Tuesday, 2023-01-17

clarkbianw: nodepool applies a regex filter on the flavor name matches iirc00:43
clarkbwe might need something like opendev$ in that field00:44
ianwclarkb: i don't think it does, it's just "in"; but i am proposing we sort it alphabetically00:46
clarkblast call on meeitng agenda updates. I'm going to send that out shortly00:49
*** soniya29|rover is now known as soniya29|rover|brb07:12
*** soniya29|rover|brb is now known as soniya29|rover08:15
*** ysandeep is now known as ysandeep|lunch08:19
*** jpena|off is now known as jpena08:21
*** soniya29|rover is now known as soniya29|rover|lunch09:16
*** ysandeep|lunch is now known as ysandeep10:09
*** soniya29|rover|lunch is now known as soniya29|rover10:44
*** rlandy|out is now known as rlandy11:14
*** dviroel|out is now known as dviroel11:44
*** gthiemon1e is now known as gthiemonge12:40
fungirackspace notified us that our trove instance named "zuul-v3" experienced an unplanned outage. am i correct in my assumption that it's safe for us to delete that now?13:29
*** obondarev_ is now known as obondarev14:02
*** dasm|off is now known as dasm14:07
*** ysandeep is now known as ysandeep|out14:28
*** soniya is now known as soniya|rover14:34
*** dviroel is now known as dviroel|lunch15:28
Tengufungi, clarkb heya! would it be possible to get https://review.opendev.org/c/opendev/system-config/+/869819 in (ansible-galaxy proxy). fungi I've answer your comment about the CacheIgnoreQueryString - imho it's safe and isolated enough15:37
Tenguthe context for this setting is: server config, virtual host15:38
fungiyeah, i forgot that this was a dedicated vhost section (due to the path gymnastics)15:38
Tenguyeah, sorry for that one15:38
fungino worries15:38
TenguI tried to check what was going on in the other vhosts, it *should* be good as-is.15:38
Tengubut my knowledge about mod_cache is rusted.15:38
Tenguto not say "close to null"15:39
fungianyway, i've already voted +2 on it, just hoping to get a second infra-root sysadmin to check it's sensible15:39
Tenguso getting a good review on that would be great15:39
Tengufungi: sure :)15:39
NeilHanlonwell, it's not a +2 but ya can have my +1 on it. seems sane to me15:44
corvusfungi: re zuul-v3 -- how do we map the name of the trove instance to the dburi used in zuul.conf?15:50
*** soniya|rover is now known as soniya|rover|out15:52
fungicorvus: i think i can find it through the api. that said, i didn't realize we were still using trove for the schedulers now15:54
fungibut maybe we never moved off of it15:54
corvusfungi: yeah, `grep dburi /etc/zuul/zuul.conf` on a scheduler will show the uri (i don't want to paste here since i think we decided that we should keep even the hostnames secret)16:13
corvusbut it does appear to be a trove instance so i think we should double check16:13
fungiyep, did you happen to notice any errors related to database disconnects? i can dig once i get a moment for a context switch16:14
corvusi only checked that it's working now, let me try the logs16:14
fungiyeah, i've been merging things successfully at least16:15
corvusfungi: sql reporting errors around 14:55 jan 1616:15
corvus14:52 -- 14:5516:17
corvus([Errno 111] Connection refused)16:18
corvusthat's in /var/log/zuul/zuul.log.1 on zuul0116:18
fungiDate: Mon, 16 Jan 2023 14:24:18 +000016:19
fungithat's when we first got notified16:19
fungiDate: Mon, 16 Jan 2023 16:48:12 +000016:20
fungiis when we were notified of it being fixed16:20
fungiinteresting that our errors were after it was supposedly fixed, but maybe it took that long for the connection tim timeout and get reset16:20
fungier, connection to timeout16:21
*** marios is now known as marios|out16:29
*** dviroel|lunch is now known as dviroel16:35
opendevreviewClark Boylan proposed opendev/system-config master: Run mirror-update scripts with no timeout by default  https://review.opendev.org/c/opendev/system-config/+/84021417:00
clarkbfungi: ^ thank you for reviews on some of those older changes. I think another one needs a rebase too. Will try to get that cleaned up this morning too17:00
fungiclarkb: looking back through some older changes, that one may also be redundant with https://review.opendev.org/77070617:02
fungior, no, i'm thinking of https://review.opendev.org/85751917:03
clarkbah ya that takes a different appraoch but both should work. Looks like ianw's is simpler. I'm happy to review that one instead17:04
*** jpena is now known as jpena|off17:26
opendevreviewClark Boylan proposed opendev/system-config master: Update Gitea to 1.17.4  https://review.opendev.org/c/opendev/system-config/+/87085018:17
opendevreviewClark Boylan proposed opendev/system-config master: Update gitea to 1.18.1  https://review.opendev.org/c/opendev/system-config/+/87085118:17
clarkbthis was something I meant to catch up on after the holidays then got sick and completely forgot. I've remembered now :)18:17
clarkbthe 1.17.4 upgrade should be straightforward. THe 1.18 upgrade deserves more careful review18:17
opendevreviewClark Boylan proposed opendev/system-config master: Explicitly install jmespath alongside ansible on bridge  https://review.opendev.org/c/opendev/system-config/+/84333018:23
opendevreviewClark Boylan proposed opendev/system-config master: Up openafs client -stat value  https://review.opendev.org/c/opendev/system-config/+/85752018:24
funginew git security updates!18:53
*** Tengu7 is now known as Tengu18:59
fungieverybody loves security holes in their tools. at least i know i do19:24
NeilHanlonswiss _is_ my favourite cheese19:25
NeilHanlongoes with the holes in my firewall rules19:25
opendevreviewMerged opendev/system-config master: mirror-update: make jobs interactive by default  https://review.opendev.org/c/opendev/system-config/+/85751919:27
opendevreviewMerged openstack/project-config master: Add FIPS job for ubuntu  https://review.opendev.org/c/openstack/project-config/+/86711219:45
*** rlandy is now known as rlandy|brb19:50
clarkbianw: I've approved the mailman2 apache vhost update. THank you for adding the sites-enabled content to log collection, made checking things super easy19:55
ianwok, thanks.  i'll go add the right variable to bridge now19:57
*** dviroel_ is now known as dviroel|afk19:57
ianwclarkb: https://review.opendev.org/q/topic:save-published-logs was two others that came out of that ACL thing as well; that should save manage-projects runs locally and stop us creating empty files20:06
opendevreviewMerged opendev/system-config master: mailman: add variable for matching UAs in Apache  https://review.opendev.org/c/opendev/system-config/+/86977920:23
*** rlandy|brb is now known as rlandy20:43
clarkbianw: I approved the first change in that topic but only +2'd the second since it affects when things are encrypted. I wasn't sure how to verify that is working: https://review.opendev.org/c/opendev/system-config/+/869892 ?21:13
clarkbit should be fine though because either we encrypt or we don't have log files and that should prevent accidental disclosure21:13
clarkbianw: anyway when you have time to apply judgement to that feel free to approve or point me at how it is tested21:27
opendevreviewMerged opendev/system-config master: infra-prod: save run logs even if being published  https://review.opendev.org/c/opendev/system-config/+/86989121:36
ianwclarkb: yeah i think that only affects production and only the manage-projects job22:05
ianw(as that's the only one with the publish flag).  i.e. i think just monitor that is the testing :)22:06
ianwclarkb: dropped an idea on a stamp-file thing for the reindex upgrade with https://review.opendev.org/c/opendev/system-config/+/870114 ... but i'm open to the suggestion it's over-engineering something we can manage manually22:57
clarkbI also wonder if it is strictly required to do an offlien reindex. I found a bunch of issues in the docs around submit requirements which makes me generally skeptical22:59
opendevreviewMerged opendev/system-config master: infra-prod: don't encrypt logs if they are being published  https://review.opendev.org/c/opendev/system-config/+/86989223:04
opendevreviewMerged opendev/system-config master: Correct (again) how ansible-galaxy proxy is configured  https://review.opendev.org/c/opendev/system-config/+/86981923:08
ianwyeah, not sure, agree sometimes the instructions around that have been unclear23:09
opendevreviewClark Boylan proposed opendev/jeepyb master: Add launchpadlib as a dep  https://review.opendev.org/c/opendev/jeepyb/+/87087323:23
opendevreviewClark Boylan proposed opendev/system-config master: Build Gerrit on top of our python-base images  https://review.opendev.org/c/opendev/system-config/+/87087423:26
opendevreviewIan Wienand proposed openstack/project-config master: linaro-us : set max-servers to 0  https://review.opendev.org/c/openstack/project-config/+/87087623:41
clarkbwe'll need to merge the 3.5 drop change before I can accurately test if gerrit under java 17 is happy (since they share the base image)23:49
clarkbI guess I can push a change up for that and let it fail23:49
fungiinfra-prod-service-mirror failed deploying the ansible-galaxy proxy change: https://zuul.opendev.org/t/openstack/build/4cc4d1a62cea44aaacd4b11da13171ad23:52
opendevreviewClark Boylan proposed opendev/system-config master: Switch Gerrit to Java 17  https://review.opendev.org/c/opendev/system-config/+/87087723:55
clarkbfungi: mirror01.regionone.linaro.opendev.org was unreachable23:56

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!