge what workflows make sense there. zuul has to make some assumptions about workflows though, and it doesn't sound like the workflow you're describing is compatible with its assumptions | 13:58 |
| @fungicide:matrix.org | zuul expects to be in charge of deciding when a pr merges, and needs to have a durable branch that it's merging into. i don't see how ephemeral topic branches for other pull requests would satisfy that | 13:59 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Marvin Becker: [zuul/nodepool] 873716: Add gpu support for k8s/openshift pods https://review.opendev.org/c/zuul/nodepool/+/873716 | 14:10 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 881144: Deduplicate circular dep jobs in check https://review.opendev.org/c/zuul/zuul/+/881144 | 15:45 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Marvin Becker: [zuul/nodepool] 873716: Add gpu support for k8s/openshift pods https://review.opendev.org/c/zuul/nodepool/+/873716 | 15:45 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 887907: Use low-level OpenStack SDK calls for server listing https://review.opendev.org/c/zuul/nodepool/+/887907 | 16:39 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/nodepool] 887356: dnm: Increase OpenStack driver cache TTL https://review.opendev.org/c/zuul/nodepool/+/887356 | 16:39 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/nodepool] 887356: dnm: Increase OpenStack driver cache TTL https://review.opendev.org/c/zuul/nodepool/+/887356 | 16:42 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 888144: Increase state machine interval and cache ttl https://review.opendev.org/c/zuul/nodepool/+/888144 | 16:57 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 887907: Use low-level OpenStack SDK calls for server listing https://review.opendev.org/c/zuul/nodepool/+/887907 | 20:17 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/nodepool] 887356: dnm: Increase OpenStack driver cache TTL https://review.opendev.org/c/zuul/nodepool/+/887356 | 20:17 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 888144: Increase state machine interval and cache ttl https://review.opendev.org/c/zuul/nodepool/+/888144 | 20:18 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 888187: Add Ansible 8 https://review.opendev.org/c/zuul/zuul/+/888187 | 20:42 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 887732: Add explicit foreign key indexes https://review.opendev.org/c/zuul/zuul/+/887732 | 21:29 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!