| opendevreview | Clark Boylan proposed opendev/system-config master: Add a mailman3 list server https://review.opendev.org/c/opendev/system-config/+/851248 | 00:00 |
|---|---|---|
| opendevreview | Clark Boylan proposed opendev/system-config master: DNM force mm3 failure to hold the node https://review.opendev.org/c/opendev/system-config/+/855292 | 00:00 |
| johnsom | Hi all, I'm not sure who to bug about the openinfra/openstack-map repo, but maybe someone here can approve this typo fix patch for Designate? It causes Designate to not be listed for tripleo. https://review.opendev.org/c/openinfra/openstack-map/+/840774 | 00:32 |
| johnsom | It's been un-reviewed since May. | 00:32 |
| ianw | seems like ttx mostly +2's things there? | 00:39 |
| opendevreview | Merged zuul/zuul-jobs master: zuul-jobs-test-ensure-python-pyenv: update matchers https://review.opendev.org/c/zuul/zuul-jobs/+/853913 | 00:52 |
| opendevreview | Merged zuul/zuul-jobs master: Fedora : update to Fedora 36 release nodes https://review.opendev.org/c/zuul/zuul-jobs/+/853909 | 01:03 |
| *** dasm|off is now known as Guest1645 | 04:00 | |
| dpawlik | ianw: hey, could you tell me what is correct DIB_RELEASE variable value for centos 9 stream image ? | 07:03 |
| dpawlik | is is: 9 or 9-stream? | 07:03 |
| ianw | dpawlik: it should be 9-stream | 07:04 |
| dpawlik | if 9, ok, because many conditions are checking just a numbers, but if 9-stream some of them needs to be fixed | 07:04 |
| ianw | in a lot of places though we are checking =~ 9 to match both a rhel and a 9-stream path though | 07:05 |
| ianw | basically, "does it start with 9" | 07:05 |
| dpawlik | yhym | 07:05 |
| dpawlik | will check more carefully. | 07:06 |
| opendevreview | daniel.pawlik proposed openstack/diskimage-builder master: Change conditions for checking if DIB_RELEASE is greater than 7 https://review.opendev.org/c/openstack/diskimage-builder/+/855315 | 07:21 |
| opendevreview | daniel.pawlik proposed openstack/diskimage-builder master: Change conditions for checking if DIB_RELEASE is greater than 7 https://review.opendev.org/c/openstack/diskimage-builder/+/855315 | 07:32 |
| dpawlik | cc ianw ^^ | 07:32 |
| *** jpena|off is now known as jpena | 07:33 | |
| ttx | johnsom: oh, I missed that one. Should probably add it to my dashboard... | 08:44 |
| ttx | sorry about that | 08:45 |
| dpawlik | clarkb: thanks for the review 855014. Still in some place it want to install python3-virtualenv https://softwarefactory-project.io/nodepool-builder/nodepool-builder.softwarefactory-project.io/upstream-centos-9-stream-0000000804.log | 08:58 |
| *** dviroel|out is now known as dviroel | 11:23 | |
| *** Guest1645 is now known as dasm | 13:50 | |
| fungi | clarkb: i've added our timing and disk usage measurements to the mm3migration pad, as well as a rough draft migration plan | 13:51 |
| fungi | i also crossed off the completed todo entries and added a couple more | 13:57 |
| clarkb | fungi: thanks left a couple of notes on your notes. But what you have lgtm | 15:00 |
| *** dviroel is now known as dviroel|lunch | 15:25 | |
| clarkb | fungi: also the newer held node has an updated apache which is set up to hopefully host pipermail archives. If you want to take a look at the updates to the change and/or test that on the held node | 15:28 |
| clarkb | fungi: I've used https://github.com/maxking/docker-mailman/search?q=html_to_plain_text_command to double check the docker images aren't using some other tool. Seems like it really is a bug that lynx isn't installed. I think we can file an issue and send a PR to address that one | 15:43 |
| clarkb | I'll put that together later today. First I need to get some exercise before the day gets too hot | 15:43 |
| *** dviroel|lunch is now known as dviroel | 16:25 | |
| fungi | i'm still just trying to update the jitsi docker env files and getting increasingly lost... looks like we "updated" them to be based on a newer config at some point, but still left a lot of options in our copy which were removed upstream | 16:41 |
| fungi | all the xmpp settings, for example. do we need those? | 16:41 |
| fungi | they were in the upstream env example at one point when we originally copied them, but we never removed them when updating to a newer env example which had dropped them | 16:41 |
| *** jpena is now known as jpena|off | 16:43 | |
| fungi | the entire "Advanced configuration options" section of the sample is actually gone | 16:43 |
| fungi | aha, that came in cb5a753 "config: simplify configuration" where they dropped most stuff if the example relied on default values | 16:45 |
| fungi | so basically i need to look through every one of these settings, see if they were default at that time and whether they're still default, and then drop them if so | 16:52 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Update Jitsi configs to latest upstream samples https://review.opendev.org/c/opendev/system-config/+/855388 | 17:08 |
| fungi | that's wip since it's just the docker envs so far | 17:08 |
| fungi | clarkb: it seems like meet.conf and settings-config.js are extracted from running containers? what's the best way to get updated versions of those? | 17:17 |
| clarkb | fungi: I think I just pulled them out of their git repo for the dockerfiles | 17:47 |
| clarkb | fungi: https://github.com/jitsi/docker-jitsi-meet/blob/master/web/rootfs/defaults/settings-config.js | 17:48 |
| clarkb | https://github.com/jitsi/docker-jitsi-meet/blob/master/web/rootfs/defaults/meet.conf ? | 17:48 |
| fungi | oh, yep there they are | 17:48 |
| fungi | sorry, i got it in my head those were in the jitsi-meet repo like the interface config | 17:49 |
| clarkb | I decided to skip filing an issue and making a PR. Just a PR seems sufficient for this one line update: https://github.com/maxking/docker-mailman/pull/552 | 17:58 |
| fungi | oh, awesome | 18:00 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Speed up log file fetching tasks https://review.opendev.org/c/zuul/zuul-jobs/+/855402 | 18:19 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Speed up log file fetching tasks https://review.opendev.org/c/zuul/zuul-jobs/+/855402 | 18:21 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Update Jitsi configs to latest upstream samples https://review.opendev.org/c/opendev/system-config/+/855388 | 18:45 |
| fungi | infra-root: okay, i *think* that's ^ complete | 18:45 |
| clarkb | fungi: my PR to the docker images seems to be breaking beacuse django has invalid i18n strings: https://app.circleci.com/pipelines/github/maxking/docker-mailman/1171/workflows/d518e14b-08f7-4af9-ba1e-a949be339788/jobs/2973?invite=true#step-106-1537 | 18:46 |
| fungi | yeesh | 18:46 |
| clarkb | fungi: do you know why .Env. prefixes were removed from some vars in https://review.opendev.org/c/opendev/system-config/+/855388/2/playbooks/roles/jitsi-meet/files/meet.conf but not all? | 18:49 |
| clarkb | fungi: left some comments, but I'll try to compare and sanity check after lunch | 18:53 |
| fungi | i do not know, but can probably go diving in their revision history to find out | 19:00 |
| *** dviroel is now known as dviroel|afk | 19:47 | |
| fungi | clarkb: is this fatal? https://zuul.opendev.org/t/openstack/build/1575fb542b584a9c9890aaae8a9d2403/log/meetpad01.opendev.org/docker/jitsi-meet-docker_web_1.txt | 19:51 |
| fungi | testinfra said it couldn't connect to 443/tcp, so i'm guessing that's it | 19:51 |
| fungi | https://github.com/jitsi/docker-jitsi-meet/blob/master/web/rootfs/defaults/meet.conf#L91 is what i think it's complaining about | 19:51 |
| clarkb | yes I suspect that is fatal | 19:57 |
| fungi | our compose file doesn't specify any particular versions, so i guess we install latest always? | 19:57 |
| clarkb | yes we install latest | 19:58 |
| clarkb | https://github.com/jitsi/docker-jitsi-meet/blob/1c93e1b74d773706a2e780458950231148ee0003/web/rootfs/defaults/nginx.conf#L59 do we override nginx.conf too? if so maybe we need to update it to pick this up | 19:58 |
| clarkb | but I don't know enough about nginx to know what it expects from that | 19:58 |
| fungi | oh | 19:59 |
| fungi | i also notice our compose file is forked from the docker-jitsi-meet repo so may need to merge from a new version of that too | 19:59 |
| clarkb | infra-root if you have time for https://review.opendev.org/c/opendev/system-config/+/855279 we suspcet that will fix refstack result uploads | 20:02 |
| fungi | it doesn't appear as though we have previously forked their nginx.conf file so it should still be the one in their master branch | 20:03 |
| fungi | looks like all that was added several months ago in https://github.com/jitsi/docker-jitsi-meet/commit/0cbfbfd | 20:08 |
| fungi | i'll let it rattle around in my noggin whilst knocking out some yardwork | 20:08 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Speed up log file fetching tasks https://review.opendev.org/c/zuul/zuul-jobs/+/855402 | 20:11 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Exercise stage-output extensions_to_txt in testing https://review.opendev.org/c/zuul/zuul-jobs/+/855434 | 20:11 |
| clarkb | fungi: googling around does show that map line I linked to is what defines that variable | 20:14 |
| fungi | i don't see it in https://zuul.opendev.org/t/openstack/build/1575fb542b584a9c9890aaae8a9d2403/log/meetpad01.opendev.org/jitsi-meet/web/nginx/nginx.conf | 20:20 |
| fungi | where would that be coming from? | 20:20 |
| fungi | maybe we're supposed to be copying that file somewhere and it's missing? | 20:22 |
| clarkb | fungi: I think https://hub.docker.com/r/jitsi/web/tags?page=1&name=latest is the issue compare to the stable tag | 20:23 |
| clarkb | I think the lesson here is that latest is unreliable because people will stop using it randomly :/ | 20:24 |
| clarkb | I suspec that if you update to the stable tag it will work | 20:24 |
| fungi | oh, interesting | 20:27 |
| fungi | but these are the configs from the master branch | 20:27 |
| clarkb | yes they stopped updating their latest tag 4 months ago and we're using configs that are newer than that | 20:27 |
| fungi | i notice there's an unstable tag mentioned as the default in their compose file on master, i guess that's it | 20:28 |
| clarkb | There is no stbale or unstable branch I think that has to do with what jitsi they install and then tag the appropriate image uploads? | 20:28 |
| fungi | right, i'm as of yet unclear on how to map latest to a git state | 20:28 |
| clarkb | ya the stable tags include jitsi meet releases | 20:28 |
| fungi | oh, latest maps to a stable image tag which then corresponds to a git tag name? | 20:29 |
| clarkb | fungi: maybe grab the config state based on when they updated the changelog? | 20:29 |
| fungi | maybe | 20:29 |
| clarkb | it maps to a jitsi meet git tag | 20:29 |
| clarkb | not a gitsi meet docker repo git tag | 20:29 |
| fungi | i'll try to take a closer look after yardwork | 20:29 |
| clarkb | oh actually maybe both. Github's filter doesn' tseem to work properly | 20:30 |
| clarkb | fungi: https://github.com/jitsi/docker-jitsi-meet/tree/stable-7648-4 that appears to be the ocntent for the most recent stable images | 20:30 |
| fungi | thanks! | 20:31 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Exercise stage-output extensions_to_txt in testing https://review.opendev.org/c/zuul/zuul-jobs/+/855434 | 20:38 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Speed up log file fetching tasks https://review.opendev.org/c/zuul/zuul-jobs/+/855402 | 20:38 |
| clarkb | fungi: I think the mailman container build issues may originate in the upstream mailman translations | 20:40 |
| opendevreview | Merged opendev/system-config master: Install refstack with openstack constraints https://review.opendev.org/c/opendev/system-config/+/855279 | 20:46 |
| clarkb | once that deploys I think we need to manually restart the service. I'm not 100% sure of that but I'll take a look later | 20:47 |
| clarkb | it apperas to have restarted itself | 20:52 |
| fungi | how nice of it! | 20:59 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Speed up log file fetching tasks https://review.opendev.org/c/zuul/zuul-jobs/+/855402 | 21:22 |
| *** dasm is now known as dasm|off | 21:40 | |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Speed up log file fetching tasks https://review.opendev.org/c/zuul/zuul-jobs/+/855402 | 21:40 |
| *** rlandy is now known as rlandy|bbl | 22:25 | |
| clarkb | fungi: you haven't used the newer of the two mm3 held nodes yet have you? I'm working on the mariadb my.cnf update for max allowed packet and would like to rotate thta held node out hwen I push the change up | 22:49 |
| fungi | i have not, no | 22:50 |
| fungi | and i don't need the older one now if you want to trash them both | 22:50 |
| opendevreview | Merged zuul/zuul-jobs master: Exercise stage-output extensions_to_txt in testing https://review.opendev.org/c/zuul/zuul-jobs/+/855434 | 23:03 |
| *** dviroel|afk is now known as dviroel | 23:12 | |
| opendevreview | Clark Boylan proposed opendev/system-config master: Add a mailman3 list server https://review.opendev.org/c/opendev/system-config/+/851248 | 23:15 |
| opendevreview | Clark Boylan proposed opendev/system-config master: DNM force mm3 failure to hold the node https://review.opendev.org/c/opendev/system-config/+/855292 | 23:16 |
| clarkb | if I didn't fail at ansible I think that will do it for us. Ne whold in place. I cycled out the old new one but will leave the old old one for now since it potentially has good data on it | 23:17 |
| clarkb | fungi: an interesting thing I learned is that by default mysqldump has a higher value than mysql. 24MB and 16MB respecitvely if I read docs properly | 23:20 |
| clarkb | The issue is mysqldump will apparently be more verbose or something in some scenarios | 23:21 |
| clarkb | anyway that update should keep a larger value for mysqldump, but we should check our backups function with the large data in the db | 23:21 |
| clarkb | In theory we can't insert anything larger than the smaller value that mysql respects so we'll be fine | 23:21 |
| fungi | cool, i should probably plan to do another openstack-discuss import on the new hold | 23:36 |
| fungi | and make sure it runs clean | 23:37 |
| clarkb | ++ | 23:39 |
| clarkb | there is at least one bug with that. Apparnetly mysqlupgrade is run by the container and it reads client configs but it doesn't support that flag | 23:51 |
| clarkb | but then also https://zuul.opendev.org/t/openstack/build/8e4d10ba5a5348f8ba104dd150750277/log/lists99.opendev.org/docker/mailman-compose_database_1.txt#88 :/ | 23:52 |
| clarkb | I'm going to fix the client issue in case that is causing the booted up db to be sad | 23:52 |
| clarkb | But I suspect something else is unhappy | 23:52 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Add a mailman3 list server https://review.opendev.org/c/opendev/system-config/+/851248 | 23:53 |
| opendevreview | Clark Boylan proposed opendev/system-config master: DNM force mm3 failure to hold the node https://review.opendev.org/c/opendev/system-config/+/855292 | 23:53 |
| clarkb | and now dinner | 23:54 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!