| opendevreview | Merged openstack/diskimage-builder master: Reduce thin pool by one more extent https://review.opendev.org/c/openstack/diskimage-builder/+/868376 | 02:34 |
|---|---|---|
| *** rlandy is now known as rlandy|out | 02:48 | |
| *** ysandeep is now known as ysandeep|afk | 04:35 | |
| *** ysandeep|afk is now known as ysandeep | 06:35 | |
| *** akahat is now known as akahat|ruck | 07:01 | |
| *** akahat|ruck is now known as akahat|rover | 07:01 | |
| *** bhagyashris|ruck is now known as bhagyashris | 07:47 | |
| *** jpena|off is now known as jpena | 08:43 | |
| *** ysandeep is now known as ysandeep|lunch | 08:58 | |
| *** rlandy|out is now known as rlandy | 11:12 | |
| *** ysandeep|lunch is now known as ysandeep | 12:00 | |
| *** tosky_ is now known as tosky | 12:55 | |
| *** ysandeep is now known as ysandeep|afk | 12:56 | |
| opendevreview | Jeremy Stanley proposed opendev/zone-opendev.org master: Dummy mailman hostname to house the list of lists https://review.opendev.org/c/opendev/zone-opendev.org/+/867981 | 13:36 |
| *** ysandeep|afk is now known as ysandeep | 13:58 | |
| *** dasm|off is now known as dasm | 14:01 | |
| fungi | okay, so i think this is the reason the magic domain guessing change for mm3 is failing: https://zuul.opendev.org/t/openstack/build/71cb4e5491294775b9e9539a85d3c3c7/log/lists99.opendev.org/docker/mailman-web.txt#40184-40239 | 14:28 |
| fungi | ValueError: The database backend does not accept 0 as a value for AutoField. | 14:28 |
| *** ysandeep is now known as ysandeep|out | 14:41 | |
| *** dviroel|out is now known as dviroel|ruck | 15:08 | |
| yadnesh|away | hello, a few tempest jobs are failing with missing rust compiler, is this a known issue | 15:17 |
| yadnesh|away | https://zuul.opendev.org/t/openstack/build/a924ffd50ce4485591182d902b335501 | 15:17 |
| yadnesh|away | https://zuul.opendev.org/t/openstack/build/87f0e4e550e24540a801b7ea0352d3d8 | 15:17 |
| yadnesh|away | https://zuul.opendev.org/t/openstack/build/c099759921034494a44ab6a83e51fd6f | 15:17 |
| yadnesh|away | they all fail with "error: can't find Rust compiler" | 15:18 |
| fungi | yadnesh|away: you want the #openstack-qa channel, they're the ones who take care of openstack's test jobs and they're also aware of and working on a fix for that problem, so please confer with them | 15:24 |
| yadnesh|away | fungi, sure, thank you | 15:30 |
| fungi | looks like the plan i had a month ago was to have the deploy playbook check for the existence of a flagfile somewhere and if it's not present then set SITE_ID=1 temporarily in settings.py, do an initial db migration run, revert the SITE_ID override, and create the flagfile so that operation is skipped in the future | 15:33 |
| fungi | sans the flagfile part, the change right now is doing lineinfile to replace that setting temporarily before running migrations. ara report says the config was successfully updated and then the migrations break anyway: | 15:38 |
| fungi | https://174ab73dc2f37b91ee54-e84e491efa8a2348e120c66acb6b4680.ssl.cf1.rackcdn.com/867987/2/check/system-config-run-lists3/71cb4e5/bridge99.opendev.org/ara-report/playbooks/9.html?status=failed&status=unreachable#results | 15:38 |
| fungi | maybe i need to `manage.py makemigrations` like the output says | 15:40 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Use magic domain guessing in Mailman 3 https://review.opendev.org/c/opendev/system-config/+/867987 | 15:42 |
| fungi | if the in-test prototype works, i'll move it into the deployment playbook with the flagfile guard | 15:43 |
| *** marios is now known as marios|out | 16:53 | |
| clarkb | fungi: looks like it is stillfailing | 16:59 |
| fungi | yeah | 17:17 |
| fungi | different error this time though: | 17:21 |
| fungi | Error response from daemon: Container f8afa4f701856e154bda9b49d95484da5bc9253d01a23d6b5f99182b119ee7b2 is restarting, wait until the container is running | 17:21 |
| fungi | how do we normally tackle that? | 17:22 |
| fungi | or maybe the restarting state is indicative of a different problem | 17:22 |
| clarkb | I would look at the container logs for that container. It may be in a restart loop due to an error | 17:23 |
| fungi | yeah, still the same underlying error | 17:23 |
| fungi | https://zuul.opendev.org/t/openstack/build/610688f76b0c457b89fd3d02b1bdf68b/log/lists99.opendev.org/docker/mailman-web.txt#1708-1760 | 17:24 |
| fungi | ValueError: The database backend does not accept 0 as a value for AutoField. | 17:24 |
| fungi | maybe i need to restart the mailman-web container after editing settings.py | 17:24 |
| clarkb | is it possible that they've restricted the value of that column now and 0 is no longer valid? | 17:26 |
| fungi | well, it was valid when i tested on a held node, still the same versions of components at the moment | 17:32 |
| fungi | the upgrade change is completely separate from that series | 17:33 |
| clarkb | fungi: https://stackoverflow.com/questions/20328905/south-migration-database-backend-does-not-accept-0-as-a-value-for-autofield | 17:33 |
| clarkb | it does seem like this is a db constraint that has been around in django for a long time? | 17:34 |
| clarkb | I dunno why it would have worked in the past. maybe you didn't migrate afterwards so it couldn't complain or doing it by hand produces different behavior? | 17:34 |
| *** jpena is now known as jpena|off | 17:35 | |
| clarkb | fungi: is it failing on the intiail deployment or in the test playbook when you try to flip it back to 0? I wonder if it just isn't valid for the initial deployment and we have to flip it to 0 afterwards | 17:53 |
| clarkb | fungi: essentially do 1 ->0 rather than 0 -> 1 -> 0 | 17:53 |
| fungi | that's a good point, maybe i need to just go ahead and move the setup steps into the deployment playbook now because the container simply isn't starting initially and we're not catching that earlier | 17:57 |
| fungi | i don't think i need to invert the flag edits, just do them earlier in the job | 17:57 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Use magic domain guessing in Mailman 3 https://review.opendev.org/c/opendev/system-config/+/867987 | 20:19 |
| *** dviroel|ruck is now known as dviroel|ruck|afk | 20:53 | |
| *** dasm is now known as dasm|off | 21:35 | |
| fungi | ensure-pip: Install backport pip | 21:44 |
| fungi | failed to fetch PPA information, error was: HTTP Error 503: Service Unavailable | 21:44 |
| fungi | ouch | 21:44 |
| fungi | https://status.python.org/ mentions one of the fastly locations in europe is under maintenance | 21:45 |
| fungi | oh, that's xenial, we install it from ppa:openstack-ci-core/python-pip | 21:48 |
| clarkb | ya we do that because its 8.0 and we need at least 9.0 for sanity or something | 21:49 |
| fungi | the good news is system-config-run-lists3 passed! | 21:50 |
| clarkb | so 1 -> 0 did address it? | 21:50 |
| clarkb | the issue with the db at least | 21:50 |
| fungi | yeah, i mean we still need to create the additional site mappings, but it seems to be able to run | 21:51 |
| fungi | the screenshots of course still show lists.opendev.org in the corner of the lists.openstack.org postorius and hyperkitty pages | 21:52 |
| fungi | i'll get the dnm change rebased onto it and an autohold set | 21:53 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: DNM force mm3 failure to hold the node https://review.opendev.org/c/opendev/system-config/+/855292 | 21:54 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!