09:53 <eric_engestrom> DavidHeidelberg[m]: FYI I believe when you tag someone by editing an MR description (or comment), it doesn't actually notify them; best to add a comment instead

09:54 <DavidHeidelberg[m]> eric_engestrom: oh, I didn't know that. Thanks ;)

13:54 <zmike> is there a gitlab setting to have MRs create themselves using the commit log from the last commit in a branch instead of the first?

13:54 <zmike> cuz that seems like it would be way more accurate as a template

13:57 <pq> +1 but I haven't seen that

14:02 <jenatali> zmike: In my experience it uses the first one that's more than just a title, if that's helpful

14:08 <pq> I never write any commit that's only a title of a message

14:08 <pq> not in public anyway :-p

14:09 <eric_engestrom> +1 to what jenatali said, but I agree that it would make more sense to use the last commit instead as usually the rest are prep work for that last one

14:22 <bentiss> daniels: dropping this here so it gets a little bit more visibility: https://gitlab.freedesktop.org/freedesktop/mr-label-maker/-/merge_requests/11 -> this should allow to retrigger mr-label-maker by removing the "mr-label-maker::stamped" label it added

14:22 <bentiss> in case anyone is interested...

14:24 <bentiss> And regarding why mr-label-maker works well in some case and not in some others: if you open a draft MR first, it leaves time for gitlab to process the MR, and so when the checnge from draft is done, the MR is already ready and can be processed by mr-label-maker. If the MR is not a draft first, mr-label-maker do not see the changes, and do not apply labels

14:39 <alatiera> headups post runner 15.11 upgrade it would refuse to pick up any jobs due to the grafana config

14:40 <alatiera> Apr 26 16:34:35 gst-htz-2 gitlab-runner[8695]: FATAL: Failed to create listener for metrics server builds=0 error=listen tcp 10.88.0.1:3807: bind: cannot assign requested address

14:40 <alatiera> Apr 26 16:34:35 gst-htz-2 systemd[1]: gitlab-runner.service: Main process exited, code=exited, status=1/FAILURE

14:40 <alatiera> Apr 26 16:34:35 gst-htz-2 systemd[1]: gitlab-runner.service: Failed with result 'exit-code'.

14:41 <alatiera> also gitlab-runner status would just claim "its running" just fine

14:44 <alatiera> in hindsight I should have probably commented out the listen_adress thing while registering

16:09 <alanc> Do you have to be a group owner to be able to add people to the group in gitlab?

16:56 <eric_engestrom> alanc: Maintainers and Owners can do add people: https://docs.gitlab.com/ee/user/project/members/index.html#add-users-to-a-project

17:05 <alanc> that's for projects - but it helped me find later in that doc that yes, for groups you have to be an owner: https://docs.gitlab.com/ee/user/group/manage.html#add-users-to-a-group

17:07 <alanc> which I guess raises the next question - what would it take for me to become an owner for the xorg group?

18:13 <bentiss> alanc: done :)

18:15 <alanc> thanks bentiss

22:50 <whot> zmike: I vaguely remember that's a feature from a higher tier

22:50 <zmike> dang

22:52 <whot> https://docs.gitlab.com/ee/user/project/merge_requests/commit_templates.html - looks like only first_commit exists as variable

22:53 <whot> IOW this doesn't seem to exist at all, even in higher tiers