ChanServ changed the topic of #freedesktop to: https://www.freedesktop.org infrastructure and online services || for questions about freedesktop.org projects, please see each project's contact || for discussions about specifications, please use https://gitlab.freedesktop.org/xdg or xdg@lists.freedesktop.org
Haaninjo has quit [Quit: Ex-Chat]
systwi_ has quit [Ping timeout: 480 seconds]
systwi has joined #freedesktop
systwi has quit [Ping timeout: 480 seconds]
swisstackle has quit []
Leopold__ has quit [Ping timeout: 480 seconds]
systwi has joined #freedesktop
gtristan has joined #freedesktop
gtristan has quit [Remote host closed the connection]
gtristan has joined #freedesktop
systwi has quit []
ximion has quit []
<whot> sigh, we need an admin in my timezone so I don't have to delete spam issues individually :)
agd5f_ has joined #freedesktop
agd5f has quit [Ping timeout: 480 seconds]
<mupuf> whot: it's gone
<whot> mupuf: thanks!
<mupuf> you're welcome :)
<mupuf> I'm not in your TZ, but I guess not *that* far
* whot wonders what the business case is for spamming gitlab.fdo
<mupuf> SEO?
<whot> heh, you're what, 8h away from me?
<mupuf> Something like that, I have in the central european time... and I tend to wake up early nowadays
<mupuf> it is 7am now
<whot> yeah, 8h then (utc+10 here)
<mupuf> hehe
<whot> i get to wake up before 5am these days. the benefit of a 3y old...
<mupuf> even worse than me then... I usually wake up at ~6am, due to the pregnant wife sleeping badly
<whot> enjoy the sleep-ins while you can
<mupuf> thanks, I will!
<whot> and hope that your kid isn't a duracell bunny
<mupuf> so far, she seems like it, if we can judge by the might and frequency of the kicks
<mupuf> she kicked a nurse yesterday during an exam :D
gtristan has quit [Ping timeout: 480 seconds]
<whot> lol
gtristan has joined #freedesktop
Mempler has joined #freedesktop
Mempler has left #freedesktop [#freedesktop]
ppascher has quit [Quit: Gateway shutdown]
jkhsjdhjs has quit [Ping timeout: 480 seconds]
alanc has quit [Remote host closed the connection]
alanc has joined #freedesktop
vbenes has quit []
MajorBiscuit has joined #freedesktop
damian_ has quit []
damian has joined #freedesktop
jkhsjdhjs has joined #freedesktop
phasta has joined #freedesktop
vbenes has joined #freedesktop
gtristan has quit [Ping timeout: 480 seconds]
* bentiss is not happy with the way the igt is creating its tags: https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/container_registry -> 7000 tags per repository is just insane. I have to import them one by one with gsutil to switch to the registry db :(
<mupuf> bentiss: I don't think the problem is creating a lot of tags, it is just that it is never removing them]
<mupuf> just like artifacts, they should get deleted
<bentiss> mupuf: well, no up to now, you couldn't remove tags, there were no gc
<mupuf> yeah
<bentiss> but here with 7000 tags per repo, that is 7000 layers I actually have to import (even if they are small)
<mupuf> and removing the old ones wouldn't help?
<bentiss> still 7000 layers times 6 + 12000 for the igt repo
<bentiss> mupuf: I am not doing that kind of precision work, blindly importing the repos for now
<mupuf> right, it is likely the best approach. Then we can set some policies in harbor to keep images for max 6 months and let it garbage collect
<mupuf> hmm, actually, we probably shouldn't do that. I can work on the IGT side on a scheduled pipeline that would remove the old images
<bentiss> actually... I might be wrong, that's roughly only 32 layers per repo, but 7000 manifests, which are just taking time to import
<mupuf> makes more sense to me
<mupuf> but yeah... that will take a lot of time
<mupuf> on the topic of containers, I may have found a way to get ci-templates to generate containers for arbitrary architectures
<mupuf> the issue was that binfmt cannot be set up in a container... but if we set it up on the host and the guest just happens to have the binaries at the expected location, buildah happily goes forward
<bentiss> and on the topic of containers, gitlab 15.9 might help us work with harbor in a more efficient way -> https://docs.gitlab.com/ee/ci/secrets/id_token_authentication.html was the missing bit to directly use the JWT token with harbor
<bentiss> mupuf: but buildah is already using the host with a chroot, so binfmt should be OK to be used, no?
<mupuf> I have been experimenting with that in valve-infra. In the end, creating a multi-arch container would be as easy as setting: FDO_DISTRIBUTION_PLATFORMS: 'linux/amd64 linux/arm64/v8'
<bentiss> cool :)
<mupuf> what you call "host" is the gitlab-runner-spawned container
<bentiss> so only one job?
<mupuf> what I call host is what runs the gitlab-runner
<bentiss> mupuf: yeah, host is the runner
<bentiss> mupuf: the runners already have binfmt IIRC
<mupuf> well, then everything should just work then :D
* bentiss waits for the MR then ;)
<mupuf> Yep, that's the plan!
<mupuf> Will make b2c's pipeline simpler, and remove custom code in valve-infra
<mupuf> bentiss: hehe, wonderful
ximion has joined #freedesktop
ximion has quit []
gtristan has joined #freedesktop
vkareh has joined #freedesktop
gtristan has quit [Ping timeout: 480 seconds]
phasta has quit [Quit: Leaving]
Leopold has joined #freedesktop
arekm has quit [Ping timeout: 480 seconds]
arekm has joined #freedesktop
<bentiss> I stand corrected of my correction: gsutil ls gs://.../gfx-ci/igt-ci-tags/igt/_layers/\*\* | wc -l -> 18065 layers
<bentiss> this is just insane
agd5f has joined #freedesktop
agd5f_ has quit [Ping timeout: 480 seconds]
vkareh is now known as Guest5669
vkareh has joined #freedesktop
vkareh has quit [Remote host closed the connection]
Guest5669 has quit []
vkareh has joined #freedesktop
MrCooper has joined #freedesktop
jarthur has quit [Ping timeout: 480 seconds]
jarthur has joined #freedesktop
agd5f_ has joined #freedesktop
agd5f has quit [Ping timeout: 480 seconds]
MajorBiscuit has quit [Quit: WeeChat 3.6]
Leopold__ has joined #freedesktop
Leopold has quit []
Jake[m] has joined #freedesktop
Jake[m] has left #freedesktop [#freedesktop]
agd5f has joined #freedesktop
vkareh has quit [Quit: Leaving]
agd5f_ has quit [Ping timeout: 480 seconds]
Haaninjo has joined #freedesktop
<DodoGTA> Why does cloning the Mesa repository fail with git clone --mirror?
ximion has joined #freedesktop
infernix has quit [Ping timeout: 480 seconds]
infernix has joined #freedesktop
agd5f_ has joined #freedesktop
agd5f has quit [Ping timeout: 480 seconds]
agd5f has joined #freedesktop
agd5f_ has quit [Ping timeout: 480 seconds]
vbenes has quit []
agd5f has quit [Ping timeout: 480 seconds]