[09:07:51] <dcaro>	 morning
[10:15:51] <dcaro>	 ahj.... when triggering a cronjob manually (toolforge jobs restart) we are manually building the job, and not taking into account any extra things added to the `CronJob.spec.jobTemplate` or the `Cronjob.spec.jobTemplate.spec.template.spec` ... 
[10:16:22] <dcaro>	 that's why the timeout option is not working when triggered manually (it adds `Cronjob.spec.jobTemplate.activeDeadlineSeconds`)
[11:47:16] <dcaro>	 it was because of the way we (jobs-api) start a job from a cronjob, git a fix
[11:47:22] <dcaro>	 https://gitlab.wikimedia.org/repos/cloud/toolforge/jobs-api/-/merge_requests/144
[11:51:39] * dcaro lunch
[14:04:57] <dcaro>	 https://docs.google.com/document/d/1WYEEWyLoD7pKYapuT8pyHmpxxEOCSmF7VKdxpnVQRFA/edit?tab=t.0 <- toolforge sync meeting notes
[14:18:30] <dcaro>	 dhinus: https://gitlab.wikimedia.org/repos/cloud/toolforge/toolforge-deploy/-/blob/main/components/calico/values/tools.yaml?ref_type=heads#L7
[16:02:43] <andrewbogott>	 In case those temp alerts about clouddumps1001 are driving anyone up the wall -- there is a plan underway to move the source of hot air away from the host. Please continue to ignore the warnings in the meantime.  https://phabricator.wikimedia.org/T383723
[16:15:11] <dcaro>	 xd
[16:15:46] <dcaro>	 the actual shape of the servers is affecting the behavior of the neighbor servers... interesting
[16:16:22] <dhinus>	 I added a 7-day silence for that specific alert
[16:18:15] <andrewbogott>	 dcaro: yeah, I'm surprised that a little server under a big server gets bullied by the big one
[18:37:00] * dcaro off
[18:37:02] <dcaro>	 cya tomorrow!
[21:19:51] <andrewbogott>	 Is there a way to "toolforge build start https://gitlab.wikimedia.org/toolforge-repos/openstack-browser.git" on a test branch, or will it always grab the main branch?
[21:20:27] * andrewbogott can merge-to-test but is trying not to be so messy
[21:26:20] * andrewbogott discovers --ref