| *** rlandy is now known as rlandy|out | 00:53 | |
| *** yadnesh|away is now known as yadnesh | 05:12 | |
| *** pojadhav is now known as pojadhav|ruck | 05:14 | |
| ianw | i'm running a prune on the backup02 | 06:11 |
|---|---|---|
| ianw | root screen | 06:11 |
| *** akahat|rover is now known as akahat | 07:05 | |
| *** pojadhav is now known as pojadhav|ruck | 07:24 | |
| *** jpena|off is now known as jpena | 08:43 | |
| *** elodilles is now known as elodilles_afk | 09:39 | |
| opendevreview | Tobias Henkel proposed zuul/zuul-jobs master: Fix creating bucket in s3 log upload https://review.opendev.org/c/zuul/zuul-jobs/+/872104 | 09:52 |
| *** rlandy|out is now known as rlandy | 11:09 | |
| *** elodilles_afk is now known as elodilles | 11:12 | |
| *** dviroel_ is now known as dviroel|rover | 11:24 | |
| *** ysandeep is now known as ysandeep|afk | 11:55 | |
| *** ysandeep|afk is now known as ysandeep | 12:56 | |
| *** tkajinam is now known as Guest2882 | 13:11 | |
| *** dasm|off is now known as dasm | 13:57 | |
| *** yadnesh is now known as yadnesh|away | 15:01 | |
| fungi | lucasagomes: our root alias got an invite from rtd to join ovn-bgp-agent, supposedly initiated by your account there. did you need something done? | 15:15 |
| *** ysandeep is now known as ysandeep|out | 15:23 | |
| fungi | okay, so getting closer on the mm3 fixups. on the new held node (104.130.219.31) it still had the initial mailman site as lists.opendev.org rather than mailman.opendev.org so i'm clearly missing something there. after creating corresponding sites and mailman domains through the django admin interface i can see both postorius and hyperkitty correctly filtering the lists for each site. | 15:57 |
| fungi | unfortunately hyperkitty still says lists.openstack.org in the top-left corner of all the sites, so i need to figure out what's missing for that part (i thought we had already addressed it in the proposed changes, but this would indicate otherwise) | 15:57 |
| fungi | i'm restarting the containers on the held node to see if anything changes | 15:58 |
| lucasagomes | fungi, hi there, ah... I created the account and added the openstackci as an admin | 16:12 |
| lucasagomes | fungi, lemme get the patch | 16:12 |
| lucasagomes | fungi, https://review.opendev.org/c/x/ovn-bgp-agent/+/872032 | 16:13 |
| lucasagomes | fungi, does it need some approval ? | 16:13 |
| Clark[m] | As a heads up slow start for me today due to a doctor's checkup | 16:13 |
| fungi | lucasagomes: i'm not super familiar with rtd, so unsure whether the refresh trigger in the rtd api requires the zuul account to be a member of the project. if so, we'll need to accept the invitation from rtd, i think? | 16:14 |
| fungi | okay, good news on the mm3 front... after a full container restart and page refresh, hyperkitty is now showing the lists' individual display names as intended | 16:15 |
| *** dviroel|rover is now known as dviroel|rover|lunch | 16:15 | |
| fungi | so it seems like all we really need at this point is a mechanism for adding the sites/domains in django and then restarting the containers maybe? (possible the container restart is only needed after initial installation and not on addition of new domains) | 16:16 |
| lucasagomes | fungi, I see... I will keep an eye on it on future updates in the doc. If needed I will ask you to accept the invitation | 16:16 |
| lucasagomes | fungi, thanks for the heads up! | 16:16 |
| fungi | lucasagomes: note that the invite has a two-week expiration on it, so try to let us know before then | 16:17 |
| lucasagomes | fungi, will do | 16:17 |
| lucasagomes | we have a patch changing the docs right now | 16:17 |
| lucasagomes | should be merged this week | 16:17 |
| fungi | lucasagomes: basically, rtd rebuilds the content periodically if it notices the repository has changed, the zuul job hits a webhook to tell it to check immediately, so in theory you should see the content update within a few minutes of the job running | 16:18 |
| lucasagomes | fungi, I see... if that's the case there's no need to have the openstack user as admin. I just added because I had another project on rtd where that openstack user was the admin | 16:21 |
| lucasagomes | so I mimic'd it | 16:21 |
| lucasagomes | but should be fine then | 16:21 |
| lucasagomes | lemme revoke the invitation and see if it will work without it | 16:22 |
| *** marios is now known as marios|out | 16:54 | |
| *** dviroel|rover|lunch is now known as dviroel|rover | 17:20 | |
| *** jpena is now known as jpena|off | 17:39 | |
| clarkb | fungi: so if I'm parsing the mm3 stuff correctly these are the steps we need to take: deploy mm3 with site id = 1 and create vhosts, switch site id to 0, add django web host entries for each domain, restart containers? | 17:47 |
| fungi | nah, the only thing that needed SITE_ID=1 is the initial database setup | 17:50 |
| fungi | so the current change temporarily switches to SITE_ID=1, does the db setup, then switches back to SITE_ID=0 and proceeds with the deployment | 17:51 |
| fungi | there's a flagfile created at ~mailman/bootstrapped which indicates that initial db setup has been performed so that ansible knows not to toggle the config again after the initial deployment | 17:52 |
| clarkb | gotcha so is the process more like set site_id = 1, create db, set site_id = 0, deploy the rest of mm3, update web hosts in django, restart containers? | 17:53 |
| fungi | i think so, yes | 17:57 |
| fungi | at least i did those later steps on the held node and they seemed to work. i didn't touch settings.py at all | 17:57 |
| fungi | https://review.opendev.org/867987 is the current state of that | 17:57 |
| fungi | looks like i added the config revert after the list creation step, but can try moving it earlier | 17:58 |
| fungi | the biggest unknown is what is involved in creating sites and domains through django's api instead of the admin webui | 18:00 |
| clarkb | fungi: there aer a number of django commands through the management script iirc. Its possible there is one that might do this? | 18:03 |
| clarkb | if not we may need to insert into the db directly (ugh). I think I looked at the mm3 api and it doesn't have it exposed that way | 18:04 |
| opendevreview | Jeremy Stanley proposed opendev/infra-manual master: General updates for the Test Environment chapter https://review.opendev.org/c/opendev/infra-manual/+/872201 | 18:17 |
| clarkb | fungi: looks like the django-mailman3 repo's MailDomain has a foreign key on django.contrib.sites.models.Site and it is this that we need to update (both the mailmaindomain table and the site table) | 18:17 |
| fungi | yeah, adding them through the admin webui you can clearly see the dependency (once you create sites, you can choose them in the drop-down while creating domains) | 18:19 |
| clarkb | I'm not seeing any obvious command line tools for this :/ | 18:20 |
| fungi | argh | 18:20 |
| fungi | i can ask on the ml | 18:20 |
| clarkb | https://pypi.org/project/django-api-admin/ would seem to indicate that django doesn't generically expose this if there is a special library made to do some of this | 18:21 |
| fungi | i asked on mailman-users, maybe someone will come back with a less fragile solution | 18:32 |
| fungi | https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message/OHAKSEYYL6REAHWEEZZRA7YZM4O7DPLH/ | 18:33 |
| fungi | clarkb: revisiting https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/thread/JCRYKMJKCRAWNTZPGH2MKGJPGPOMCHJZ/ the last message seems to indicate that poking things directly into the db is indeed the alternative to using the admin webui, but hopefully a response to my post will confirm or refute that | 18:40 |
| clarkb | ack | 18:50 |
| fungi | also i think my addition of a mailman.opendev.org site didn't do anything because we don't create it if we don't have any lists to add to it, so i'll update the change to make a test list and see if that works | 18:54 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Revert "Update git in gitea images" https://review.opendev.org/c/opendev/system-config/+/872209 | 19:11 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Revert "hound: use updated git packages" https://review.opendev.org/c/opendev/system-config/+/872211 | 19:13 |
| clarkb | if you accidentall use -d instead of -t git-review fails, but without a clear indication of why (potential improvement here if someone has time) | 19:14 |
| fungi | oh -d without a change number? | 19:15 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Flip Gerrit image to install distro git https://review.opendev.org/c/opendev/system-config/+/872212 | 19:17 |
| clarkb | fungi: ya -d foo_topic | 19:17 |
| clarkb | my typing driver failed me there :) | 19:18 |
| clarkb | its a super minor thing, just thought I'd mention it | 19:19 |
| ianw | #status log ran backup prune on backup02.ca-ymq-1 | 20:43 |
| opendevstatus | ianw: finished logging | 20:43 |
| opendevreview | Ade Lee proposed zuul/zuul-jobs master: Add ubuntu to enable-fips role https://review.opendev.org/c/zuul/zuul-jobs/+/866881 | 21:35 |
| opendevreview | Ade Lee proposed openstack/project-config master: Add base openstack FIPS job https://review.opendev.org/c/openstack/project-config/+/872222 | 21:36 |
| *** dviroel|rover is now known as dviroel|out | 21:54 | |
| opendevreview | Ian Wienand proposed zuul/zuul-jobs master: promote-docker-image: also accept OCI manifest formats https://review.opendev.org/c/zuul/zuul-jobs/+/872230 | 22:40 |
| opendevreview | Merged zuul/zuul-jobs master: promote-docker-image: also accept OCI manifest formats https://review.opendev.org/c/zuul/zuul-jobs/+/872230 | 23:03 |
| *** rlandy is now known as rlandy|bbl | 23:14 | |
| opendevreview | Merged opendev/system-config master: Revert "hound: use updated git packages" https://review.opendev.org/c/opendev/system-config/+/872211 | 23:56 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!