| *** mugsie has quit IRC | 02:01 | |
| tristanC | mordred: excellent :) | 07:08 |
|---|---|---|
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 07:43 |
| *** pcaruana has joined #zuul | 07:50 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 08:19 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 08:35 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 08:43 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 08:57 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 09:09 |
| *** jesusaur has quit IRC | 09:34 | |
| *** jesusaur has joined #zuul | 09:56 | |
| *** ssbarnea_ has quit IRC | 09:56 | |
| *** ssbarnea|bkp2 has joined #zuul | 09:56 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 10:44 |
| *** ssbarnea_ has joined #zuul | 10:53 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: WIP: Test using install-kubernetes role https://review.openstack.org/605803 | 10:55 |
| *** ssbarnea|bkp2 has quit IRC | 10:55 | |
| *** sshnaidm|off is now known as sshnaidm | 13:01 | |
| *** rlandy has joined #zuul | 13:04 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: docker-compose quickstart example https://review.openstack.org/608344 | 14:04 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Merger: automatically add new hosts to the known_hosts file https://review.openstack.org/608453 | 14:04 |
| corvus | mordred, fungi: ^ if you could take a quick look at 608453 that'd be swell. it's a one-liner that makes zuul a lot easier for first use. and, um, we've run into it a lot too. :) | 14:06 |
| corvus | mordred: 608344 has gotten to the point where i'm writing zuul config and complete sentences. still needs an apache container to serve logs and to use zuul-jobs, but it's getting close. | 14:07 |
| fungi | thanks for the heads-up! | 14:19 |
| fungi | corvus: comparing to the manpage for ssh_config(5) i'm not sure your assertion in the commit message is entirely true. have you tested it? | 14:25 |
| corvus | fungi: hrm, did i misread ssh_config? i'll test it now | 14:30 |
| *** rlandy has quit IRC | 14:33 | |
| corvus | fungi: hrm, i did -- if the key changes it just adds a new entry :/ | 14:34 |
| corvus | fungi: aha! latest ssh (in bionic) adds 'accept-new': http://manpages.ubuntu.com/manpages/bionic/en/man5/ssh_config.5.html | 14:36 |
| fungi | is that too new? | 14:37 |
| corvus | it seems very new yes | 14:37 |
| corvus | i feel like requiring it for zuul is maybe a bit premature | 14:37 |
| fungi | that's what i get for running sid everywhere, i guess ;) | 14:37 |
| corvus | it's probably not in centos7 | 14:37 |
| fungi | yeah, almost certainly not | 14:37 |
| fungi | maybe centos>7 (whatever they end up calling it) | 14:38 |
| fungi | i suppose it could be worked around with some fancy error handling logic, like trap for an error suggesting there's no known-hosts entry and only then proceed to use StrictHostKeyChecking=no or something | 14:42 |
| fungi | but that could be fragile if openssh doesn't use separate exit codes | 14:42 |
| corvus | on the scheduler, with paramiko, we have a 'missing host key policy' -- so if there is no host key for a host, we warn, but accept, but don't add them to the file. basically we make host keys optional. | 14:42 |
| corvus | there's also a 'autoaddpolicy', which would add new keys to the file (but still only on the first connection to a host) | 14:43 |
| corvus | maybe we could perform a connection with paramiko in the mergers to get that behavior before we call out to git | 14:43 |
| fungi | that sounds like it might do it | 14:43 |
| corvus | http://docs.paramiko.org/en/2.4/api/client.html#paramiko.client.AutoAddPolicy | 14:44 |
| corvus | i just looked up the history of our use of WarningPolicy -- it goes back to the initial commit. so it doesn't look like we've tried AutoAddPolicy before. | 14:47 |
| corvus | i left some notes on the change; i'll tackle that later, i'm going to sunday now | 14:47 |
| fungi | have a good sunday! sorry to rain on your ssh parade | 14:48 |
| fungi | heading out shortly to catch an eddie izzard performance in norfolk tonight | 14:49 |
| pabelanger | fungi: nice always liked the death star canteen bit from him | 15:36 |
| *** rlandy has joined #zuul | 17:03 | |
| *** rlandy has quit IRC | 17:21 | |
| fungi | yeah, he's a hoot | 17:21 |
| *** rlandy has joined #zuul | 22:46 | |
| *** ianw_pto has quit IRC | 22:54 | |
| *** ianw_ has joined #zuul | 22:54 | |
| *** ianw_ has left #zuul | 23:02 | |
| *** ianw has joined #zuul | 23:12 | |
| *** ianw is now known as ianw_pto | 23:15 | |
| *** ianw_pto is now known as ianw | 23:16 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul-jobs master: Create ansible-galaxy-import role https://review.openstack.org/608472 | 23:52 |
| pabelanger | mnaser: mordred: ^ | 23:52 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!