Tuesday, 2026-08-11

@arnaudm:matrix.orghey team, do you know who is managing https://www.stackalytics.io/ ? It seems the barbican metrics are not collected correcly10:37
@mnasiadka:matrix.orgI think the tool that OpenInfra and Linux Foundation are exposing for this case is https://insights.linuxfoundation.org - stackalytics has been sort of on and off and incosistently managed outside of OpenDev/OpenInfra11:07
@arnaudm:matrix.orgack thanks mnasiadka 12:06
@fungicide:matrix.orghuh, gerrit seems to maybe be down, i'm having trouble reaching it12:24
@fungicide:matrix.orgwill check the nova status12:24
@fungicide:matrix.orgno, looks like maybe i'm having ipv6 problems, i can reach it over ipv412:25
@fungicide:matrix.orgyeah, i can't reach v6 addresses in other networks either12:27
@fungicide:matrix.orglooks like my isp's v6 routing has broken somewhere past the first hop, since i can get responses from my firewall internally and it reports its v6 default gateway is reachable12:39
@fungicide:matrix.orgcorvus: please let me know when you're done editing the passwords list13:15
@jim:acmegating.comfungi: done13:23
@fungicide:matrix.orgthanks corvus!13:29
-@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 1000395: Document domain creation for Mailman https://review.opendev.org/c/opendev/system-config/+/100039514:04
@mnasiadka:matrix.orghttps://review.opendev.org/c/opendev/system-config/+/999795 needs a second +2 and we'll get node-exporter on the prometheus hsot16:07
@mnasiadka:matrix.org* https://review.opendev.org/c/opendev/system-config/+/999795 needs a second +2 and we'll get node-exporter on the prometheus host16:08
@fungicide:matrix.org#status log Manually cleaned up cronjobs for the retired backup02.ca-ymq-1.vexxhost Borg target from all production servers16:15
@status:opendev.org@fungicide:matrix.org: finished logging16:15
@clarkb:matrix.orgfungi: thank you16:37
@clarkb:matrix.orgmnasiadka: ya maybe fungi will be able to review that change? its on the meeting agenda to followup on that too later today16:38
@clarkb:matrix.orgalso apparently its the beginning of august when service coordinator election stuff should happen and not the end... I'll work on a last minute draft proposal that we can discuss in the meeting really quickly before making something official later today16:39
@clarkb:matrix.orgReminder that the weekly team meeting will be in the new #opendev-meeting:opendev.org room today and not the old #opendev-meeting channel on OFTC18:26
@clarkb:matrix.orgfungi: backup03 is the failing backup target19:16
@clarkb:matrix.orgfungi: you can see that in the backup03 log file on review03. Looks like a permissions error on the target side. Maybe due to the compaction?19:17
@fungicide:matrix.orgah, possible something with the compaction triggered those errors to start19:17
@clarkb:matrix.orgits possible we need to prune and compact as the user owning the backups rather than as root19:17
@fungicide:matrix.orgyeah, entirely possible. i think that was the one i did by hand before i edited the script to do the others19:17
@fungicide:matrix.orgmaybe umask differed19:17
@clarkb:matrix.orgso probably just need to cross check the perms and ownership of the data files on some other backup target then correct review03's data files19:19
@fungicide:matrix.orgyeah, i can look shortly. between meetings and discovering my isp has broken their ipv6 routing while i was sleeping, my day hasn't been all that productive19:20
@fungicide:matrix.orgokay, should be fixed now. indeed some of the files were root-owned because i missed that the backup operations are run in the context of the per-server users19:32
@fungicide:matrix.orgi did a `chown -R borg-review03:borg-review03 /opt/backups/borg-review03/backup` on backup03.ca-ymq-1.vexxhost19:32
@fungicide:matrix.orgthe others should be fine because i edited the command into the prune-borg-backups script to do them, which runs commands as the correct user for each repository19:34
@fungicide:matrix.orgas for the `Found errors in Matrix room configuration!` failure on 1000376, i can reproduce it locally where it's also uselessly silent, but at least i can try to step through the checker and find where it goes slantways19:43
@clarkb:matrix.orgfungi: corvus its the channel collision19:49
@clarkb:matrix.orgsince limnoria is still in #opendev-meeting in irc19:49
@clarkb:matrix.orgI think thats fine since we know not to run a meeting there for now. But we can fix that up and get the change merged19:49
@fungicide:matrix.orgyeah, i just spotted that as well19:49
@fungicide:matrix.organd had previously been considering checking whether we had a change to drop the bot from that channel19:49
@fungicide:matrix.orgi'll propose a change that adds some actual output for that check ;)19:50
@jim:acmegating.comthanks, i'll fixup the original change and leave the linter output to fungi19:50
@jim:acmegating.comthough...19:51
@jim:acmegating.comi mean, there isn't a conflict19:51
@jim:acmegating.comso maybe the change should be to not make that a conflict :)19:51
@clarkb:matrix.orgdo they write to the same log files?19:51
@clarkb:matrix.orgI think that may be the concern19:51
@clarkb:matrix.orgwith the logging I mean19:51
@jim:acmegating.comyes, but that's set explicitly; so a check for colliding log files would make sense19:52
@jim:acmegating.com(it's not because they have the same name)19:52
@clarkb:matrix.orgyes I think the check is specifically on the log paths19:52
@jim:acmegating.comok if that's what it's checking, then just, like outputting the error would be good then :)19:53
@clarkb:matrix.organyway the bot should leave matrix in ~ 15 minutes and then we can figure it out from there19:53
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [opendev/system-config] 1000376: Add matrix-eavesdrop to new #opendev-meeting room https://review.opendev.org/c/opendev/system-config/+/100037619:54
@fungicide:matrix.orgbecause of my broken ipv6 i have to wait for git to time out and fall back to v4, git-review doesn't seem to handle the `-4` option like some git subcommands19:55
@fungicide:matrix.orgso eventually i'll have the change for that showing up, but i lost the race to have it fail stacked on top of the broken state of 1000376,119:55
@clarkb:matrix.orghow concerned should we be about corrupting the log for the opendev-meeting channel/room? I think its not a big deal19:59
@clarkb:matrix.orgworst case we have incomplete data from today and we can call those growing pains19:59
@fungicide:matrix.orgi'm sure it's not a big deal19:59
@clarkb:matrix.orgcool we will need to manually restart limnoria to pick up the channel drop there20:00
@clarkb:matrix.orgI'm going to grab lunch now. Then will send an email about service coordinator elections. Then can help with the bot fixup/cleanup stuff20:00
-@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 1000559: Meetbot linting: Add output for room collisions https://review.opendev.org/c/opendev/system-config/+/100055920:01
@fungicide:matrix.orgsgtm, thanks!20:11
-@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [opendev/system-config] 1000376: Add matrix-eavesdrop to new #opendev-meeting room https://review.opendev.org/c/opendev/system-config/+/100037620:51
@clarkb:matrix.orgI just sent the election email. Hopefully I got all the dates right21:00
@clarkb:matrix.orgfungi: re ipv6 I have a small sysctl script to disable and enable ipv6 globally so that my wireguard traffic doesn't go over ipv6 and skip the tunnel. You could do something like that until your ISP corrects the upstream issue21:01
@fungicide:matrix.orgyeah, i've been hoping it gets corrected soon, but with it having been this way for most of a day already my hopes are dwindling. i'll take more aggressive measures if it's ongoing tomorrow21:02
@clarkb:matrix.org1000376's deploy failed in the base job on a lock contention issue with apt/dpkg on mirror02.regionone.osuosl.opendev.org21:05
@clarkb:matrix.orgthats "ok" the hourly job will deploy it momentarily21:05
@clarkb:matrix.orgI don't think we need to reenqueue anything21:05
@clarkb:matrix.orglooks like apt-get has been running since 11:43 UTC21:06
@clarkb:matrix.orgfungi: should I try to kill that process then run a manual apt-get update (the command that is stuck) to see if it succeeds?21:06
@clarkb:matrix.orgalso both bot configs have udpated at this point. Any objections to me restarting limnoria now?21:10
@clarkb:matrix.orgthe osuosl mirror's apt-get process was killed then I reran apt-get update by hand and that succeeded21:11
@clarkb:matrix.orgok no objections I am restarting the limnoria bot now21:16
@fungicide:matrix.orglooking21:43
@fungicide:matrix.orgClark: i don't see apt-get running, maybe it finished?21:44
@clarkb:matrix.orgfungi: I killed it22:08
@clarkb:matrix.orgthen reran apt-get update by hand to make sure it would exit in the current state of things22:08

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