kevko | morning | 08:07 |
---|---|---|
kevko | anybody here ... I have a question :) | 08:07 |
mnasiadka | morning | 08:49 |
kevko | mnasiadka: replied to your sigterm patches | 08:50 |
kevko | or one of them | 08:50 |
mnasiadka | nice, thanks | 08:50 |
kevko | mnasiadka: i am really curious what do you think ... | 08:50 |
kevko | or anybody already working :P bbezak ? | 08:50 |
kevko | frickler ? | 08:50 |
mnasiadka | replied | 08:54 |
mnasiadka | I think dumb-init is not actively maintained, so another topic for PTG - but would be good to explore this before PTG to be able to decide on a direction | 08:54 |
kevko | mnasiadka: currently trying to test tozd/dinit | 08:55 |
kevko | mnasiadka: will inform | 08:55 |
kevko | from desc it looks nice | 08:55 |
kevko | mnasiadka: and these trivials ? mariadb check mode https://review.opendev.org/c/openstack/kolla-ansible/+/907971 and some bash kolla-ansible destroy handling inventory https://review.opendev.org/c/openstack/kolla-ansible/+/908429 | 08:56 |
opendevreview | Seunghun Lee proposed openstack/kayobe master: Make OVS container build regex independent to OVN https://review.opendev.org/c/openstack/kayobe/+/907721 | 09:05 |
opendevreview | Seunghun Lee proposed openstack/kayobe master: Make OVS container build regex independent to OVN https://review.opendev.org/c/openstack/kayobe/+/907721 | 09:06 |
kevko | mnasiadka: dinit -> nice -> https://paste.openstack.org/show/bU0cgHFT5iQr2WTIxscE/ | 10:11 |
kevko | mnasiadka: inside container https://paste.openstack.org/show/b1SgxJXP7lXHV1f3iW9e/ | 10:12 |
mnasiadka | jangutter: around? | 11:19 |
jangutter | I am! | 11:19 |
mnasiadka | jangutter: etcd seems to be having issues in cephadm jobs - https://86679a46b4c1d4c5128d-a5c518380d1db6458f4f9d5f7d87c4d0.ssl.cf5.rackcdn.com/908405/5/check/kolla-ansible-ubuntu-cephadm/74cdc72/secondary2/logs/kolla/cinder/cinder-volume.txt | 11:27 |
mnasiadka | it happens on Ubuntu surely | 11:27 |
mnasiadka | third time in a row in this patch | 11:27 |
mnasiadka | https://review.opendev.org/c/openstack/kolla-ansible/+/908405 | 11:27 |
mnasiadka | any ideas? | 11:28 |
mnasiadka | (I don't have good feelings towards etcd) :) | 11:28 |
opendevreview | Ivan Halomi proposed openstack/kolla-ansible master: Refactor of docker worker https://review.opendev.org/c/openstack/kolla-ansible/+/908295 | 11:29 |
jangutter | yeaaah, lemme check the logs. | 11:29 |
jangutter | mnasiadka, flakey or 100% consistent failures? | 11:31 |
jangutter | Etcd is complaining about slow disk: https://zuul.openstack.org/build/74cdc7264d384144b0f6d58653e183c7/log/primary/logs/container_logs/etcd.txt#114 | 11:33 |
jangutter | and a leader election happened at exactly the cinder traceback. | 11:34 |
jangutter | It says it lost a couple of messages during the election: https://zuul.openstack.org/build/74cdc7264d384144b0f6d58653e183c7/log/secondary1/logs/container_logs/etcd.txt#117 | 11:36 |
jangutter | One of the SRE's I work with says "etcd is quite sensitive to load"... I have a nasty feeling it's looking at the CAP theorem and deciding to give neither consistency, nor availability. | 11:42 |
mnasiadka | Flakey CI does not help, I think we were better with redis in those jobs | 11:44 |
mnasiadka | and yes, etcd always complains about slow disk :) | 11:44 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Switch cephadm jobs to redis https://review.opendev.org/c/openstack/kolla-ansible/+/908895 | 11:46 |
jangutter | mnasiadka: yeah, I agree, memory backed coordination makes a heck of a lot of more sense. | 11:51 |
jangutter | mnasiadka, paradoxically, I think, if it's single node (i.e. running etcd in "degraded" mode) these problems aren't evident. | 11:52 |
mnasiadka | true, let's see how this patch will change the flakiness | 11:52 |
opendevreview | Seunghun Lee proposed openstack/kayobe master: Make OVS container build regex independent to OVN https://review.opendev.org/c/openstack/kayobe/+/907721 | 11:53 |
opendevreview | Ivan Halomi proposed openstack/kolla-ansible master: Refactor of docker worker https://review.opendev.org/c/openstack/kolla-ansible/+/908295 | 12:02 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: cinder: Stop using admin service token https://review.opendev.org/c/openstack/kolla-ansible/+/908405 | 12:23 |
mnasiadka | for the multinode upgrade jobs - cinder is working on a fix here: https://review.opendev.org/c/openstack/cinder/+/908578 (with kolla-ansible DNM testing here https://review.opendev.org/c/openstack/kolla-ansible/+/908581) | 13:34 |
wncslln | morning kolla o/ anyone to help me? | 14:56 |
wncslln | anyone saw this error "nova | Ensure RabbitMQ users exist (5 retries left)." on rabbitmq reconfigure? | 14:56 |
wncslln | when i list vhosts and users inside rabbit container, nothing appears | 14:56 |
opendevreview | Michal Arbet proposed openstack/kolla master: Replace dumb-init with dinit https://review.opendev.org/c/openstack/kolla/+/908913 | 16:07 |
opendevreview | Michal Arbet proposed openstack/kolla master: Replace dumb-init with dinit https://review.opendev.org/c/openstack/kolla/+/908913 | 16:16 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Disable testing volume to image https://review.opendev.org/c/openstack/kolla-ansible/+/908914 | 16:30 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: Disable testing volume to image https://review.opendev.org/c/openstack/kolla-ansible/+/908914 | 16:30 |
opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: cinder: Stop using admin service token https://review.opendev.org/c/openstack/kolla-ansible/+/908405 | 16:38 |
mnasiadka | kevko: although still would be nice to merge those sigterm patches, I doubt we'll backport new init - and I don't know if it's going to fix all of our issues ;-) | 16:45 |
opendevreview | Michal Nasiadka proposed openstack/kolla master: WIP: Switch to Ubuntu 24.04 https://review.opendev.org/c/openstack/kolla/+/907589 | 16:49 |
opendevreview | Michal Arbet proposed openstack/kolla master: Replace dumb-init with dinit https://review.opendev.org/c/openstack/kolla/+/908913 | 16:56 |
opendevreview | Matúš Jenča proposed openstack/kolla-ansible master: Implement Redis as caching backend https://review.opendev.org/c/openstack/kolla-ansible/+/903978 | 17:05 |
kevko | mnasiadka: well, yeah, but let me try something before | 17:10 |
opendevreview | Michal Arbet proposed openstack/kolla master: Replace dumb-init with dinit https://review.opendev.org/c/openstack/kolla/+/908913 | 17:59 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Test build with dinit and fix some CI stuff https://review.opendev.org/c/openstack/kolla-ansible/+/908941 | 19:26 |
opendevreview | Pierre Riteau proposed openstack/kayobe master: Add missing space in help output https://review.opendev.org/c/openstack/kayobe/+/908942 | 19:27 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Test build with dinit and fix some CI stuff https://review.opendev.org/c/openstack/kolla-ansible/+/908941 | 19:51 |
opendevreview | Pierre Riteau proposed openstack/kayobe master: CI: support .gitreview of unmaintained branch https://review.opendev.org/c/openstack/kayobe/+/908950 | 22:08 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!