| *** ysandeep|out is now known as ysandeep | 03:19 | |
| *** ysandeep is now known as ysandeep|brb | 04:55 | |
| *** ysandeep|brb is now known as ysandeep | 05:22 | |
| *** ykarel is now known as ykarel|mtg | 06:02 | |
| *** bhagyashris_ is now known as bhagyashris|rover | 07:08 | |
| *** sshnaidm|off is now known as sshnaidm | 07:26 | |
| *** jpena|off is now known as jpena | 07:31 | |
| *** ykarel|mtg is now known as ykarel | 07:34 | |
| *** ysandeep is now known as ysandeep|lunch | 08:12 | |
| *** aluria_ is now known as aluria | 09:03 | |
| *** ykarel is now known as ykarel|lunch | 09:09 | |
| *** ysandeep|lunch is now known as ysandeep | 09:17 | |
| *** ykarel|lunch is now known as ykarel | 10:10 | |
| *** rlandy is now known as rlandy|ruck | 10:32 | |
| *** jcapitao is now known as jcapitao_lunch | 10:47 | |
| *** fressi_ is now known as fressi | 10:54 | |
| *** jpena is now known as jpena|lunch | 11:29 | |
| *** ysandeep is now known as ysandeep|brb | 11:39 | |
| *** ysandeep|brb is now known as ysandeep | 12:13 | |
| *** jcapitao_lunch is now known as jcapitao | 12:20 | |
| *** jpena|lunch is now known as jpena | 12:22 | |
| *** ysandeep is now known as ysandeep|brb | 12:29 | |
| *** ysandeep|brb is now known as ysandeep | 12:58 | |
| sean-k-mooney | is this the full set of nodepool lables avaiable https://zuul.openstack.org/labels or are there others ? | 13:07 |
|---|---|---|
| sean-k-mooney | hum it looks to be the same on all tenats so ya i assume so | 13:09 |
| fungi | sean-k-mooney: that's the full set configured in our deployment, yes | 13:09 |
| sean-k-mooney | i was wonderign if we had ubuntu-latest or similar | 13:11 |
| sean-k-mooney | basically i was looking for a simple way to test with a newer version of libvirt | 13:12 |
| sean-k-mooney | debian and fedora 34 willl work | 13:12 |
| fungi | we have debian-bullseye | 13:12 |
| fungi | yeah | 13:12 |
| sean-k-mooney | yep that has 7.0.0 | 13:12 |
| fungi | also opensuse-tumbleweed is a rolling release | 13:12 |
| sean-k-mooney | we are hitting an issue we tghink is fixed after 6.2.0 | 13:12 |
| ykarel | recently fedora-latest was switched to fedora-34 if that works | 13:21 |
| ykarel | fedora-latest nodeset | 13:22 |
| sean-k-mooney | it likely will but tempest jobs are less stable on fedora so trying to not boild the ochen here and keep things as similar as possible | 13:22 |
| sean-k-mooney | so that is why i was wondering about 21.04 or the upcuming 21.10 as they will be most simiarly to the current 20.04 job. we are trying centos 8 stream now but i suspect that will use the os defautl libivrt not the advacne vert version we could certenly change that in devstack but right now we are just smoke testing a theory | 13:23 |
| sean-k-mooney | we have a potinealy workaorund for older release which si to jsut tweek the libvirt keepalive interval to avoid the current ci issues | 13:24 |
| sean-k-mooney | so trying with a newer version is mainly to confirm that issue is fixed | 13:25 |
| *** akekane__ is now known as abhishekk | 14:42 | |
| *** ykarel is now known as ykarel|away | 14:42 | |
| elodilles | fyi, I'm planning to run the eol'd branch removal script to clean up a couple of old branches. (as far as I see there is nothing gerrit related activity now) | 15:57 |
| fungi | elodilles: sounds good, thanks for the heads up, you should be all clear to do whatever you need | 16:05 |
| elodilles | cool, thanks! | 16:06 |
| *** ysandeep is now known as ysandeep|dinner | 16:06 | |
| frickler | sean-k-mooney: devstack on bullseye should be working: https://review.opendev.org/c/openstack/devstack/+/789083 | 16:09 |
| frickler | I'm just trying to find out why qemu seems to be much more memory hungry there | 16:09 |
| frickler | like cirros in a 256M flavor uses < 256M rss on focal, but 512M on bullseye | 16:10 |
| *** akekane_ is now known as abhishekk | 16:28 | |
| *** jpena is now known as jpena|off | 16:34 | |
| sean-k-mooney | frickler: likely a change in the default machine type or something along those lines | 16:36 |
| sean-k-mooney | thats just a guess | 16:36 |
| frickler | sean-k-mooney: that's a good idea, that would be visible in the libvirt xml I assume? | 16:46 |
| sean-k-mooney | am yes it should be | 16:53 |
| sean-k-mooney | if its defaulting to say q35 that dose use more ram | 16:53 |
| sean-k-mooney | i dont know if debian has its own device models or if it uses the upstream ones | 16:54 |
| *** ysandeep|dinner is now known as ysandeep|out | 17:25 | |
| elodilles | and the today's eol'd branch clean-up result: https://paste.opendev.org/show/809773/ | 17:51 |
| fungi | thanks for the details! | 17:52 |
| bcafarel | elodilles: thanks for the housekeeping :) and stable/train CI is fixed, so that should make a nice drop in periodic job failures (at least the vpnaas ones) | 20:03 |
| fungi | yay! can't wait to see what the daily message count for the stable-maint ml falls to now | 20:06 |
| bcafarel | oslo folks were also talking about EOL'ing some branches too - it should make a nice drop too in that ml activity | 20:10 |
| fungi | yep | 20:23 |
| *** rlandy|ruck is now known as rlandy|ruck|bbl | 22:28 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!