| NeilHanlon | noonedeadpunk: sorry just saw the messages here about libzstd... I think yes we need to just get it from EPEL | 02:34 |
|---|---|---|
| NeilHanlon | fyi i had some free time this evening and went and did a bunch of reviewing 😈 | 02:34 |
| opendevreview | Merged openstack/openstack-ansible-os_neutron master: [doc] Use neutron_ovn_distributed_fip instead of the override https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/945944 | 02:35 |
| opendevreview | Merged openstack/openstack-ansible master: Stop reffering to tests repo for ansible requirements https://review.opendev.org/c/openstack/openstack-ansible/+/945982 | 02:51 |
| opendevreview | Merged openstack/ansible-role-systemd_service master: Update requirements.yml to be installable without force flag https://review.opendev.org/c/openstack/ansible-role-systemd_service/+/941517 | 04:02 |
| opendevreview | Merged openstack/openstack-ansible-os_ceilometer master: Revert the logic of applying overrides to pipeline https://review.opendev.org/c/openstack/openstack-ansible-os_ceilometer/+/945268 | 04:13 |
| opendevreview | Merged openstack/openstack-ansible-plugins master: Skip image upload if checksum is provided https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/945248 | 04:23 |
| opendevreview | Merged openstack/openstack-ansible-plugins master: Automatically import ssh connection plugin options from the base class https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/939957 | 04:32 |
| noonedeadpunk | NeilHanlon: amazing, thanks a ton! | 07:01 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 07:31 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 07:32 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 07:35 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 07:35 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 07:35 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Update 2023.1 maintenance state https://review.opendev.org/c/openstack/openstack-ansible/+/946975 | 07:36 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: [doc] Update 2023.1 maintenance state https://review.opendev.org/c/openstack/openstack-ansible/+/946975 | 07:37 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-plugins master: Add ability to decompress images for upload https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/946918 | 08:48 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-plugins master: Add ability to decompress images for upload https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/946918 | 09:32 |
| opendevreview | Merged openstack/openstack-ansible master: Move argument parsing for dynamic_inventory to generate https://review.opendev.org/c/openstack/openstack-ansible/+/945025 | 09:50 |
| opendevreview | Merged openstack/openstack-ansible master: Remove lxb driver support https://review.opendev.org/c/openstack/openstack-ansible/+/946149 | 09:59 |
| noonedeadpunk | such a great merge progress..... | 10:03 |
| f0o | https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/tasks/swift_rings_md5sum.yml + https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/tasks/swift_rings_check.yml as well as https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/tasks/swift_rings_post_distribution_check.yml#L16-L30 seem a bit pointless to me. | 10:08 |
| f0o | it's almost 3x the same check for no apparently reason? | 10:08 |
| noonedeadpunk | Um, the only guess is that they might happen before or after the transfer | 10:09 |
| noonedeadpunk | so that you want to compare initial and resulting state | 10:09 |
| f0o | 1x happens before yes, 2 and 3 does not | 10:10 |
| noonedeadpunk | and then could be different tasks for different group of hosts, but not sure | 10:10 |
| f0o | just seems a bit superflous when the rsync passes then you alerady know that the contents are the same | 10:11 |
| noonedeadpunk | but both swift_rings_check and swift_rings_post_distribution_check seem to ruin o nthe same set of hosts.... | 10:12 |
| noonedeadpunk | well, depends... | 10:13 |
| noonedeadpunk | I don't think it specifically pointles to ensure the SHA after the sync is done | 10:13 |
| noonedeadpunk | at least to ensure that you synced correct thing | 10:13 |
| f0o | not really | 10:13 |
| noonedeadpunk | but indeed, it's more a CI check.... | 10:13 |
| noonedeadpunk | but the also `swift_rings_md5sum` is included twice... | 10:14 |
| noonedeadpunk | um, yeah, that all looks weird | 10:15 |
| f0o | :D | 10:15 |
| f0o | I had to remove it from my setup | 10:15 |
| noonedeadpunk | looking at really old code here, and it looked same: https://opendev.org/openstack/openstack-ansible-os_swift/src/tag/rocky-eol/tasks/swift_rings.yml | 10:15 |
| f0o | mainly because it would randomly fail for two hosts | 10:15 |
| noonedeadpunk | but the for some reason the check is failing in CI? | 10:15 |
| noonedeadpunk | but why does it, if the content is the same? | 10:16 |
| f0o | so that's because cat /etc/swift/*.builder !== cat /etc/swift/ring_build_files/*.builder | 10:16 |
| noonedeadpunk | like all jobs failed as https://zuul.opendev.org/t/openstack/build/e37e9e2aaa0d48f48771f59d122a8078/log/job-output.txt#11045-11052 | 10:16 |
| f0o | so at very least https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/tasks/swift_rings_md5sum.yml#L18 has the wrong path | 10:17 |
| noonedeadpunk | feels I need to spawn a test env to understand that | 10:17 |
| f0o | I opt to remove that superflous check | 10:18 |
| noonedeadpunk | as the command on itself look also weird | 10:18 |
| f0o | rsync command's success implies that sync happened correctly | 10:18 |
| noonedeadpunk | I think biggest question for me now - why it was passing before and not passing now | 10:19 |
| f0o | because of the paths | 10:19 |
| noonedeadpunk | if there was not mess-up in rsync, like missing or extra `/` which will change behaviour | 10:19 |
| noonedeadpunk | and copy with folder or without it | 10:20 |
| noonedeadpunk | (for instance) | 10:20 |
| f0o | https://paste.opendev.org/show/br2TD7iY9PdPqt8DjpKG/ this is the file structure I got | 10:21 |
| f0o | but its still not fully working. I will have to dig into it over the weekend | 10:22 |
| f0o | anyway it was just a curiousity/observation with those md5sum tasks | 10:23 |
| f0o | I wont alter them in my sshca changes, just get them to work (somehow) | 10:23 |
| noonedeadpunk | f0o: so you don't need 2 rsyncs here, imo: https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/946865/10/tasks/swift_rings_distribute.yml#16 | 10:24 |
| f0o | I do :) | 10:24 |
| noonedeadpunk | as the first `synchronize` was pulling data to the deploy host | 10:24 |
| f0o | because md5sum requires the files in /ring_build_files | 10:25 |
| noonedeadpunk | oh, and sedcond is in the loop | 10:25 |
| f0o | and swift itself requires them in / | 10:25 |
| noonedeadpunk | um | 10:25 |
| f0o | :D | 10:25 |
| f0o | it's a mess | 10:25 |
| noonedeadpunk | /o\ | 10:25 |
| f0o | technically only _first_host needs ring_build_files | 10:25 |
| f0o | whatever that magic tag was | 10:25 |
| f0o | fun right? | 10:26 |
| f0o | so if we remove the md5sum checks then we can also remove the 2nd rsync since we only care about the rings really. Then we can also tighten the rsync to only do the .gz's and not _everything_ | 10:27 |
| f0o | it's a fun rabbithole I got myself into | 10:27 |
| f0o | want to fix just ssh-keys end up almost being the maintainer now | 10:28 |
| noonedeadpunk | well, I mean, it's kind of intreresting... | 10:28 |
| f0o | that's one way to see it :D | 10:28 |
| noonedeadpunk | at least for me as I somehow enjoy messing up with ansible :D | 10:31 |
| f0o | :D | 10:32 |
| f0o | I also notice that the rings themselves are just empty | 10:32 |
| noonedeadpunk | but also, frankly speaking - I think https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/tasks/swift_rings_build.yml should be even delegated to repo_server | 10:32 |
| f0o | https://paste.opendev.org/show/bRXGjvddCqCgbflIdDcX/ | 10:32 |
| noonedeadpunk | then we could delegate all the tasks to the repo host, and fetch required rings from it via http.... and probably don't need SSH at all? | 10:34 |
| f0o | repo_server might actually be a good idea | 10:34 |
| noonedeadpunk | or are they syncing during runtime with cron or smth? | 10:34 |
| f0o | nop | 10:34 |
| f0o | one time only | 10:34 |
| noonedeadpunk | so ssh is only for this init distribution? | 10:34 |
| f0o | the only issue is that the amount of rings is not known | 10:34 |
| f0o | but I guess we can script that with an index of all files | 10:35 |
| noonedeadpunk | but can't you download the folder recuirsively? | 10:35 |
| f0o | suppose you can do that too | 10:35 |
| f0o | dont like recursive wgets, they get messy fast :D | 10:35 |
| noonedeadpunk | and if we place a checksum files separately... we can pass and compare them... | 10:35 |
| noonedeadpunk | yeah, true... | 10:35 |
| f0o | sounds good to me, in theory | 10:35 |
| f0o | right now rings_build doesnt even work as all my rings are empty xD | 10:36 |
| f0o | but at least they're distributed! | 10:36 |
| f0o | I think I will abandon the current changeset and create a new one with smaller patches now that I can isolate the sshca stuff from the rest | 10:37 |
| f0o | and do what you guys told me from the start, multiple commits and git revew | 10:37 |
| noonedeadpunk | ok, but then there's also an rsyncd service for some reason | 10:37 |
| f0o | rsyncd is used for swift itself | 10:38 |
| f0o | think they use that for replication | 10:38 |
| noonedeadpunk | but then I guess SSH CA stil lrequired for that? | 10:38 |
| f0o | dont think they use ssh for that | 10:38 |
| noonedeadpunk | but like... how rsyncd would work.... | 10:38 |
| f0o | I'd check if I had a working env lol | 10:39 |
| noonedeadpunk | yeah... | 10:39 |
| f0o | I think its anonymous rsync | 10:39 |
| noonedeadpunk | well, I can try to spawn an aio env and check at least some things with you | 10:39 |
| f0o | hah rsyncd isnt even enabled in systemd | 10:40 |
| f0o | so no clue how thats even supposed to work | 10:40 |
| f0o | but it is configured | 10:40 |
| noonedeadpunk | it should be https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/handlers/main.yml#L28-L32 | 10:41 |
| f0o | https://paste.opendev.org/show/bM6bphmLxpW204pq6Fh6/ | 10:41 |
| f0o | noonedeadpunk: I think I never got to the point because of random failures | 10:42 |
| f0o | like empty rings prevents swift from starting | 10:42 |
| noonedeadpunk | yeah... I need to run out to the bank now and do some irl stuff, but I'd love to help out in polishing the role tbh | 10:44 |
| noonedeadpunk | as it's indeed quite outdated... | 10:44 |
| noonedeadpunk | I guess we were fine just because it was passing metal jobs | 10:45 |
| f0o | no worries, i cant work on this today anyway. You've been more than helpful already throughout the week on this. I'll just do some planing and will close/reopen my changesets to make them less feature-creepy and just start refactoring this role | 10:45 |
| noonedeadpunk | (and then i nCI glance is confogured to use swift iirc) | 10:45 |
| noonedeadpunk | so it looked like be working overall | 10:45 |
| f0o | https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/templates/ring.contents.j2#L62-L66 << So if a drive is not in any group, it is simply ignored? | 10:46 |
| f0o | https://docs.openstack.org/openstack-ansible-os_swift/latest/configure-swift-config.html barely mentions groups and the examples give me a sense that it would've been optional | 10:48 |
| noonedeadpunk | `container_vars: -> swift_vars` is kind of related to groups | 10:51 |
| noonedeadpunk | but a very weird way to put this in | 10:51 |
| f0o | :D | 10:51 |
| f0o | but this explains my empty rings | 10:51 |
| noonedeadpunk | so p[retty much that would translate to having `swift` var defined inside of /etc/openstack_deploy/host_vars/swift-node5 | 10:52 |
| f0o | my issue is that it doesnt mention that groups is a _required_ setting | 10:53 |
| f0o | it just mentions that there exists a group setting, not saying why I would need it and the examples show it being used together with storage policies but factually it is always required to be set even for a plain setup | 10:54 |
| f0o | https://opendev.org/openstack/openstack-ansible-os_swift/src/branch/master/templates/ring.contents.j2#L64 something in this neseted defaults it also doesnt seem to actually default to the 'account,container' def_groups at the top | 10:56 |
| f0o | urgh I wish I had access to the env from remote now to debug this because I feel like somewhere is a bad value that makes either drives or groups empty | 10:58 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-os_swift master: Migrate role to SSHCA https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/946988 | 11:28 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-os_swift master: Migrate ring distribution to SSHCA https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/946990 | 11:36 |
| f0o | I'm liking this git review | 11:36 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-os_swift master: Migrate cronjobs to systemd-timers https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/946991 | 11:38 |
| f0o | later in the afternoon I will try out these split patches in my env and see how far I get. Now that I know that I need groups ;0 | 11:40 |
| f0o | you guys might just make me a gerrit believer after decades of gitflow | 11:44 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 11:47 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 11:49 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 11:55 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 11:56 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 12:00 |
| f0o | so flush_handles wasnt a good idea | 12:36 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-os_swift master: Migrate role to SSHCA https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/946988 | 12:37 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-os_swift master: Migrate ring distribution to SSHCA https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/946990 | 12:44 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-os_swift master: Migrate cronjobs to systemd-timers https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/946991 | 12:44 |
| noonedeadpunk | f0o: > might just make me a gerrit believer after decades of gitflow | 13:18 |
| noonedeadpunk | I told you so :p | 13:18 |
| f0o | :p | 13:19 |
| noonedeadpunk | at first you o_O at wtf is that | 13:19 |
| noonedeadpunk | but with some time.... | 13:19 |
| f0o | yep yep | 13:20 |
| f0o | I still need to find a good way to trigger the trusted_ca regeneration without flushing handlers | 13:24 |
| f0o | probably will have to change ansible-plugins to replicate what keystone does | 13:24 |
| f0o | unless there's a nicer/proper way | 13:24 |
| jrosser | the handler can always be a regular task | 13:26 |
| f0o | I think the regen task is deffo safe to be executed always | 13:27 |
| jrosser | use `register: <blah>` on one task, and then another can be `when: <blah> is changed` | 13:27 |
| jrosser | the linter might moan about it | 13:27 |
| jrosser | execute always will not be idempotent | 13:27 |
| jrosser | aim for zero changed tasks when there is nothing to do, thats really the ansible way | 13:28 |
| jrosser | though it kind of depends on the details, some tasks themselves are idempotent so the condition is not always needed | 13:29 |
| f0o | so basically move https://opendev.org/openstack/openstack-ansible-plugins/src/branch/master/roles/ssh_keypairs/handlers/main.yml into tasks and create the when: 'Create sshd trusted certificate config files' is changed ? | 13:32 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-plugins master: Move regenerate handler into task to trigger more reliably https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/946998 | 13:37 |
| f0o | ^ like that? | 13:38 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-plugins master: Move regenerate handler into task to trigger more reliably https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/946998 | 13:38 |
| opendevreview | Merged openstack/openstack-ansible master: Install inventory-manage as a console_script https://review.opendev.org/c/openstack/openstack-ansible/+/945032 | 13:40 |
| NeilHanlon | f0o: welcome to the gerrit side of the force. we only have one rule: proselytize gerrit to not-yet-belivers :P | 14:25 |
| NeilHanlon | fwiw https://graphite.dev is a similar concept for other forges like GH | 14:26 |
| NeilHanlon | noonedeadpunk: wow! only one 'needs approval' in the queue!! | 14:26 |
| noonedeadpunk | I think you can use https://gerrithub.io/ ? | 14:28 |
| noonedeadpunk | or it's smth different? | 14:28 |
| NeilHanlon | oh wow, i had no idea about gerrithub lol | 14:28 |
| NeilHanlon | graphite is something different.. just a similar concept to how gerrit manages stacking | 14:29 |
| noonedeadpunk | well. it doesn't have own storage - storage is github. But you can leverage the favorite flow there :D | 14:29 |
| noonedeadpunk | ah | 14:29 |
| noonedeadpunk | but it makes totall sense.... | 14:29 |
| NeilHanlon | not being able to relate pull requests on github is really annoying | 14:30 |
| NeilHanlon | especially coming from a gerrit mindset heh | 14:30 |
| noonedeadpunk | especially if multiple ppl want to collaborate on some feature development | 14:31 |
| noonedeadpunk | yeah... broken by gerrit mindset - which is really hard to un-see once used to it | 14:32 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947008 | 14:57 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947008 | 14:58 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: fix small mistake in Deployment hosts https://review.opendev.org/c/openstack/openstack-ansible/+/947008 | 14:59 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: small mistake in Deployment hosts https://review.opendev.org/c/openstack/openstack-ansible/+/947008 | 15:02 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Update roles for 31.0.0.0b1 release https://review.opendev.org/c/openstack/openstack-ansible/+/947014 | 15:22 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Move zuul preparation for role/collection bootstrap https://review.opendev.org/c/openstack/openstack-ansible/+/939151 | 15:23 |
| noonedeadpunk | seem we have now to land https://review.opendev.org/c/openstack/openstack-ansible/+/947014 | 15:24 |
| noonedeadpunk | and I need to look into bump script about wtf is wrong with it | 15:24 |
| noonedeadpunk | as apparently< it's taking HEAD without accounting for the tree or smth like that | 15:24 |
| opendevreview | Merged openstack/openstack-ansible-haproxy_server master: Make sysctl configuration path configurable https://review.opendev.org/c/openstack/openstack-ansible-haproxy_server/+/939601 | 15:32 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2024.2: Use first utility host as default host for openstack_resources https://review.opendev.org/c/openstack/openstack-ansible/+/947019 | 15:36 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Update roles for 31.0.0.0b1 release https://review.opendev.org/c/openstack/openstack-ansible/+/947014 | 15:36 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Unfreeze roles after milestone release https://review.opendev.org/c/openstack/openstack-ansible/+/946281 | 15:40 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Move zuul preparation for role/collection bootstrap https://review.opendev.org/c/openstack/openstack-ansible/+/939151 | 15:40 |
| darkhackernc | noonedeadpunk++ thank you very much | 15:56 |
| f0o | NeilHanlon: One of Us. One of Us. One of Us. :D | 16:03 |
| f0o | noonedeadpunk: I have a working Swift setup! there's just one caveat, I had to entirely exclude the md5 shenanigans that are in the role. I hoenstly fail to see the value in those since rsync takes care of it already. I will add a patch to my series to remove those and we can have the discussion about it on gerrit | 16:05 |
| opendevreview | Daniel Preussker proposed openstack/openstack-ansible-os_swift master: Remove md5 checks of rings/builders https://review.opendev.org/c/openstack/openstack-ansible-os_swift/+/947021 | 16:07 |
| noonedeadpunk | yeah, I'm not sure I like the idea of dropping it as a whole. but I think I can see how it can be done better at least | 16:22 |
| f0o | I'd say the patch to drop it is a conversation starter | 16:23 |
| noonedeadpunk | yeah | 16:24 |
| f0o | if we want to keep it (in any form or shape) we will need to addresss the chicken-egg issues of initial deployment. Which is what was attempted with the comparisson to empty md5sum. it's execution is just poorly | 16:24 |
| f0o | I had multiple instances where rexecuting the same playbook would yield fails or passing seemingly random | 16:25 |
| f0o | either way, might feature creep it with the idea of moving the ring-building onto the repo_server | 16:26 |
| noonedeadpunk | I guess I'd be re-implementing it with `ansible.builtin.find` and then `ansible.builtin.stat` across find results | 16:26 |
| f0o | that would entirely get rid of it too since we can just download the md5sum file and use md5sum itself | 16:26 |
| f0o | speaking of something else | 16:26 |
| f0o | any idea why https://zuul.opendev.org/t/openstack/build/1e8dd970e06340b5a149909c19abdd2e is failing? | 16:26 |
| noonedeadpunk | yeah, I mean, moving to repo server looks somehow cleaner... | 16:27 |
| f0o | the tempest assertion error is not really useful to me | 16:27 |
| noonedeadpunk | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_1e8/openstack/1e8dd970e06340b5a149909c19abdd2e/logs/openstack/aio1-utility-container-9fe07d1c/utility/stestr_results.html | 16:27 |
| f0o | I'm just happy my swift setup is working now - highlight of the week hah | 16:27 |
| noonedeadpunk | it could be intermittent as well | 16:27 |
| noonedeadpunk | yeah, right, and now you can enlighten me if I get things right or not | 16:28 |
| f0o | how can I retrigger the CI? or is that limited to a select few? | 16:29 |
| noonedeadpunk | (and I can spawn a sandbox I guess) | 16:29 |
| noonedeadpunk | just leave the comment with `recheck - <reason>` | 16:29 |
| f0o | oh coolio | 16:29 |
| noonedeadpunk | somehow rocky is unhappy with SSH :( | 16:30 |
| noonedeadpunk | wonder why it fails to auth | 16:30 |
| noonedeadpunk | https://zuul.opendev.org/t/openstack/build/e1e19c5644bc4672a36583e8f7a225da/log/job-output.txt#10214-10218 | 16:30 |
| f0o | yeah dunno at least I'm happy it's non-voting xD | 16:31 |
| noonedeadpunk | I think it is | 16:32 |
| f0o | oh youre right | 16:32 |
| f0o | I thought it was the distro variant | 16:32 |
| noonedeadpunk | jsut all rocky are unhappy, and centos stream is fine at the same time | 16:33 |
| noonedeadpunk | so it's weird | 16:33 |
| f0o | I can look into it over the weekend | 16:33 |
| f0o | can spin up a rocky vm and just trace it | 16:33 |
| noonedeadpunk | btw - https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/946998 is a valid failure | 16:36 |
| f0o | it is? | 16:36 |
| noonedeadpunk | https://zuul.opendev.org/t/openstack/build/5f46157668544794ab007e56f7e6d49b/log/job-output.txt#12574-12597 | 16:36 |
| f0o | https://zuul.opendev.org/t/openstack/build/5f46157668544794ab007e56f7e6d49b/log/job-output.txt#12163-12166 | 16:37 |
| f0o | this was skipped | 16:37 |
| f0o | I guess the xyz.changed didnt work | 16:37 |
| noonedeadpunk | I'd guess it should be `xyz is changed` is th right syntax | 16:38 |
| f0o | but I honestly didnt know what I ws doing there anyway | 16:38 |
| f0o | I googled it a bit and some guide said to do a register and then .changed on whatever you register | 16:38 |
| f0o | guess it/I was wrong | 16:38 |
| noonedeadpunk | but tbh, I still somehow like it done with the handler | 16:39 |
| f0o | I do need the handler to trigger before I do all the rsync stuff tho | 16:39 |
| noonedeadpunk | sand that flushing handler in role doesn't do the trick | 16:39 |
| f0o | and flush handlers broke a lot of things because it started triggering the swift handlers (systemd) which all failed since well it wasnt yet time for them | 16:40 |
| noonedeadpunk | yeah, right | 16:40 |
| f0o | maybe I can re arrange the tasks somehow to make key setup way earlier | 16:40 |
| f0o | idk if that solves it entirely or just masquerades it | 16:40 |
| noonedeadpunk | yeah, I guess it doesn't solve it properly | 16:43 |
| noonedeadpunk | it's more of indeed moving out key setup out of role to pre_tasks then... | 16:43 |
| f0o | but that's a change in ansible-plugins then right? | 16:45 |
| noonedeadpunk | yeah | 16:46 |
| noonedeadpunk | but again. let's sleep with it a bit... | 16:46 |
| f0o | +1 | 16:46 |
| f0o | friday afternoon now, I'm not gonna do much now anyway | 16:46 |
| noonedeadpunk | I'm somehow exhausted atm and can't think very clear. NExt week will spawn a sandbox and really wanna play with swift | 16:46 |
| f0o | no rush | 16:47 |
| f0o | the patches I made might not be perfect but they're a good inspiration to get moving forward | 16:47 |
| f0o | anyway I'm off now, have a great friday and weekend ahead (: | 16:49 |
| noonedeadpunk | yup, thanks for your work! | 16:49 |
| noonedeadpunk | much appreciated! | 16:49 |
| noonedeadpunk | and have a good one | 16:49 |
| NeilHanlon | yes, thanks f0o! Have a great weekend :) | 17:18 |
| opendevreview | Neil Hanlon proposed openstack/openstack-ansible master: Update roles for 31.0.0.0b1 release https://review.opendev.org/c/openstack/openstack-ansible/+/947014 | 17:27 |
| NeilHanlon | noonedeadpunk: made some grammar changes for the release notes there | 17:27 |
| opendevreview | Merged openstack/openstack-ansible master: docs: small mistake in Deployment hosts https://review.opendev.org/c/openstack/openstack-ansible/+/947008 | 18:28 |
| opendevreview | Merged openstack/openstack-ansible master: docs: little fix for Zed release https://review.opendev.org/c/openstack/openstack-ansible/+/946961 | 18:28 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 19:43 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 19:44 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 19:46 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 19:48 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 19:50 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 19:54 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: WIP https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 19:59 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: minor fixes with OpenStack-Ansible naming https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 20:01 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: minor fixes with OpenStack-Ansible naming https://review.opendev.org/c/openstack/openstack-ansible/+/947039 | 20:03 |
| opendevreview | Merged openstack/openstack-ansible master: Symlink console_scripts to /usr/local/bin/ https://review.opendev.org/c/openstack/openstack-ansible/+/945033 | 23:14 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!