Thursday, 2026-05-14

-@gerrit:opendev.org- Michal Nasiadka marked as active: [openstack/diskimage-builder] 988472: CI: Switch dib-devstack to use openstack.cloud https://review.opendev.org/c/openstack/diskimage-builder/+/98847211:33
@fungicide:matrix.orgthe mirror.ubuntu vos release (including resolute and the last month of updates for noble) started at 02:51:28 today, barring any errors it will likely continue through the weekend13:30
@fungicide:matrix.orghuh, or not! i guess because the mirror volumes only replicate to another fileserver in the same cloud region, it copied very quickly14:11
@fungicide:matrix.orgi'm running it again just to make sure it's a no-op14:11
@clarkb:matrix.orgfungi: disk utilization looks reasonable which is nice14:44
@clarkb:matrix.orgfungi: when the second pass is done should we send an email to service-announce indicating the resolute labels s hould be ready for general consumption? Maybe with a note about the python 3.14 GC issue too?14:45
@fungicide:matrix.orgyeah, there were a handful of new packages pulled in since yesterday, but reprepro is at the point of regenerating the indices now, and grafana no longer shows the volume a month stale since the previous run15:08
@fungicide:matrix.orgit does look like my estimate was about double what resolute actually ended up consuming, i guess because i was basing it on older releases that now include years of fixes and security updates15:09
@fungicide:matrix.orgwe could in theory add resolute arm64 mirroring at the moment without putting too much pressure on /vicepa if it's approximately the same size as amd64 (earlier comparisons indicate it may be a few percent larger, but still comparable)15:11
@fungicide:matrix.orgbut as these grow, we'll probably be under a bit of a deadline to free up space or add more storage15:12
@clarkb:matrix.orgMakes sense15:32
@fungicide:matrix.orgokay, not sure why that took as long as it did, maybe the first reprepro pass after adding new files needs more time for the deleteunreferenced checking and such, but it started to vos release finally15:54
@fungicide:matrix.orgvos release finished, i guess it's open season on resolute amd64 mirrors now16:04
@clarkb:matrix.orgfungi: do you want to send the announcement to service-annouce or should I?16:12
@fungicide:matrix.orgi can16:17
-@gerrit:opendev.org- Zuul merged on behalf of cid: [openstack/project-config] 984159: Ends project gating for `virtualpdu` https://review.opendev.org/c/openstack/project-config/+/98415917:06
-@gerrit:opendev.org- cid proposed: [openstack/project-config] 988655: Remove VirtualPDU repository references https://review.opendev.org/c/openstack/project-config/+/98865517:21
-@gerrit:opendev.org- cid proposed: [openstack/project-config] 988655: Remove VirtualPDU repository references https://review.opendev.org/c/openstack/project-config/+/98865517:42
-@gerrit:opendev.org- cid proposed: [openstack/project-config] 988655: [EOL]: Remove VirtualPDU repository references https://review.opendev.org/c/openstack/project-config/+/98865517:44
-@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 988678: Use named matches in jitsi config https://review.opendev.org/c/opendev/system-config/+/98867818:42
-@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [opendev/system-config] 988678: Use named matches in jitsi config https://review.opendev.org/c/opendev/system-config/+/98867820:34
@fungicide:matrix.orgthat ^ deployed, i'll make sure stuff's still reachable20:42
@clarkb:matrix.orgthanks20:43
@clarkb:matrix.orgI'm almost done testing the 3.12 -> 3.13 upgrade. Then will run through the downgrade. I don't see anything unexpected yet20:43
@fungicide:matrix.orgi'm in https://meetpad.opendev.org/isitbroken and stuff still seems to be functional as best i can tell20:44
@clarkb:matrix.orgI can join momentarily20:44
@clarkb:matrix.orgI'm in there and don't see you20:45
@clarkb:matrix.orgalso it doesn't look like nginx was restarted20:45
@fungicide:matrix.orgstarting back up now20:47
@clarkb:matrix.orglooks good thanks again20:49
@fungicide:matrix.orgno sweat20:49
@fungicide:matrix.orgthanks for the change20:49
@fungicide:matrix.orgi'm working on seeing if i can switch the bridge node label in the arm base job now20:50
-@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 988698: Use mixed arches for system-config-run-base-arm64 https://review.opendev.org/c/opendev/system-config/+/98869820:58
@fungicide:matrix.orgi think that's ^ it20:58
@clarkb:matrix.orgthis is interesting while 3.12 -> 3.13 doesn't change the index schemas/versions it does change the gerrit_file_diff h2 cache version from 9 to 10 and it will prune entries out that aren't the right version20:59
@clarkb:matrix.orgsimilarly when you downgrade it prunes out the newer versions stuff so that it can replace the entries with old versions20:59
@clarkb:matrix.orgI don't think this is a big deal. Just something I noticed that is interesting20:59
@fungicide:matrix.orghuh, yeah since it's just caches i think that's fine21:00
@clarkb:matrix.orgok https://etherpad.opendev.org/p/gerrit-upgrade-3.13 is updated with notes from actually running through the process to upgrade and downgrade on a held node21:00
@clarkb:matrix.orgfungi: there are a couple things that may be working looking at. The first is I updated the commands for checking the image shas to use podman commands (since docker commands don't work anymore) and I tried to make it more clear that is a two step process (first find the image id then inspect the image id). Then second in the dowgrade process I marked two steps as not being done for this upgrade but kept them in there and struck them through with a note why we skip them21:01
@clarkb:matrix.orgI'm curious to know if you find that to be clearly communicated in both cases where I chagned things compared to the last upgrade21:02
@clarkb:matrix.orgI haven't seen anything concerning in this process. I think the next step is landing https://review.opendev.org/c/opendev/system-config/+/987375 and upgrading prod to 3.12.7 then I can retest with the latest images and work through the TODO items on that etherpad against the latest images21:03
@clarkb:matrix.orgmaybe we can do that tomorrow if things are calm21:03
@fungicide:matrix.orgyeah, i can take a look tomorrow, i need to do a quick outing to pick up our grocery order but otherwise plan to be around all day21:48
@fungicide:matrix.orghuh, testinfra failures (ssh permission denied) on system-config-run-base-arm64 now: https://zuul.opendev.org/t/openstack/build/d2df0429fd1246bf9c23514e6205952321:50
@fungicide:matrix.orgi'll wip the change and look closer at that tomorrow too21:51
@clarkb:matrix.orgwe might have some bad assumptions about using local connectivity or non public ips in the test infra testing21:55
@clarkb:matrix.orgwhich now that bridge is on another cloud maybe creating problems?21:55
@clarkb:matrix.orgbut if we got far enough to run testinfra that implies ansible itself is working?21:55

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!