| opendevreview | Merged openstack/openstack-ansible-lxc_hosts stable/xena: Do not mess up with distro-provided machines mount https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/843169 | 00:49 |
|---|---|---|
| noonedeadpunk | would be great to get another vote on https://review.opendev.org/c/openstack/openstack-ansible/+/843600 | 05:21 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Drop nova-key from scale documentation and scripts https://review.opendev.org/c/openstack/openstack-ansible/+/843703 | 05:30 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Add release note about Ceph-Ansible deprecation https://review.opendev.org/c/openstack/openstack-ansible/+/843704 | 05:35 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Update RabbitMQ version to 3.10 https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/843705 | 05:45 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_octavia master: Use PKI role for certificate generation https://review.opendev.org/c/openstack/openstack-ansible-os_octavia/+/839068 | 05:58 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_octavia master: Move installation variables collection to beginning https://review.opendev.org/c/openstack/openstack-ansible-os_octavia/+/843706 | 06:05 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_octavia master: Use PKI role for certificate generation https://review.opendev.org/c/openstack/openstack-ansible-os_octavia/+/839068 | 06:05 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Add Octavia certificates upgrade playbook https://review.opendev.org/c/openstack/openstack-ansible/+/843711 | 07:02 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_octavia master: Use PKI role for certificate generation https://review.opendev.org/c/openstack/openstack-ansible-os_octavia/+/839068 | 07:04 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Add Octavia certificates upgrade playbook https://review.opendev.org/c/openstack/openstack-ansible/+/843711 | 07:31 |
| noonedeadpunk | jrosser_: have you seen https://bugs.launchpad.net/openstack-ansible/+bug/1975718 btw? | 07:33 |
| noonedeadpunk | seems we might have a problem with ssh keypairs | 07:34 |
| jrosser_ | do we test this in the infra jobs I wonder | 07:36 |
| jrosser_ | maybe for keystone role test we should ensure 3 keystone containers | 07:37 |
| noonedeadpunk | yeah, as we always have just single one | 07:47 |
| jrosser_ | argh lots of zuul logs are broken | 07:48 |
| jrosser_ | was just going to check if we have 'keystone' in SCENARION in a keystone job | 07:48 |
| noonedeadpunk | but I'm pretty sure we never have 3 keystone containers, except in functional jobs | 07:48 |
| jrosser_ | that would be an easy hook increasing the affinity to 3 | 07:48 |
| noonedeadpunk | every role is parsed and added to scenarios_extended | 07:48 |
| jrosser_ | right, but that always has keystone as we add it to everything? | 07:49 |
| noonedeadpunk | ah, yes... | 07:49 |
| noonedeadpunk | true | 07:49 |
| jrosser_ | i was thinking to make 3 keystones when we test os_keystone only | 07:49 |
| noonedeadpunk | mabe we should add this for infra scanrio indeed.... and replace keystone lxc jobs with infra ones... | 07:50 |
| noonedeadpunk | oh, but I think keystone should appear not in extended as well | 07:50 |
| jrosser_ | ok role name should be in all scenarios https://github.com/openstack/openstack-ansible/blob/master/zuul.d/playbooks/pre-gate-scenario.yml#L40 | 07:51 |
| jrosser_ | so we can check that rather than the expanded one | 07:51 |
| noonedeadpunk | yeah | 07:52 |
| noonedeadpunk | and place a comment for the future :D | 07:53 |
| jrosser_ | we don't define shared-infra_hosts very cleary | 07:57 |
| jrosser_ | clearly | 07:57 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Create three keystone containers when the test scenario includes 'keystone' https://review.opendev.org/c/openstack/openstack-ansible/+/843713 | 08:01 |
| jrosser_ | ^ bit of a guess | 08:02 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-os_keystone master: DNM - test multiple keystone instances for os_keystone tests https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/843714 | 08:03 |
| noonedeadpunk | I think it should work | 08:04 |
| noonedeadpunk | except we don't have lxc test for keystone | 08:05 |
| jrosser_ | doh, we kind of need to fix this generally for all roles becasue we do merge a few things that are ok for metal but not lxc | 08:05 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_keystone master: DNM - test multiple keystone instances for os_keystone tests https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/843714 | 08:05 |
| noonedeadpunk | well, I kind of didn't like switching all CI to metal back in stein, but it's really way faster.. | 08:06 |
| jrosser_ | i guess adding an infra scenario job (maybe just one), would also make 3 keystones with 843713 | 08:20 |
| jrosser_ | that might be a good compromise | 08:20 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_keystone master: DNM - test multiple keystone instances for os_keystone tests https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/843714 | 08:35 |
| noonedeadpunk | well, don't mind that either. but we also miss upgrade jobs there... | 08:35 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_keystone master: DNM - test multiple keystone instances for os_keystone tests https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/843714 | 08:36 |
| noonedeadpunk | but I can recall some cases when we were failing on glance during service_setup, as keystone was malfunctioning... But yeah, then tempest would fail... | 08:37 |
| jrosser_ | i'm out for the rest of the day now but i'll try to check in on this later | 08:38 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_keystone master: DNM - test multiple keystone instances for os_keystone tests https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/843714 | 08:39 |
| * noonedeadpunk has a flight in 4 hours as well | 08:39 | |
| noonedeadpunk | Hey, praveen77 sorry, I guess today most of ppl are away | 09:44 |
| damiandabrowski[m] | noonedeadpunk: just watched project upgrade, good job! | 12:23 |
| NeilHanlon | Agreed! great update, and I'm not only saying that because I got a shout out :P | 17:38 |
| jrosser_ | so multinode keystone fails on master because of this https://github.com/openstack/openstack-ansible-os_keystone/blob/054f7b545cb282c3ac5d07d65dfc2463f913ee96/tasks/keystone_fernet.yml#L20 | 18:06 |
| jrosser_ | the role runs serial (first, the rest), so on the first play hosts the ssh keypairs have not been created yet on the remaining nodes so the rsync fails | 18:06 |
| jrosser_ | the old code worked because everything happened on the first play host, like this https://github.com/openstack/openstack-ansible-os_keystone/blob/stable/xena/tasks/keystone_key_distribute.yml#L20-L22 | 18:09 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!