12:11 <alatiera[m]> DavidHeidelberg: do we have a room for marge?

12:11 <alatiera[m]> irc or matrix

14:30 <pq> What do people do with Gitlab branches they have used for a MR that didn't make it? I'd prefer to rename the branch from mr/laalaa to closed/laalaa, but that does not seem in any way possible.

14:43 <MrCooper> not sure I'm quite following: the MR didn't make it, but remains open?

14:44 <pq> the MR is closed, and won't re-open

14:44 <MrCooper> then what is the problem with renaming the branch?

14:45 <pq> I didn't try it, but since it is not possible to point an existing MR to another branch, nor to rename a branch in Gitlab, I presume renaming would be equivalent to deleting the branch.

14:46 <MrCooper> right, and? :)

14:46 <pq> I tried deleting the branch, and Gitlab warned about losing data, but looks like it finds the commits for a moment still at least.

14:46 <pq> I wouldn't want to make a MR page say code not found, but I also have no use for the branch myself.

14:47 <pq> Is deleting the branch safe, in that the MR page doesn't lose anything?

14:48 <MrCooper> my assumption has been that the MR keeps the commits alive, I'm not sure though

14:48 <pq> hmm, I might accidentally later create a branch with the same name for an unrelated purpose, that might mess up?

14:48 <MrCooper> does the warning specifically mention MRs?

14:48 <pq> I don't think it did.

14:50 <MrCooper> IIRC the branch name of a merged MR can be re-used, not sure about closed though, since it could be re-opened anytime

14:50 <svuorela> I think I repeatedly have (mostly on other gitlab instances) made misc-fixes branches

16:05 <eric_engestrom> pq, MrCooper: yeah gitlab never deletes commits, so it's perfectly safe to delete a branch when you close an MR (and deleting the branch will auto-close the MR), the MR will continue to show the commits as they were when the branch was deleted, and if you create a new branch later with the same name it won't confuse gitlab and the old MR will not be touched

16:08 <MrCooper> eric_engestrom: I wonder what happens if the MR is reopened though

16:09 <eric_engestrom> if the branch doesn't exist, iirc it throws an error; if the branch does exist, it updates the MR to the new state of the branch

16:10 <eric_engestrom> (not 100% sure on the "error if branch doesn't exist", I haven't done that in forever)

16:10 <eric_engestrom> (but also that's not an interesting case ^^)

16:17 <MrCooper> that could be confusing when re-using the same name for an unrelated branch

16:35 <ndufresne> Is there any way to workaround gitlab file size limit ? we have code we can't code review in gitlab because of this (these are new files, not changes), https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5865/diffs

16:46 <heftig> ndufresne: you could open the MR in webide. it will open up to 10 diffs by default, and you can diff the rest by right-clicking on the affected files in the tree and choosing to compare with the merge request base

16:49 <psykose> does it not work to press expand file? expands for me and allows leaving a comment though i didn't try submitting it

17:03 <DavidHeidelberg> alatiera[m]: nope, no room

17:48 <ndufresne> psykose: the error is after tryign to expand, so no

17:48 <ndufresne> heftig: I'll try webide, I need to comment inline, hopefully that works

17:51 <heftig> ndufresne: it doesn't

17:51 <heftig> or at least i didn't see it

17:53 <ndufresne> yeah, no, I does not give me anything useful

17:53 <ndufresne> its a 6K line of code file, but as its brand new, it would be weird to split just to workaround tool limitations

17:59 <heftig> ah right, that file doesn't display at all, so you can't inline comment

18:37 <glehmann> a630 CI looks dead? https://gitlab.freedesktop.org/mesa/mesa/-/pipelines/1308339

23:55 <jenatali> alatiera: After trying to clean up disk space, we've been hitting issues with our runner: https://gitlab.freedesktop.org/nadaouf/mesa/-/jobs/66436123. Any ideas?

23:56 <alatiera> yes actually

23:56 <alatiera> I spend all of saturday figuring that out

23:56 <alatiera> one sec

23:56 <jenatali> Oh great

23:56 <alatiera> jenatali https://gitlab.com/gitlab-org/gitlab-runner/-/issues/38267

23:56 <alatiera> didn't test downgrading, but creating a network by hand and adding it to the toml works fine

23:56 <jenatali> Thanks!

23:57 <alatiera> the weird thing is I only saw this on a new runner, not existing ones that got upgraded

23:57 <jenatali> Weird...

23:57 <alatiera> but they might jsut explode anytime I guess

23:58 <alatiera> [runners.feature_flags]

23:58 <alatiera> FF_NETWORK_PER_BUILD = true

23:58 <alatiera> [runners.docker]

23:58 <alatiera> network_mode = "mynat"

23:58 <alatiera> and

23:58 <alatiera> docker network create --driver nat --subnet=192.168.111.0/24 mynat

23:59 <alatiera> (as mentioned in a comment on the issue)

23:59 <jenatali> Yeah I see that, thanks, that should help

23:59 <jenatali> In the future if you happen to see something like this, any chance you'd drop me a note? :)