| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 896909: Don't schedule merge repo state for branch/ref https://review.opendev.org/c/zuul/zuul/+/896909 | 05:34 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 896909: Don't schedule initial merge for branch/ref items https://review.opendev.org/c/zuul/zuul/+/896909 | 05:38 | |
| @harbott.osism.tech:regio.chat | the logging of the inheritance path is lacking the `negate` parameter, see this example from https://zuul.opendev.org/t/openstack/build/b9266d8c66444c6e8dc3d6b4f3880a0e where the next-to-last match is actually the negated version of the match shown: | 05:39 |
|---|---|---|
| ``` | ||
| zuul: | ||
| _inheritance_path: | ||
| - '<Job base branches: None source: opendev/base-jobs/zuul.d/jobs.yaml@master#25>' | ||
| - '<Job multinode branches: None source: zuul/zuul-jobs/zuul.d/general-jobs.yaml@master#33>' | ||
| - '<Job grenade-base branches: {MatchAny:{ImpliedBranchMatcher:master}} source: | ||
| openstack/grenade/.zuul.yaml@master#26>' | ||
| - '<Job grenade branches: {MatchAny:{ImpliedBranchMatcher:master}} source: openstack/grenade/.zuul.yaml@master#158>' | ||
| - '<Job grenade-skip-level-always branches: {MatchAny:{ImpliedBranchMatcher:master}} | ||
| source: openstack/grenade/.zuul.yaml@master#381>' | ||
| - '<Job grenade-skip-level branches: {MatchAny:{BranchMatcher:master}} source: | ||
| openstack/grenade/.zuul.yaml@master#396>' | ||
| - '<Job grenade-skip-level branches: {MatchAny:{BranchMatcher:^stable/(train|ussuri|victoria|wallaby|xena|zed|2023.2)$}} | ||
| source: openstack/grenade/.zuul.yaml@stable/2023.2#399>' | ||
| - '<Job grenade-skip-level branches: None source: openstack/grenade/.zuul.yaml@master#1>' | ||
| ``` | ||
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 897409: Don't trigger timer events for dynamic branches https://review.opendev.org/c/zuul/zuul/+/897409 | 10:23 | |
| @clarkb:matrix.org | > <@yoctozepto:matrix.org> observed this in OpenDev with https://review.opendev.org/c/nebulous/slo-violation-detector/+/896885/ | 12:35 |
| It would probably be helpful to link the actual builds that you observed this in rather than a high level change. Not that provides/requires should be between buildsets for multiple changes not within a single change. Within a buildset for a single change you use job dependencies. | ||
| https://review.opendev.org/c/zuul/zuul/+/895988 this spec for improving circular dependencies would as a side effect reduce the complexity in dependencies/provides/requires as well | ||
| @clarkb:matrix.org | * In reply to @yoctozepto:matrix.org | 12:35 |
| observed this in OpenDev with https://review.opendev.org/c/nebulous/slo-violation-detector/+/896885/ | ||
| It would probably be helpful to link the actual builds that you observed this in rather than a high level change. Note that provides/requires should be between buildsets for multiple changes not within a single change. Within a buildset for a single change you use job dependencies. | ||
| https://review.opendev.org/c/zuul/zuul/+/895988 this spec for improving circular dependencies would as a side effect reduce the complexity in dependencies/provides/requires as well | ||
| @yoctozepto:matrix.org | Clark: but it works in a buildset as long as the jobs come from different projects; just not on retry | 12:44 |
| @yoctozepto:matrix.org | let me link to the details in the bug report anyways | 12:45 |
| @clarkb:matrix.org | You can't have jobs from different projects in one buildset. Not today. That is what the spec will address and why it improves the situation here | 12:45 |
| @yoctozepto:matrix.org | Clark: well, I do have 🙃 | 12:45 |
| @clarkb:matrix.org | Buildsets currently run jobs for one change and a change can only belong to one project | 12:46 |
| @yoctozepto:matrix.org | it seems to have been working thus far | 12:46 |
| @clarkb:matrix.org | Provides/requires apply relationships between buildsets | 12:47 |
| @yoctozepto:matrix.org | but maybe it was a fluke that it worked | 12:47 |
| @clarkb:matrix.org | My comment is mostly to point out linking to a specific change when discussing provides requires is insufficient information | 12:47 |
| @yoctozepto:matrix.org | thanks for pointing this out, I will dig further | 12:47 |
| @yoctozepto:matrix.org | yes, it was a fluke; I see it now with the verbose info that zuul status produces | 12:51 |
| @yoctozepto:matrix.org | thanks Clark you and fungi either solve issues for me or point me into the right direction; I hope to meet you two in person some day | 12:52 |
| @yoctozepto:matrix.org | I updated and closed the bug report | 12:54 |
| @clarkb:matrix.org | The proposed future state in that spec should simplify things and make it a lot more intuitive | 12:55 |
| @clarkb:matrix.org | But still a ways away | 12:55 |
| @yoctozepto:matrix.org | I tabbed that spec and will read it in some freer time | 12:56 |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org: [zuul/zuul-website] 890040: Blog article titles link to themselves https://review.opendev.org/c/zuul/zuul-website/+/890040 | 16:10 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!