| @clarkb:matrix.org | corvus: I think https://paste.opendev.org/show/bS1Udj5RpRm8dkcFaVac/ shows a few different possible races in the upload process for common blobs. I think larger blobs just make the racing more likely. I didn't note this in the paste but I also think it is possible that the two uploads are overwriting each others chunk files? | 00:00 |
|---|---|---|
| @jim:acmegating.com | Clark: other people might be using that as an registry with a filesystem backend, so if it made it worse, sounds like a revert is good. | 00:10 |
| @clarkb:matrix.org | ya I'm still nto sure if it made it worse or not. I think it didn't make it better so ya no objection from me if you +A | 00:11 |
| @jim:acmegating.com | * Clark: other people might be using that as an intermediate registry with a filesystem backend, so if it made it worse, sounds like a revert is good. | 00:11 |
| @clarkb:matrix.org | I've got 831064 running to see if it is the two pushes colliding with each other that causes the problem or not | 00:11 |
| @jim:acmegating.com | +3 -- can always unrevert | 00:11 |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul-registry] 831054: Revert "Atomically concatenate blob objects" https://review.opendev.org/c/zuul/zuul-registry/+/831054 | 00:31 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 831069: Fix error in sql reporter upgrade https://review.opendev.org/c/zuul/zuul/+/831069 | 00:45 | |
| @jim:acmegating.com | zuul-maint: ^ we should merge that as an emergency fix | 00:45 |
| @jim:acmegating.com | i'm not sure anyone is around and that's affecting opendev; so i'm going to approve that now with the understanding that anyone is free to -1 or -2 it at any time before it merges, or even retro-review it afterwords. | 00:49 |
| @tristanc_:matrix.org | corvus: looks to me, I wonder if mypy would have catched this error. | 00:57 |
| @tristanc_:matrix.org | * corvus: looks good to me, I wonder if mypy would have catched this error. | 00:57 |
| @jim:acmegating.com | that's a sqlalchemy magic attribute, so i doubt it, but would be truly impressed if it did :) | 01:00 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 831069: Fix error in sql reporter upgrade https://review.opendev.org/c/zuul/zuul/+/831069 | 05:33 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 831082: Refresh pipeline change list and status when refreshing the state https://review.opendev.org/c/zuul/zuul/+/831082 | 18:11 | |
| @jim:acmegating.com | zuul-maint: ^ that is a fix minor/moderate UX bug which you can see on the zuul tenant status page right now. it's not affecting jobs, but it's certainly confusing. | 18:12 |
| @jim:acmegating.com | (see the periodic pipeline) | 18:12 |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 831082: Refresh pipeline change list and status when refreshing the state https://review.opendev.org/c/zuul/zuul/+/831082 | 19:04 | |
| -@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 831085: Calculate elapsed and remaining times in javascript https://review.opendev.org/c/zuul/zuul/+/831085 | 20:54 | |
| @jim:acmegating.com | zuul-maint: ^ and that another UI issue i noticed after yesterday's restart. those are the only issues i've seen; hopefully that's the last of it. :) | 20:55 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 831082: Refresh pipeline change list and status when refreshing the state https://review.opendev.org/c/zuul/zuul/+/831082 | 21:59 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!