| opendevreview | melanie witt proposed openstack/releases master: Release oslo.limit 1.5.0 for Yoga https://review.opendev.org/c/openstack/releases/+/826204 | 00:18 |
|---|---|---|
| melwitt | hberaud: for your review ^ | 00:19 |
| *** amoralej|off is now known as amoralej | 08:04 | |
| opendevreview | Riccardo Pittau proposed openstack/releases master: Release ironic 18.2.1 for xena https://review.opendev.org/c/openstack/releases/+/826254 | 08:33 |
| opendevreview | Merged openstack/releases master: Release oslo.limit 1.5.0 for Yoga https://review.opendev.org/c/openstack/releases/+/826204 | 09:55 |
| opendevreview | Merged openstack/releases master: Release ironic-python-agent 8.2.1 for xena https://review.opendev.org/c/openstack/releases/+/826126 | 09:55 |
| opendevreview | Merged openstack/releases master: Release bifrost 11.2.1 for xena https://review.opendev.org/c/openstack/releases/+/826125 | 10:01 |
| elodilles | it seems we have multiple release job failures :S | 10:25 |
| elodilles | please don't approve release patches | 10:25 |
| elodilles | hberaud ttx : ^^^ | 10:25 |
| elodilles | https://zuul.opendev.org/t/openstack/build/a490703c83514eb7ba9d0ff6307f7371/log/job-output.txt#641-647 | 10:26 |
| elodilles | for me it seems related to some gerrit config change maybe? | 10:27 |
| hberaud | ack | 10:32 |
| elodilles | i'm a bit puzzled about the second one, as it has some similar sympthoms but it is weirdly failed at some different point: https://zuul.opendev.org/t/openstack/build/00d3f765ea6847a3bf7ea53294cf2a88/log/job-output.txt#979-980 | 10:33 |
| elodilles | and same thing on the 3rd fail: https://zuul.opendev.org/t/openstack/build/d8aaeff52dac442e89fd3e253596e0c2/log/job-output.txt#980 | 10:34 |
| hberaud | elodilles: I think something went wrong with ssh and keys or something like that | 10:37 |
| elodilles | yes, it seems :S | 10:37 |
| elodilles | i'll send a mail to ML as well | 10:45 |
| elodilles | (and the mail: http://lists.openstack.org/pipermail/openstack-discuss/2022-January/026898.html ) | 10:55 |
| hberaud | apparently gerrit has been updated, the UI seems a bit different today | 13:10 |
| elodilles | yepp | 13:27 |
| *** amoralej is now known as amoralej|lunch | 13:31 | |
| *** amoralej|lunch is now known as amoralej | 14:02 | |
| fungi | just a heads up, those host key errors are because when we upgraded from gerrit 3.3 to 3.4 yesterday it added at least one new host key type and we didn't know to add it to the known hosts entries in the release_ssh_key secret, so i'll work on updating that now | 14:08 |
| fungi | please hold approving any further release requests until i get that patched and let you know | 14:08 |
| hberaud | ack | 14:17 |
| fungi | oh, though that looks like it may have merely been a warning, the error looks permissions-related | 14:18 |
| fungi | just an update, the host key warning was a red herring, though i've pushed a fix to make it stop. i've tested manually and confirmed the real problem seems to be a regression in gerrit 3.4, for which we're testing a possible workaround | 15:12 |
| clarkb | now a couple possible workarounds | 15:16 |
| fungi | yep | 15:19 |
| elodilles | sounds promising! thanks! | 15:22 |
| clarkb | elodilles: fungi: a good improvement to the release tools would be to capture the output from gerrit for debugging in these cases | 15:31 |
| clarkb | it seems to have consumed the message indicating we need extra permissions | 15:31 |
| fungi | well, the log did error about insufficient permissions to push a tag | 15:34 |
| clarkb | oh it did? I guess the linked lines about the ssh key were very distracting :) | 15:34 |
| fungi | yep, the error is farther down | 15:39 |
| elodilles | sorry, it was me, i thought the cause is the SSH host key thing o:) | 15:43 |
| *** marios is now known as marios|out | 16:50 | |
| opendevreview | Elod Illes proposed openstack/releases master: [vitrage] Propose to switch the release model https://review.opendev.org/c/openstack/releases/+/826328 | 17:03 |
| opendevreview | Elod Illes proposed openstack/releases master: [swift] Propose to switch the release model https://review.opendev.org/c/openstack/releases/+/826329 | 17:04 |
| opendevreview | Elod Illes proposed openstack/releases master: [ovn-octavia-provider] Propose to switch the release model https://review.opendev.org/c/openstack/releases/+/826331 | 17:09 |
| opendevreview | Elod Illes proposed openstack/releases master: [ironic] Propose to switch the release model https://review.opendev.org/c/openstack/releases/+/826332 | 17:14 |
| opendevreview | Elod Illes proposed openstack/releases master: [Quality Assurance] Propose to switch the release model https://review.opendev.org/c/openstack/releases/+/826336 | 17:17 |
| opendevreview | Eyal proposed openstack/releases master: release new version of vitrage and vitrage-dashboard https://review.opendev.org/c/openstack/releases/+/826344 | 17:37 |
| *** amoralej is now known as amoralej|off | 17:44 | |
| opendevreview | sean mooney proposed openstack/releases master: release os-traits 2.7.0 https://review.opendev.org/c/openstack/releases/+/826353 | 18:37 |
| sean-k-mooney | not that im complaining but why did zuul set PTL-Approved on https://review.opendev.org/c/openstack/releases/+/826353 i know im set as the release liasion for nova https://github.com/openstack/releases/blob/a2c89d98d1ab2ffdd4b4da5c99e1d5cf259a626f/data/release_liaisons.yaml#L85-L88 and os-trait is a deliverable of nova in the governace repo | 18:53 |
| sean-k-mooney | https://github.com/openstack/governance/blob/master/reference/projects.yaml#L1228-L1230 but does zull take account of that? | 18:53 |
| sean-k-mooney | the state is not nessisarly wrong i just want to understand how it currently works | 18:56 |
| sean-k-mooney | https://zuul.openstack.org/build/814484bd5c724ba0a083bcf67746ced8/log/job-output.txt#1048 looks like it is picking that up | 18:58 |
| sean-k-mooney | ah its likely part of check-release-approval https://github.com/openstack/releases/blob/a2c89d98d1ab2ffdd4b4da5c99e1d5cf259a626f/.zuul.yaml#L40-L42 | 19:00 |
| sean-k-mooney | yes https://zuul.openstack.org/build/4791dcec6b47425193f603b0e69894ab/log/job-output.txt#64 | 19:02 |
| elodilles | sean-k-mooney: yes, it's because you are release liaison :) | 19:02 |
| elodilles | for nova | 19:02 |
| elodilles | :) | 19:02 |
| sean-k-mooney | yep | 19:02 |
| sean-k-mooney | ok that all makes sense then looking at the job | 19:02 |
| elodilles | and the PTL-Approval flag gets triggered even in case you are proposing the patch :) | 19:03 |
| sean-k-mooney | i was confused becaue i cant actully set that feild in gerrit | 19:03 |
| elodilles | oh, can you? :-o | 19:03 |
| sean-k-mooney | no i dont have permission to set it PTL-approved in gerrit but zuul does | 19:04 |
| sean-k-mooney | which is fine | 19:04 |
| elodilles | ahh, sorry, i read 'i can' :) | 19:04 |
| elodilles | yes, it cannot be set actually | 19:05 |
| elodilles | it only gets triggered whenever a PTL or release liaison +1's a patch | 19:05 |
| sean-k-mooney | ah ok | 19:05 |
| elodilles | via the check-release-approval job that you have found out already :) | 19:06 |
| sean-k-mooney | i guess that avoid the need to manage gerrit groups when that chagnes | 19:06 |
| sean-k-mooney | you can simply manage it https://github.com/openstack/releases/blob/master/data/release_liaisons.yaml | 19:07 |
| elodilles | this makes the release managers job easier, because we see that someone who is responsible for the release was involved (uploaded the patch, or approved it) | 19:07 |
| sean-k-mooney | yep | 19:07 |
| sean-k-mooney | ok well ill proably call it a day there | 19:07 |
| sean-k-mooney | o/ | 19:07 |
| elodilles | me, too, have a nice evening o/ | 19:07 |
| opendevreview | Vishal Manchanda proposed openstack/releases master: Release horizon 21.0.0(Yoga) https://review.opendev.org/c/openstack/releases/+/826362 | 19:43 |
| opendevreview | Vishal Manchanda proposed openstack/releases master: Release horizon for stable/victoria https://review.opendev.org/c/openstack/releases/+/825773 | 20:05 |
| fungi | elodilles et al: per my latest reply on the ml, it's safe to resume approving review requests and i can reenqueue the failed ones if needed, just let me know | 21:08 |
| fungi | (i think just approving new releases changes will rerun the tag-releases job and push any missing tags though, right?) | 21:08 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!