| openstackgerrit | Pedro Henrique Pereira Martins proposed openstack/kolla-ansible master: Add suport to OpenID Connect Authentication flow https://review.opendev.org/695432 | 00:22 |
|---|---|---|
| *** rphillips_ has joined #openstack-kolla | 00:40 | |
| *** rphillips_ is now known as rphillips | 00:40 | |
| rphillips | I have a train install, and I'm trying to add swift. The docker swift images are coming up but exiting with | 00:45 |
| rphillips | + sudo -E kolla_set_configs | 00:45 |
| rphillips | sudo: no tty present and no askpass program specified | 00:45 |
| *** igordc has quit IRC | 00:45 | |
| rphillips | anybody know what is up with it? | 00:45 |
| rphillips | the base install with nova is working fine. | 00:46 |
| *** benj_ has quit IRC | 01:56 | |
| *** benj_ has joined #openstack-kolla | 01:58 | |
| *** mrunge_ has joined #openstack-kolla | 02:03 | |
| *** mrunge has quit IRC | 02:05 | |
| openstackgerrit | Merged openstack/kolla-ansible stable/train: CI: Ignore more file paths when considering ansible runs https://review.opendev.org/695535 | 02:38 |
| openstackgerrit | Dincer Celik proposed openstack/kolla master: Ensure lsb-release exists on debian/ubuntu https://review.opendev.org/695635 | 04:18 |
| *** tonythomas has quit IRC | 04:44 | |
| *** msalo has joined #openstack-kolla | 04:51 | |
| *** msalo has quit IRC | 04:56 | |
| *** zhanglong has joined #openstack-kolla | 06:43 | |
| *** zhanglong has quit IRC | 07:14 | |
| *** zhanglong has joined #openstack-kolla | 07:16 | |
| *** msalo has joined #openstack-kolla | 08:11 | |
| *** msalo has quit IRC | 08:15 | |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: Enable repos only when needed https://review.opendev.org/695465 | 08:24 |
| mnasiadka | rphillips: please raise a bug - swift is not something we test in CI | 08:26 |
| *** zhanglong has quit IRC | 08:49 | |
| *** andrein has quit IRC | 09:04 | |
| *** andrein has joined #openstack-kolla | 09:04 | |
| openstackgerrit | Dincer Celik proposed openstack/kolla master: Ensure lsb-release exists on debian/ubuntu https://review.opendev.org/695635 | 09:15 |
| yoctozepto | mnasiadka: so you mean I should follow up with swift testing, probably worth to test ceph rgw as well (real functioning as replacement of swift) | 09:31 |
| mnasiadka | yoctozepto: don’t test what is deprecated :) | 09:32 |
| yoctozepto | mnasiadka: it is not, only internal is | 09:33 |
| yoctozepto | your integration must pass the tests. period. :-) | 09:33 |
| mnasiadka | yoctozepto: and I’m afraid swift role needs a really big ramp up - it’s not in the modern format :) | 09:33 |
| yoctozepto | yeah, it looks bad | 09:33 |
| mnasiadka | yoctozepto: well, we might skip ceph-ansible - last week Sage told me to go ceph orchestrator way | 09:33 |
| mnasiadka | ceph-ansible might be dead-ish in May | 09:34 |
| yoctozepto | mnasiadka: do whatever is best nowadays | 09:34 |
| mnasiadka | That’s the plan | 09:34 |
| yoctozepto | mnasiadka: last time I looked orchestrator used ceph-ansible | 09:34 |
| yoctozepto | so not sure what that means in fact | 09:35 |
| mnasiadka | yoctozepto: yeah, in Nautilus | 09:35 |
| hrw | hey folks | 09:35 |
| mnasiadka | yoctozepto: but in Octopus not really - they use direct ssh and run python code | 09:35 |
| yoctozepto | ah, not followed then | 09:35 |
| yoctozepto | if it works fine... | 09:35 |
| mnasiadka | yoctozepto: so I prefer to do something that is easily upgradable... | 09:35 |
| yoctozepto | yeah, exactly | 09:36 |
| yoctozepto | do they use containers? | 09:36 |
| yoctozepto | docker/podman? | 09:36 |
| yoctozepto | hi hrw | 09:36 |
| hrw | does someone know how to get to jinja2 template vars from python function registered as jinja2 method? | 09:36 |
| *** egonzalez has quit IRC | 09:36 | |
| yoctozepto | hrw: care to give some example? not sure if it is too early but not following | 09:37 |
| hrw | yoctozepto: https://review.opendev.org/695465 - enable_repos() method | 09:37 |
| patchbot | patch 695465 - kolla - WIP: Enable repos only when needed - 5 patch sets | 09:37 |
| yoctozepto | hrw: geez, which part? | 09:37 |
| hrw | yoctozepto: now it gets 4 vars: repos + 3 template vars | 09:37 |
| hrw | yoctozepto: methods.py | 09:37 |
| yoctozepto | hrw: ack | 09:37 |
| hrw | I need 4th var now. | 09:38 |
| *** egonzalez has joined #openstack-kolla | 09:38 | |
| hrw | may need another one | 09:38 |
| yoctozepto | ah, you mean you don't want to pass them around | 09:38 |
| hrw | yep | 09:38 |
| yoctozepto | you can use context | 09:38 |
| hrw | so access to template vars or kolla conf is needed | 09:38 |
| hrw | yoctozepto: tell me more | 09:38 |
| yoctozepto | jinja2 context decorator | 09:38 |
| yoctozepto | look at my ipv6 change | 09:38 |
| yoctozepto | https://review.opendev.org/#/c/681573/70/kolla_ansible/kolla_address.py | 09:40 |
| patchbot | patch 681573 - kolla-ansible - Implement IPv6 support in the control plane (MERGED) - 70 patch sets | 09:40 |
| yoctozepto | @contextfilter | 09:40 |
| yoctozepto | def kolla_address(context | 09:40 |
| yoctozepto | and the drama in function body | 09:40 |
| yoctozepto | hrw: ^ | 09:40 |
| hrw | yoctozepto: thanks! | 09:41 |
| yoctozepto | hrw: yw! | 09:41 |
| hrw | so all templates vars goes from context? | 09:42 |
| yoctozepto | hrw: as much as you want them | 09:47 |
| yoctozepto | if they are all shared, you can go all context | 09:47 |
| yoctozepto | for the most variable keep it in the function | 09:47 |
| yoctozepto | all context is probably rarely useful | 09:47 |
| yoctozepto | I did not analyze your use case though | 09:47 |
| hrw | yoctozepto: I have a macro which enables repos. you do call from Dockerfile.j2: {{ macros.enable_extra_repos(['ceph', 'grafana']) }} and it takes care of distro/arch etc | 09:51 |
| hrw | yoctozepto: but I need rpm/deb, yum/dnf, debian/ubuntu, arch... | 09:51 |
| yoctozepto | so it makes sense to contextify all except repo names | 09:51 |
| hrw | yoctozepto: sure, macro can give all vars but there are more and more... | 09:51 |
| hrw | yes | 09:51 |
| yoctozepto | just what I mean with variability | 09:52 |
| yoctozepto | all right, you are all set | 09:52 |
| hrw | @contextfunction | 09:54 |
| hrw | and my patch requires python 3.6+ :D | 09:55 |
| yoctozepto | what's the minimum to support in Ussuri? not 3.6? ;p | 09:56 |
| hrw | yep | 09:56 |
| yoctozepto | then np, go ahead | 09:56 |
| hrw | commands += f' --enable {repo_list[repo]}' | 09:56 |
| hrw | clean | 09:57 |
| hrw | + commands += f'echo "{repo_list[repo]}" ' | 09:57 |
| hrw | + commands += f'>/etc/apt/sources.list.d/{repo}.list; ' | 09:57 |
| hrw | and if you use Debian (or derived): "apt-mark hold python2" after uninstalling python2 package will mark it as not installable | 09:58 |
| hrw | so unless you install exactly python2 package then it will not get installed (and anything which depend on it) | 09:58 |
| hrw | https://review.opendev.org/#/c/695465/ goes nice on x86 :D | 09:59 |
| patchbot | patch 695465 - kolla - WIP: Enable repos only when needed - 5 patch sets | 09:59 |
| hrw | on monday will merge @contextfunction stuff with !x86 stuff and then it should be ready for detailing | 10:00 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: Enable repos only when needed https://review.opendev.org/695465 | 10:43 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: Enable repos only when needed https://review.opendev.org/695465 | 11:23 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: Enable repos only when needed https://review.opendev.org/695465 | 11:55 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: test-requirements: bump pyflakes to get f-strings support https://review.opendev.org/695796 | 11:57 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: WIP: Enable repos only when needed https://review.opendev.org/695465 | 11:58 |
| *** spiette has quit IRC | 14:43 | |
| *** mchlumsky has quit IRC | 14:45 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI/Debian: Push images after building https://review.opendev.org/694243 | 15:05 |
| *** luksky has joined #openstack-kolla | 15:06 | |
| yoctozepto | mnasiadka: google does not help me with this new ceph orchestrator - is there any worthy source of information on it? I wanted to know mostly about "why" ceph-ansible is passe | 15:46 |
| mnasiadka | yoctozepto: that is not said in any materials - I’ll ask the ceph-ansible guys on Monday and let you know | 16:03 |
| yoctozepto | mnasiadka: thanks :-) | 16:04 |
| *** msalo has joined #openstack-kolla | 16:12 | |
| *** pcaruana has quit IRC | 16:31 | |
| *** luksky has quit IRC | 16:44 | |
| *** msalo has quit IRC | 16:50 | |
| *** msalo_ has joined #openstack-kolla | 16:50 | |
| openstackgerrit | Merged openstack/kolla-ansible stable/stein: Patch to fix RPC selection problem after upgrade https://review.opendev.org/695049 | 17:01 |
| openstackgerrit | Merged openstack/kolla-ansible master: Fixes the compatibility issue at HAProxy role with Ansible 2.9 https://review.opendev.org/694058 | 17:02 |
| *** msalo_ has quit IRC | 17:06 | |
| *** msalo has joined #openstack-kolla | 17:06 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Cleanups, refactors and testing of Qinling https://review.opendev.org/695192 | 17:08 |
| *** msalo has quit IRC | 17:10 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Cleanups, refactors and testing of Qinling https://review.opendev.org/695192 | 17:11 |
| *** msalo has joined #openstack-kolla | 17:11 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Cleanups, refactors and testing of Qinling https://review.opendev.org/695192 | 17:17 |
| *** spiette has joined #openstack-kolla | 17:21 | |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Cleanups, refactors and testing of Qinling https://review.opendev.org/695192 | 17:22 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Cleanups, refactors and testing of Qinling https://review.opendev.org/695192 | 17:27 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Cleanups, refactors and testing of Qinling https://review.opendev.org/695192 | 17:28 |
| *** dr_feelgood has quit IRC | 17:56 | |
| *** msalo has quit IRC | 18:00 | |
| *** msalo has joined #openstack-kolla | 18:20 | |
| *** msalo has quit IRC | 18:25 | |
| *** numans has quit IRC | 20:17 | |
| *** mchlumsky has joined #openstack-kolla | 20:20 | |
| yoctozepto | mgoddard, mnasiadka: tacker drama continued: https://bugs.launchpad.net/tacker/+bug/1853715 | 20:26 |
| openstack | Launchpad bug 1853715 in kolla-ansible ussuri "[train] [kolla-ansible] [tacker] Incomplete config results in ERRORs from tacker-server" [High,Triaged] - Assigned to Radosław Piliszek (yoctozepto) | 20:26 |
| yoctozepto | this time I went all into their source code | 20:27 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: CI: Cleanups, refactors and testing of Qinling https://review.opendev.org/695192 | 20:30 |
| openstackgerrit | Radosław Piliszek proposed openstack/kolla-ansible master: [Train] Fix Tacker config for the local Glance store https://review.opendev.org/695804 | 20:48 |
| *** goldyfruit_ has quit IRC | 20:50 | |
| *** msalo has joined #openstack-kolla | 21:48 | |
| *** goldyfruit_ has joined #openstack-kolla | 23:12 | |
| *** msalo has quit IRC | 23:14 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!