| bluca | andersson1234: I noticed a small issue with the results reporting - if a job fails and the failed result is attached to the PR, and then I trigger it manually (because it's flaky or I change something in the config), then the job goes back to "pending" in the PR | 13:02 |
|---|---|---|
| bluca | but after a few minutes, the previous failed result is attached again | 13:02 |
| bluca | the new job keeps running and eventually completes, but it's result is not reported back, the previous one stays in the PR result | 13:02 |
| bluca | this only happens when re-triggering manually via the http api - force pushing to the PR branch doesn't show any issue | 13:03 |
| andersson1234 | bluca: Thanks for reporting this. I'll note it down. Do you consider this a blocker? Or can you work around it with force pushes? | 15:39 |
| bluca | not a blocker and definitely not urgent | 15:43 |
| bluca | another question: is it possible to cancel a pending/running job from the http api? | 15:50 |
| Skia | sadly not yet, but we're working on something there. No ETA though | 15:58 |
| bluca | no worries, not urgently needed, thanks | 16:08 |
| bluca | ppc64el jobs seem to be failing to boot the testbed | 19:52 |
| bluca | happened multiple times | 19:52 |
| bluca | <VirtSubproc>: failure: setup script failed with code 1: /home/ubuntu/autopkgtest/ssh-setup/nova open | 19:52 |
| bluca | https://autopkgtest.ubuntu.com/results/autopkgtest-noble-upstream-systemd-ci-systemd-ci/noble/ppc64el/s/systemd-upstream/20240304_172916_d4069@/log.gz | 19:52 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!