| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_skyline master: Do not define a random password for each run https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/912332 | 08:43 |
|---|---|---|
| jrosser | i think we might have circular dependancies for skyline | 08:49 |
| jrosser | linters fail for this https://review.opendev.org/c/openstack/openstack-ansible/+/859446 | 08:49 |
| jrosser | which appears to be fixed by https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/912370/6 | 08:49 |
| jrosser | but that depends-on the first one | 08:49 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Use container setup role from plugins repo https://review.opendev.org/c/openstack/openstack-ansible/+/905004 | 09:56 |
| noonedeadpunk | jrosser: yeah, you're right I guess.... though I hoped they were fixed by 912333 | 09:59 |
| jrosser | it might be the difference between 9 and "9" in the meta/main.yml | 10:00 |
| jrosser | but it is not good how lint passes on 912333 but fails on 859446 | 10:02 |
| noonedeadpunk | yeah, though I hoped that meta is not verified in integrted repo | 10:03 |
| noonedeadpunk | we can actually make another patch I guess to cover this meta thing | 10:04 |
| jrosser | oh you mean just merge that first into the os_skyline? | 10:04 |
| noonedeadpunk | yeah | 10:06 |
| jrosser | i can make a patch for that | 10:06 |
| noonedeadpunk | jrosser: do you remember how you workarounded unsafe condition previously for python_venv_build? | 10:11 |
| noonedeadpunk | as seems we have common thing here https://zuul.opendev.org/t/openstack/build/d2c51c94f37e48419a73143266eb699d/log/job-output.txt#9486-9490 | 10:11 |
| jrosser | yeah i added `use: "{{ ansible_facts['pkg_mgr'] }}"` to the `package` module | 10:12 |
| jrosser | which makes it then not try to template that ^^ inside the action plugin | 10:13 |
| noonedeadpunk | ok, it's different then I assume | 10:15 |
| jrosser | its related but yes different | 10:15 |
| jrosser | looks lke that totally deserves a new bug | 10:15 |
| jrosser | as it's a different side-effect of the same change to the types they made | 10:16 |
| jrosser | oh well perhaps its not, as it says `[WARNING]: conditional statements should not include jinja2 templating` | 10:17 |
| jrosser | right before it fails | 10:17 |
| noonedeadpunk | yeah, true, though I would assume that it could be fine for assert... anyway will check | 10:19 |
| jrosser | how did this not fail before | 10:20 |
| jrosser | oh that job is maybe conditional | 10:20 |
| noonedeadpunk | yeah | 10:21 |
| noonedeadpunk | will check how to transform | 10:21 |
| jrosser | i think i found it pretty hard to escape from having an AnsibleUnsafeText | 10:25 |
| jrosser | and the | type_debug thing was pretty handy when trying to work out what was happening | 10:26 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-os_skyline master: Add quotes for EL version in meta/main.yml https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/914439 | 10:30 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: [Feature] Add skyline deployment capability https://review.opendev.org/c/openstack/openstack-ansible/+/859446 | 10:30 |
| jrosser | urgh `[MIRROR] systemd-devel-252-32.el9.x86_64.rpm: Status code: 404 for https://mirror-int.ord.rax.opendev.org/centos-stream/9-stream/AppStream/x86_64/os/Packages/systemd-devel-252-32.el9.x86_64.rpm (IP: 10.209.96.36)` | 10:32 |
| jrosser | and more `fatal: [localhost]: FAILED! => {"attempts": 5, "changed": false, "msg": "Failed to download packages: pcp-6.2.0-2.el9.x86_64: Cannot download, all mirrors were already tried without success", "results": []}` | 10:34 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Remove Jinja from conditions https://review.opendev.org/c/openstack/openstack-ansible/+/914461 | 10:42 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Remove Jinja from conditions https://review.opendev.org/c/openstack/openstack-ansible/+/914461 | 10:42 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: [Feature] Add skyline deployment capability https://review.opendev.org/c/openstack/openstack-ansible/+/859446 | 10:42 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Remove Jinja from conditions https://review.opendev.org/c/openstack/openstack-ansible/+/914461 | 10:42 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: [Feature] Add skyline deployment capability https://review.opendev.org/c/openstack/openstack-ansible/+/859446 | 10:42 |
| noonedeadpunk | ugh | 10:43 |
| noonedeadpunk | ok, but now that should pass, I assume | 10:44 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_skyline master: Add EL distro support https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/912370 | 10:44 |
| jrosser | i hope so | 10:44 |
| opendevreview | James Denton proposed openstack/openstack-ansible-os_skyline master: Support large uploads via Skyline https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/914149 | 10:44 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_skyline master: Install skyline-console through yarn https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/914405 | 10:44 |
| gokhan | hello folks, after upgrade to antelope, live migration is not working. I can migrate but live migrate is not | 10:44 |
| gokhan | when I check with nova user on compute hosts, I can not ssh between compute hosts. It waits when trying to ssh | 10:45 |
| noonedeadpunk | iirc, what changed in terms of live migration, is that libvirt should be service tls since antelope | 10:45 |
| noonedeadpunk | and also ports were changed iirc | 10:46 |
| noonedeadpunk | regarding SSH - it's done with SSH certs right now, but SSH is used only for offline migration | 10:46 |
| gokhan | my env is without ssl :( | 10:46 |
| noonedeadpunk | IIRC tunnel migrations were deprecated | 10:47 |
| jrosser | ssl between haproxy and the api is not the same as live migtation over tls between libvirt<>libvirt | 10:47 |
| noonedeadpunk | https://docs.openstack.org/nova/latest/configuration/config.html#libvirt.live_migration_tunnelled | 10:47 |
| noonedeadpunk | so, we have that by default: https://opendev.org/openstack/openstack-ansible-os_nova/src/branch/master/defaults/main.yml#L454-L455 | 10:48 |
| noonedeadpunk | but we kept old logic there actually: https://opendev.org/openstack/openstack-ansible-os_nova/src/branch/master/templates/nova.conf.j2#L271-L277 | 10:49 |
| noonedeadpunk | though I guess it's broken? | 10:49 |
| noonedeadpunk | as keyfile is probably wrong? | 10:49 |
| noonedeadpunk | huh, let me check quickly | 10:49 |
| gokhan | ok thanks jrosser fyi | 10:49 |
| gokhan | noonedeadpunk, my libvirt config is in nova https://paste.openstack.org/show/bPYwsAhtOFxZUoJbxhYO/ | 10:50 |
| noonedeadpunk | yeah, ok, so you should be running live migrations through TLS | 10:51 |
| noonedeadpunk | if you're using firewall, just ensure that required ports are open, as they're different | 10:53 |
| noonedeadpunk | so it should be 16514 and 49152:49215 for management network | 10:54 |
| noonedeadpunk | I again forgot how ssh certs are working /o\ | 10:58 |
| noonedeadpunk | ssh 10.X.X.X -l nova -i /var/lib/nova/.ssh/id_rsa throws Permission denied (publickey) | 10:58 |
| gokhan | nova conductor throws no valid host found error https://paste.openstack.org/show/broAn9S8ZiszrJVzenXc/ | 10:59 |
| noonedeadpunk | ok, well, that is different | 10:59 |
| gokhan | all compute service is up | 11:00 |
| noonedeadpunk | anything in scheduler logs? | 11:02 |
| noonedeadpunk | like some "info" of why | 11:02 |
| gokhan | noonedeadpunk, Mar 27 10:35:37 dev-infra3-nova-api-container-d4035159 nova-scheduler[65766]: 2024-03-27 10:35:37.376 65766 INFO nova.scheduler.host_manager [req-15e24c17-4b77-4911-a4a5-f99315dcc173 req-e06bd0de-9490-44ec-8c9a-ca75321d8744 4841276dcdbe4ab096ef60b1744c4fa9 f2e52a5c5d1c4ca1b51274619b517e0e - - default default] Host filter only checking host dev-compute1 and node dev-compute1 | 11:04 |
| gokhan | Mar 27 10:35:37 dev-infra3-nova-api-container-d4035159 nova-scheduler[65766]: 2024-03-27 10:35:37.377 65766 INFO nova.scheduler.host_manager [req-15e24c17-4b77-4911-a4a5-f99315dcc173 req-e06bd0de-9490-44ec-8c9a-ca75321d8744 4841276dcdbe4ab096ef60b1744c4fa9 f2e52a5c5d1c4ca1b51274619b517e0e - - default default] Host filter ignoring hosts: dev-compute1 | 11:04 |
| noonedeadpunk | Well, I guess it's worth checking what is special about the instance or host - maybe some aggregates, AZs, server groups... | 11:07 |
| opendevreview | Merged openstack/openstack-ansible-os_nova stable/2023.2: Ensure nova_device_spec is templated as JSON string https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/913501 | 11:17 |
| gokhan | noonedeadpunk, https://paste.openstack.org/show/bBgug7t7cvPqKnHwoepK/ | 11:17 |
| noonedeadpunk | ok, it was my internal issue not being able to login via ssh :) | 11:18 |
| noonedeadpunk | (as nova user) | 11:18 |
| gokhan | noonedeadpunk, I can't ssh to compute nodes with command which you sent | 11:19 |
| noonedeadpunk | actually this reminds me of some bug | 11:19 |
| noonedeadpunk | gokhan: SSH is used only for offline migration with the config you have | 11:21 |
| noonedeadpunk | for online migration it is not needed at all | 11:21 |
| noonedeadpunk | and you obviously have some scheduling issue rather then anything else | 11:21 |
| gokhan | noonedeadpunk, ok so this is different. | 11:22 |
| gokhan | noonedeadpunk, I am using default scheduler filter in osa | 11:23 |
| noonedeadpunk | I guess the question here is, why `Host filter ignoring hosts: dev-compute1` | 11:24 |
| noonedeadpunk | as that's the only candidate | 11:24 |
| noonedeadpunk | and it could be quite some reasons actually | 11:25 |
| gokhan | in nova config resizeonsamehost is true | 11:25 |
| jrosser | noonedeadpunk: i just also checked the nova<>nova ssh and it works for me here | 11:25 |
| noonedeadpunk | yeah, just our internal mess dropped Include /etc/ssh/sshd_config.d/*.conf out of sshd_config | 11:26 |
| gokhan | live migration is not working in all my antelope envs | 11:28 |
| gokhan | filters are there https://paste.openstack.org/show/bSnInnwo29kLjTB79oc0/ | 11:29 |
| jrosser | gokhan: i think you need to do debugging to find the root cause in your specific environment | 11:31 |
| jrosser | it's not possible to understand just from the config what is happening | 11:31 |
| gokhan | jrosser, I think it is about scheduling. I debugged nova scheduler but I didn't find root cause. In scheduler logs it lastly says There are 0 hosts available but 1 instances requested to build. these are scheduler logs https://paste.openstack.org/show/b1CktfA1CRBk95ZR6KYS/ | 11:43 |
| noonedeadpunk | yeah, so `Host filter ignoring hosts: dev-compute1` | 11:46 |
| noonedeadpunk | so it's a specific filter that filtered out it | 11:46 |
| opendevreview | Merged openstack/openstack-ansible-os_horizon stable/2023.2: Do not change mode of files recursively https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/914015 | 11:47 |
| noonedeadpunk | question - do you provide destination for the vm explicitly? | 11:47 |
| gokhan | noonedeadpunk, yes I am giving destination explicitly | 11:47 |
| noonedeadpunk | and what if you don't?:) | 11:48 |
| noonedeadpunk | same - no hosts available? | 11:48 |
| jrosser | you can look in the nova compute log on the dev-compute1 to see if the migration was rejected there | 11:49 |
| jrosser | and you can also look in placement logs to see if the resource claim is rejected there | 11:50 |
| gokhan | noonedeadpunk, yes similar error :( Host filter ignoring hosts: dev-compute3, dev-compute1, dev-compute2 | 11:50 |
| jrosser | gokhan: are you migrating between antelope hosts? or is this migration part of your upgrade to clean out / upgrade compute nodes? | 11:51 |
| noonedeadpunk | well, from what I do see in nova code - there should be ignored host somehow explicitly to get filtered in a way it does https://opendev.org/openstack/nova/src/branch/master/nova/scheduler/host_manager.py#L595-L598 | 11:51 |
| noonedeadpunk | I do recall one possible bug in nova, but it was sorted quite some time ago | 11:51 |
| jrosser | there is this https://bugs.launchpad.net/nova/+bug/2023035 | 11:52 |
| noonedeadpunk | nah, this one is kernel issue iirc | 11:52 |
| noonedeadpunk | it's like specific set of cpu and kernel - at least that's what we had | 11:53 |
| noonedeadpunk | sec | 11:53 |
| noonedeadpunk | will try to find what I'm talking about | 11:53 |
| gokhan | jrosser, yes now I checked nova-compute and it throws this bug. it is about cpu doesn2t have compability | 11:54 |
| gokhan | jrosser, yes this is migration part of upgrade to clean out / upgrade compute nodes | 11:55 |
| noonedeadpunk | jrosser: so, there's regression between kernel 3.14 and 3.17 where with Intel Gold kernel announces an extra cpu flag regardless of what's requested or present in cpu_map | 11:56 |
| noonedeadpunk | so migrating vms from some old E5 to Gold is one way ticket, unless you have hwe kernel | 11:56 |
| jrosser | ahhhh ok we would not have come across that | 11:57 |
| jrosser | and on focal i think we were running HWE | 11:57 |
| jrosser | for antelope we needed to patch nova for migrations to work https://github.com/bbc/nova/commits/bbc-antelope-27.1.0/ | 11:58 |
| gokhan | we are running also hwe on focal | 11:58 |
| jrosser | gokhan: it is pretty much normal for us to test this completely in a lab and the result of that is patched versions of a few services | 11:59 |
| jrosser | i think it's been a while since a completely stock install worked for us | 12:00 |
| noonedeadpunk | huh, it does for us mostly... | 12:01 |
| gokhan | jrosser, so I need this commit https://github.com/bbc/nova/commit/159869cde16fbd3e780a2a5bfa59e999890e6511 | 12:02 |
| jrosser | nova/neutron/magnum/keystone we need to fork currently | 12:03 |
| jrosser | gokhan: perhaps - this is not anything i guarantee for you | 12:03 |
| jrosser | thats based on our tests in our lab] | 12:03 |
| gokhan | jrosser, I will test it now in test env | 12:04 |
| jrosser | we discussed this with the nova team some time ago in their weekly meeting, and there suggestion for a quick-fix was to make that reverty | 12:05 |
| jrosser | *revert | 12:05 |
| jrosser | where "we" is my team here | 12:05 |
| jrosser | not OSA | 12:06 |
| gokhan | jrosser, it worked and now I can migrate thanks. | 12:09 |
| jrosser | it is probably worth replying to the bug to say that you are also affected | 12:11 |
| gokhan | jrosser, yes I am replying now. | 12:12 |
| gokhan | thanks jrosser noonedeadpunk for your help :) | 12:19 |
| gokhan | jrosser, rebuilf rom volume based image is also not working. It seem you also fixed this problem. https://github.com/bbc/nova/commit/0411adc64c96fe2f83ef7cf1584e2b1ee5b602d3 | 12:26 |
| jrosser | gokhan: ah well yes but i think you really have noonedeadpunk to thank there https://bugs.launchpad.net/nova/+bug/2040264 | 12:28 |
| jrosser | and that looks to be fixed in nova 2023.1 anyway now, so would be good idea to see why you don't have that already | 12:30 |
| gokhan | jrosser, I am getting again Image 4ef6efed-3ee1-4360-b366-3c6d69eedc09 is unacceptable: Unable to rebuild with a different image for a volume-backed server. | 12:34 |
| gokhan | I think this feature is added in zed but now it is not working | 12:34 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Use container setup role from plugins repo https://review.opendev.org/c/openstack/openstack-ansible/+/905004 | 13:01 |
| nixbuilder | Anyone run across this after installing with 28.0.1??? https://paste.openstack.org/show/bAgJFWLfCr6VTIdTkNEK/ | 13:03 |
| nixbuilder | It's rabbitmq errors :-( | 13:04 |
| jrosser | ideally you'd be using 28.1.0 | 13:06 |
| nixbuilder | jrosser: OK... will try that! | 13:07 |
| jrosser | though i don't specifically expect a magic fix there, just that it's the latest tagged release | 13:07 |
| noonedeadpunk | WRONG_VERSION_NUMBER? | 13:08 |
| noonedeadpunk | it feels that it tries to reach rabbit on tls port without tls or smth like that | 13:09 |
| noonedeadpunk | So yeah. I wouldn't put too much on 28.1.0 in this case | 13:09 |
| jrosser | so in my rabbitmq.conf i have management.ssl.versions.1 = tlsv1.2 | 13:10 |
| jrosser | and then in my nova.conf the rabbitmq connection string ends `@10.48.240.167:5671//nova?ssl=1&ssl_version=TLSv1_2&ssl_ca_file=` | 13:11 |
| jrosser | nixbuilder: ^ so you can see there that ssl is enabled and the tls version is matching at both ends | 13:12 |
| jrosser | thats probably a good place to start debugging whats happening | 13:12 |
| noonedeadpunk | jrosser: finally found bug I was talking about /o\ https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2036675 | 13:12 |
| nixbuilder | jrosser: Thanks for the tip.. I am re-installing with 28.1.0 but I imagine I will have to debug my issue after my latest system comes back up! | 13:14 |
| noonedeadpunk | it was time when I spent like 2 days trying to build a kernel patch to live-apply through kpatch, but failed in building even kernel itself lol | 13:15 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_skyline master: Define amount of gunicorn workers through config https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/914516 | 13:36 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: [Feature] Add skyline deployment capability https://review.opendev.org/c/openstack/openstack-ansible/+/859446 | 13:37 |
| noonedeadpunk | ok, this one passed ^ but I had to update it a bit as yarn jobs were in retry_limit | 13:37 |
| noonedeadpunk | and this one also looks like good to go https://review.opendev.org/c/openstack/openstack-ansible-os_glance/+/914275 | 13:40 |
| jrosser | relation chain on some of the skyline patches is pretty messed up | 13:40 |
| jrosser | i dare not try to rebase that | 13:40 |
| noonedeadpunk | haha | 13:40 |
| jrosser | this in particular https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/912332 | 13:42 |
| jrosser | i'm always wary when the relation chain on the patch is pretty different to how it looks on other patches | 13:42 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_skyline master: Do not define a random password for each run https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/912332 | 13:43 |
| noonedeadpunk | there're just couple of changes on top of https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/912370/7 as the one which should run at least playbooks towards skyline role | 13:44 |
| noonedeadpunk | still not good enough testing, but at least role gets executed | 13:44 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-openstack_hosts master: Place sysctl options to it's own file https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/908912 | 13:49 |
| opendevreview | Merged openstack/openstack-ansible-openstack_hosts unmaintained/wallaby: Update .gitreview for unmaintained/wallaby https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/913087 | 13:59 |
| noonedeadpunk | I;m looking through apt_key patches and see how ppl just rush to irc/launchpad filling bugs about failed deployments... | 14:00 |
| noonedeadpunk | while indeed that might make most sense right now | 14:01 |
| jrosser | becasue the data structure for confgiruing custom repos pretty much changes? | 14:02 |
| opendevreview | Merged openstack/openstack-ansible-openstack_hosts unmaintained/xena: Update .gitreview for unmaintained/xena https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/913144 | 14:02 |
| noonedeadpunk | yeah | 14:03 |
| noonedeadpunk | and not always obvious | 14:03 |
| jrosser | no, though the new module is pretty neat and there is an opportunity to do a bunch of tidy up | 14:03 |
| jrosser | it is good to get feedback / review though, and i also think that people will trip over this | 14:03 |
| noonedeadpunk | I was thinking if we can get some pre-upgrade check or smth to fail early | 14:04 |
| noonedeadpunk | or dunno | 14:04 |
| jrosser | we could do that if we changed the name of the var perhaps | 14:04 |
| jrosser | then assert old var is not defined | 14:04 |
| noonedeadpunk | well, I tried to do some wiring, but then realized it's also so different for 3 repos in topic | 14:05 |
| noonedeadpunk | and vars are same for centos/ubuntu which is another complication | 14:06 |
| noonedeadpunk | we could probably to check for some required keys in the list | 14:07 |
| jrosser | well perhaps this is then "all or nothing" | 14:07 |
| noonedeadpunk | but I' not sure what is required except name | 14:07 |
| jrosser | that we need a big transition to the new module, everywhere | 14:07 |
| jrosser | i realised pretty early that there was no way to retrofit the old data structure to the new module | 14:08 |
| jrosser | but the new module is so flexible that pretty much any possibility is covered if we just expose all the module params and put in the right places for overrides | 14:09 |
| jrosser | it would make it possible to no longer need to pre-stage apt keys for internal repos for example, they could be in data | 14:10 |
| noonedeadpunk | yeah, I know and kinda agree on that... but also I'm quite sure that at least 2 times I will be pinged about that itnerally after upgrade failures | 14:28 |
| jrosser | i also didnt make a releasenote for this as it felt like one overall one was needed | 14:30 |
| jrosser | ^ in the individual patches | 14:30 |
| noonedeadpunk | Ithink you did? | 14:30 |
| jrosser | oh did i?! long ago :_ | 14:30 |
| noonedeadpunk | ie https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/907833/4/releasenotes/notes/rabbitmq_repo_deb822-b47ef07ff462193f.yaml | 14:30 |
| noonedeadpunk | or well, not everywhere, but somewhere | 14:30 |
| noonedeadpunk | and https://review.opendev.org/c/openstack/openstack-ansible-galera_server/+/907752/5/releasenotes/notes/galera_repo_deb822-fc1aa6b88ee33b57.yaml | 14:31 |
| noonedeadpunk | ok, so only openstack_hosts is not covered :D | 14:31 |
| jrosser | hah | 14:31 |
| noonedeadpunk | and debian bullseye seems not liking that at all | 15:03 |
| noonedeadpunk | https://zuul.opendev.org/t/openstack/build/86b88fc0c94146a0b6d5868d53b00840/log/job-output.txt#11505 | 15:04 |
| noonedeadpunk | same for mariadb | 15:04 |
| noonedeadpunk | sounds like bullseye expects smth different for signed-by then should be | 15:08 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-galera_server master: Implement installation method selection for MariaDB role https://review.opendev.org/c/openstack/openstack-ansible-galera_server/+/914530 | 15:36 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-galera_server master: Implement installation method selection for MariaDB role https://review.opendev.org/c/openstack/openstack-ansible-galera_server/+/914530 | 15:37 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_skyline master: Install skyline-console through yarn https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/914405 | 17:10 |
| noonedeadpunk | so this looks quite good now from what I can tell https://review.opendev.org/c/openstack/openstack-ansible/+/859446 | 17:11 |
| noonedeadpunk | the biggest question if 914405 pass :D | 17:12 |
| jrosser | this is really cool stuff | 17:20 |
| jrosser | damiandabrowski: mgariepy NeilHanlon is anyone around to look at https://review.opendev.org/c/openstack/openstack-ansible/+/914461 ? | 17:22 |
| * NeilHanlon is around | 17:23 | |
| jrosser | cool, thanks! | 17:24 |
| NeilHanlon | of course :) | 17:25 |
| NeilHanlon | backport candidate? | 17:25 |
| noonedeadpunk | nah | 17:26 |
| noonedeadpunk | or well... dunno | 17:26 |
| noonedeadpunk | that's good question actually | 17:26 |
| noonedeadpunk | as we probably should bump ansible version on stable braches to cover cve | 17:26 |
| NeilHanlon | mmm | 17:28 |
| NeilHanlon | i mean, probably not incredibly urgent | 17:28 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron master: Fix multiline yaml formatting in neutron systemd services https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/914544 | 17:32 |
| * jrosser throws rocks at ansible lint :/ | 17:32 | |
| noonedeadpunk | well, we can at least on 2023.2 | 17:33 |
| jrosser | 914544 can be checked with something like this 914544 | 17:33 |
| jrosser | argh | 17:33 |
| jrosser | https://paste.opendev.org/show/b4rOfwsuGBVJsLwLr4F7/ | 17:33 |
| jrosser | noonedeadpunk: out of interest, where does the actual skyline-console wheel get built? | 18:19 |
| jrosser | oh i see it gives skyline_console_yarn_build_path as the source location for skyline-console to python_venv_build role | 18:23 |
| opendevreview | Merged openstack/openstack-ansible master: Remove Jinja from conditions https://review.opendev.org/c/openstack/openstack-ansible/+/914461 | 19:43 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!