openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Add log entry when unable to find a project during reenqueing https://review.openstack.org/481778 | 00:02 |
---|---|---|
jeblair | currently zuulv3.o.o has a cached change object for 479084 with change.project=None, which should not happen. i can't figure out why that is. the only opportunity i can think of for that to be set to None involves reconfiguration, though i have my doubts as to whether that's what happened. | 00:04 |
jeblair | i'd like to merge 481778 and restart zuulv3.o.o to try to narrow things down further. | 00:05 |
jeblair | mordred, clarkb, fungi, SpamapS: unless one of you is still around, i'm going to self-approve that change ^ | 00:06 |
fungi | jeblair: i'm back, looking now | 00:14 |
openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Add log entry when unable to find a project during reenqueing https://review.openstack.org/481778 | 00:18 |
*** xinliang has quit IRC | 00:48 | |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: Add some indexes to allow efficient dashboard sql https://review.openstack.org/481614 | 00:52 |
*** xinliang has joined #zuul | 01:00 | |
*** xinliang has quit IRC | 01:00 | |
*** xinliang has joined #zuul | 01:00 | |
openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: Add support for zuul.d configuration split https://review.openstack.org/473764 | 02:19 |
openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: Add support for zuul.d configuration split https://review.openstack.org/473764 | 02:40 |
*** xinliang has quit IRC | 10:56 | |
*** xinliang has joined #zuul | 11:10 | |
*** xinliang has quit IRC | 11:10 | |
*** xinliang has joined #zuul | 11:10 | |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Make sure bindep is on the node https://review.openstack.org/481817 | 13:46 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Rework bindep role to be more ansible and less shell https://review.openstack.org/481818 | 13:46 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Update debug-ansible role to put files in a dir https://review.openstack.org/481819 | 13:46 |
*** isaacb has joined #zuul | 14:11 | |
openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Permit config shadowing https://review.openstack.org/479084 | 14:58 |
openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Fix race in test_queue_rate_limiting_dependent https://review.openstack.org/481824 | 14:58 |
jeblair | mordred: retry limit log link fix is next on my list. | 15:00 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: Add support for marking module results uninteresting https://review.openstack.org/481825 | 15:14 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Clean up debug-ansible role https://review.openstack.org/481819 | 15:14 |
openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Rename debug-ansible to validate-host https://review.openstack.org/481826 | 15:14 |
mordred | jeblair: yay! | 15:15 |
mordred | jeblair: I think once we get https://review.openstack.org/481819 in - our logs will be a bit nicer - but there's a bit to debug first obvs | 15:15 |
jeblair | mordred: oh, interesting approach. that's different than ps1. | 15:17 |
mordred | jeblair: yah - I'm playing around with some different things :) | 15:17 |
jeblair | mordred: there's a potential downside to using a module -- it means it has to go into a trusted repo and can't be speculatively tested.... if folks want to add new debugging stuff that may be slightly less convenient. | 15:17 |
mordred | jeblair: (I'll still come back to making that go into the base job - but I think I can clean it up a bunch more in zuul-jobs before we do) | 15:17 |
mordred | jeblair: that's not true - modules should fine - it's plugins that aren't | 15:18 |
mordred | modules execute remotely - it's the plugins we block because they have the potential to execute locally | 15:18 |
jeblair | mordred: oh right | 15:18 |
jeblair | mordred: of course if it's going in the base job anyway... | 15:18 |
mordred | yah. there is that part | 15:18 |
mordred | jeblair: otoh- it's also a role/module that's super easy to test pointed at a node | 15:19 |
jeblair | mordred: yep. and i think for changes like that, we have the 'base-test' pattern we can use. should make it fairly easy and safe to change the base job. | 15:20 |
mordred | jeblair: so we can write a test job for the trusted repo that runs the role against a target so that we can at least test the role | 15:20 |
mordred | (or for wherever the role lives) | 15:21 |
mordred | it doesn't need much in the way of zuul structure around it - it's just a normal ansible role - and then yes, there's also base-test too | 15:21 |
mordred | so I think we should be able to provide some mitigations to it being in the trusted repo | 15:21 |
jeblair | ++ | 15:22 |
mordred | jeblair: I'm also toying wiht a thought around providing some ability to differentiate in the logs between pre, run and post output | 15:27 |
jeblair | mordred: seems useful | 15:27 |
mordred | jeblair: I'm ALMOST all the way to an actual suggestion :) | 15:27 |
openstackgerrit | James E. Blair proposed openstack-infra/zuul feature/zuulv3: Return RETRY_LIMIT on completion of last build https://review.openstack.org/481835 | 15:47 |
*** openstackgerrit has quit IRC | 16:17 | |
*** isaacb has quit IRC | 21:20 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!