[07:43:09] <blancadesal>	 morning 
[07:43:48] <blancadesal>	 is there a reason jobs-cli doesn't use pre-commit, or has it just not been migrated to that setup yet? 
[07:44:40] <arturo>	 yeah, nobody did it just yet
[07:44:43] <arturo>	 morning 
[07:46:13] <blancadesal>	 ok. I might do it, or if not, create a task for it
[08:20:29] <dcaro>	 I started using ruff on the api-gateway, feel free to use the one from there instead of the many linters (note that there's also some config in pyproject.toml)
[08:32:11] <blancadesal>	 lurking on your weekend, huh? :))
[08:33:47] <blancadesal>	 btw, I discovered ruff needs some extra config to sort imports like isort, it doesn't do it by default https://github.com/blancadesal/fastapi-toolhunt/blob/main/pyproject.toml#L28
[09:36:15] <arturo>	 blancadesal: the builds-cli repo contains a bunch of commits in the history that are probably not meant to be there
[09:36:47] <arturo>	 maybe a MR with multiple commits that wasn't squashed before merging to main
[09:42:55] <blancadesal>	 arturo: yup, that's what happened
[09:45:30] <blancadesal>	 I rarely do more than a single commit, but that time I did and forgot squashing isn't set up as the default option on merge
[09:47:22] <arturo>	 I'm tempted to rewrite the history and force everyone to rebase
[09:47:50] <arturo>	 opinions?
[09:51:07] <blancadesal>	 ok for me
[09:52:46] <arturo>	 do you know how to do this yourself?
[09:53:51] <blancadesal>	 I can probably figure it out, unless you'd loooove to fix my mess xd
[09:54:31] <arturo>	 probably the most sensitive thing is to carefully change the gitlab repo settings to allow you to force push to main with the rewritten history
[09:55:09] <arturo>	 the local squash is probably no big deal, first result in google search should do it
[09:58:23] <blancadesal>	 ok, I'll finish up a couple other things first
[10:03:46] <arturo>	 ok, ping me later and I will assist you
[10:07:39] <blancadesal>	 thx!
[10:14:54] <taavi>	 note to self: openstack will just say "the request you have made requires authentication" if you try to log in with an user that does not have any roles assigned anywhere
[10:37:05] <taavi>	 arturo: are you interested in reviewing this? https://gerrit.wikimedia.org/r/c/operations/puppet/+/1040115
[10:37:22] <arturo>	 sure
[10:42:03] <arturo>	 taavi: the patch LGTM, my only comment is a meta-question of whether we should stop doing the ::base indirection
[10:43:39] <taavi>	 yeah. I don't want to introduce a new model for that in this patch, but I am only using the ::codfw1dev:: class to set the settings that are not set in the role hiera (keystone url and region)
[10:44:55] <arturo>	 +1'd
[10:46:34] <arturo>	 taavi: may I ask for your +1 here? https://gitlab.wikimedia.org/repos/cloud/toolforge/tools-webservice/-/merge_requests/37
[10:48:43] <taavi>	 yes, in a moment
[11:26:46] <arturo>	 heads up, I'm deploying kyverno to toolsbeta as a way to test this MR https://gitlab.wikimedia.org/repos/cloud/toolforge/toolforge-deploy/-/merge_requests/238
[11:27:59] <blancadesal>	 ack
[11:28:40] <blancadesal>	 arturo: I'm about to unprotect main on builds-cli to be able to force push the history rewrite
[11:28:53] <arturo>	 blancadesal: ACK
[11:29:03] <arturo>	 do you have the clean history already on your local git?
[11:29:22] <blancadesal>	 yes
[11:29:33] <arturo>	 ok!
[11:34:25] <blancadesal>	 done, main is protected again
[11:35:37] <arturo>	 thanks! the result looks good to me
[11:36:58] <blancadesal>	 \o/
[12:59:11] <arturo>	 kubernetes is failing to schedule pods in lima-kilo @ laptop because
[12:59:39] <arturo>	 insufficient CPU
[13:48:46] <blancadesal>	 is it possible to allocate more?
[14:25:15] <arturo>	 I'm looking into having more nodes in lima-kilo k8s
[14:46:23] <arturo>	 blancadesal: are you around?
[14:51:17] <arturo>	 oh, nevermind! I wanted to compare your setup to mine, but I think I found my problem
[15:53:30] <blancadesal>	 what was it?
[15:56:11] <arturo>	 adding more nodes to the kind cluster is not trivial :-(
[15:56:41] <arturo>	 I wanted to check something about the base config without revert and rebuild the lima-vm
[15:56:54] <arturo>	 this is what I'm exploring at the moment https://gitlab.wikimedia.org/repos/cloud/toolforge/lima-kilo/-/merge_requests/139
[15:58:57] <arturo>	 have a good weekend
[15:58:59] * arturo offline
[18:17:37] <taavi>	 oh, lovely, someone has put an open registration wiki at https://language-translate2.wmcloud.org/index.php/Special:RecentChanges and the internet has found that
[18:24:17] <taavi>	 T366934
[18:24:18] <stashbot>	 T366934: language-translate2 filled with spambots - https://phabricator.wikimedia.org/T366934
[18:32:19] <taavi>	 in which taavi gets frustrated about misconfigured things: T366935 T366936
[18:32:20] <stashbot>	 T366935: Cloud VPS mail servers should drop mail sent from non-supported domains - https://phabricator.wikimedia.org/T366935
[18:32:20] <stashbot>	 T366936: Reject outbound traffic to port 25 (SMTP) from instances without public IPs - https://phabricator.wikimedia.org/T366936