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
JanC is now known as Guest15388
JanC has joined #freedesktop
Guest15388 has quit [Ping timeout: 480 seconds]
mairacanal has joined #freedesktop
agd5f has quit [Read error: No route to host]
agd5f has joined #freedesktop
Paddi3 has joined #freedesktop
Paddi has quit [Read error: Connection reset by peer]
Paddi3 is now known as Paddi
kasper93 has quit [Read error: Connection reset by peer]
kasper93 has joined #freedesktop
JanC has quit [Ping timeout: 480 seconds]
kasper93 has quit [Read error: Connection reset by peer]
kasper93 has joined #freedesktop
JanC has joined #freedesktop
scrumplex_ has joined #freedesktop
scrumplex has quit [Ping timeout: 480 seconds]
DragoonAethis has quit [Quit: hej-hej!]
DragoonAethis has joined #freedesktop
alarumbe has joined #freedesktop
JanC is now known as Guest15393
JanC has joined #freedesktop
Guest15393 has quit [Ping timeout: 480 seconds]
alanc-away has joined #freedesktop
alanc has quit [Remote host closed the connection]
JanC has quit [Ping timeout: 480 seconds]
JanC has joined #freedesktop
airlied has quit [Read error: Connection reset by peer]
airlied has joined #freedesktop
JanC is now known as Guest15398
JanC has joined #freedesktop
Guest15398 has quit [Ping timeout: 480 seconds]
alanc-away has quit [Remote host closed the connection]
alanc-away has joined #freedesktop
ximion has quit [Ping timeout: 480 seconds]
ximion has joined #freedesktop
_whitelogger_ has joined #freedesktop
_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #freedesktop
_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #freedesktop
_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #freedesktop
ximion has quit [Remote host closed the connection]
alanc_work has joined #freedesktop
alanc-away has quit [Remote host closed the connection]
_whitelogger_ has quit [Remote host closed the connection]
gnuiyl has joined #freedesktop
_whitelogger has joined #freedesktop
swatish2 has joined #freedesktop
alarumbe has quit [Quit: ZNC 1.8.2+deb3.1+deb12u1 - https://znc.in]
_whitelogger has joined #freedesktop
jsa1 has joined #freedesktop
_whitelogger has joined #freedesktop
swatish21 has joined #freedesktop
swatish2 has quit [Ping timeout: 480 seconds]
<dj-death> I'm getting pretty consistent panvk timeouts :
<dj-death> 00:15:44.367: ERROR - dEQP error: Incomplete job or timeout
<dj-death> 00:15:44.367: ERROR - Test dEQP-GLES3.functional.fragment_out.array.int.r16i_lowp_int: Crash: See "/results/c21.r1.log"
<dj-death> trying to land !34109
<dj-death> any objection if I put the tests in the flaky list?
<dj-death> looks like there are already flaky ones in that category
tzimmermann has joined #freedesktop
<mupuf> dj-death: yeah, I grew tired yesterday and just documented them
<mupuf> I think we could just use a wildcard here
<hakzsam> is the atom feed for commits broken now too?
_whitelogger has joined #freedesktop
<daniels> dj-death, mupuf: seems like you could probably flake fragment_out.array..* until someone doubles back and has a look at midgard
jsa1 has quit [Ping timeout: 480 seconds]
jsa1 has joined #freedesktop
<mupuf> daniels: agreed, I did not dare yesterday, but I was *this* close šŸ˜
<mupuf> kusma: ^
<kusma> I think I saw this being added to flakes somewhere this morning, no?
<kusma> Really curious why this started flaking hard recently...
sima has joined #freedesktop
<mupuf> kusma: yeah, it was my commit yesterday
<mupuf> but guidance or blessings for dj-death would be appreciated ;)
<kusma> I mean, let's add it flakes, sure. That's what it's for, no? :)
<daniels> ^
<kusma> That 100% has my blessings :)
<daniels> from my pov, the guidance has always very clearly been - if it occurs in current main then it is by definition expected, so if the list of expectations doesn't match reality, fix the list
<kusma> dj-death: No objections indeed
<daniels> (I mean obv ideally fix reality to have no bugs in v5 GPUs when we're currently mostly focusing on v10-13, buuuutttt ...
<daniels> dj-death: so yeah, no need to ask for permission at any point - just add them in as you see them (which is hopefully ~never ofc), then either tag relevant people in the MR to let them know you've done it or ping it over on IRC
<mupuf> I meant wildcard or add to the list;)
<dj-death> daniels: kusma: thanks
jsa1 has quit [Ping timeout: 480 seconds]
JanC is now known as Guest15413
JanC has joined #freedesktop
<mupuf> bebti
<mupuf> bentiss: another casualty of anubis: RSS feeds
<mupuf> can you use GET arguments to decide to o use Anubis or not in the fastly config?
Guest15413 has quit [Ping timeout: 480 seconds]
<emersion> i think anubis broke my GitLab runner
<daniels> emersion: is it still broken now? all the external runners were broken a day or two ago but that did get fixed
<emersion> yeah still broken
<emersion> seems to GET a 404 when fetching a file which exists via the API
jsa1 has joined #freedesktop
pjakobsson_ has joined #freedesktop
<hakzsam> what is anubis btw?
<pq> A cute manga character that demoralizes evil robots attacking servers.
pjakobsson has quit [Remote host closed the connection]
<mupuf> bentiss: ack, thanks!
<bentiss> mupuf: actually `set_pass(true)` disables fastly caching. We might want some cache for RSS (docs are at https://www.fastly.com/documentation/solutions/examples/rust/ and https://docs.rs/fastly/latest/fastly/index.html)
<daniels> hakzsam: hey, that one is new!
<daniels> looks like the gitlab-runner instance hosting all of Collabora's jobs caught a segfault and just dumped everything on the floor - it doesn't keep persistent state
<daniels> when that happens, the service doesn't realise that it hasn't timed out (sadly there's no keepalive ping) so we just lost them all
<daniels> sorry about that
Caterpillar has joined #freedesktop
<hakzsam> np
<mupuf> bentiss: thanks
<mupuf> daniels: niiiice
<daniels> mupuf: doubly impressive given that it's written in go
AbleBacon has quit [Quit: I am like MacArthur; I shall return.]
<mupuf> yeah! Podman sometimes crashes on corrupted imagestores...
<mupuf> daniels: one more reason to go towards the gitlab runner with job prioritization
<mupuf> sinceit doesn't oersist
<mupuf> and it is ran once per DUT
<daniels> yeah, what happened with that in the end? the WIP looked OK to me but it was still being worked on last I heard
<mupuf> daniels: it's been deployed for the general purpose runners since the migration
<mupuf> there is no enforcement of the policy though, that's for later
<daniels> yeah, we'd still quite greedily pick jobs in order to push ourselves to the front of the scheduler so we don't lose machines out to kernel testing, but at least one-shot would contain some of the damage
<daniels> either way I haven't seen gitlab-runner silently segfault in years, so hopefully it continues being a once-every-several-years problem
<mupuf> daniels: no way to limit projects to a certain percentage of the pool?
<mupuf> that would be trivial to do with a design like lava's, no?
<daniels> sure, we can designate certain DUTs as for certain projects only, but otoh if we say that KCI can only ever use 30% of the pool and Mesa can only ever use 70% of the pool then realistically we spend a lot of time with machines occupying space on a shelf doing nothing a tall
<daniels> (they're also doing KCI boot and functional testing, DRM CI, linux-media testing, etc etc)
swatish21 has quit [Ping timeout: 480 seconds]
<__tim> daniels: re. "all the external runners were broken a day or two ago but that did get fixed" how was that / what was fixed? we're still having problems uploading artefacts, getting 503 errors
<__tim> maybe that's just hetzner's S3 being on the fritz again
<karolherbst> so how is the load on our gitlab looking like now
<hakzsam> mupuf: daniels: https://gitlab.freedesktop.org/mesa/mesa/-/pipelines/1419919 it happened again
<daniels> __tim: hmm, https://gitlab.freedesktop.org/mesa/mesa/-/jobs/75923230#L1505 is totally fine from an external runner
<daniels> hakzsam: flake glob fragment_out.random..* as well pls
<daniels> __tim: can I guess that the artifacts are huge?
<__tim> yes they are
<hakzsam> daniels: what do you mean?
<daniels> hakzsam: in panfrost-t860-flakes.txt, just add fragment_out.random..*
<daniels> the prior commit looks like it only added the array group as the problematic one, but presumably some subset of the random tests pull in arrays as well, and it's not worth hardcoding specific numbers of the random seed in, so until we've fixed the array, might as well also mark the entire random group as known-flake
ds` has quit [Quit: ...]
<hakzsam> daniels: that seems unrelated to my pipeline, no? it seems to fail to download docker images
swatish2 has joined #freedesktop
ds` has joined #freedesktop
<emersion> seems like the API issue is due to a broken 301 redirect
ds` has quit [Quit: ...]
ds` has joined #freedesktop
<__tim> so frustrating that you can't even trust status.hetzner.com to know if there are outages or not
<__tim> the other day there was a 14h S3 outage and it took them half a day to even acknowledge that somewhere
mripard has quit [Quit: WeeChat 4.6.2]
JanC is now known as Guest15417
JanC has joined #freedesktop
Guest15417 has quit [Ping timeout: 480 seconds]
guludo has joined #freedesktop
JanC is now known as Guest15418
JanC has joined #freedesktop
Guest15418 has quit [Ping timeout: 480 seconds]
ds` has quit [Quit: ...]
ds` has joined #freedesktop
mvlad has joined #freedesktop
bozo16 has joined #freedesktop
haaninjo has joined #freedesktop
sentry has joined #freedesktop
<daniels> hakzsam: er? the first pipeline you posted (around 5h ago) was the one that got timed out to oblivion by a gitlab-runner crash - the second one has a single fail in t860
<hakzsam> still pending, 5h later
jsa1 has quit [Ping timeout: 480 seconds]
jsa1 has joined #freedesktop
<daniels> hakzsam: 1419919 is the first, from earlier this morning https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34859#note_2901137
<daniels> 1419977 is the second, which as you can see from Marge, reported a single failure in the T860 job which I discussed above https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34859#note_2901232
<daniels> I've cancelled the first pipeline (which would never complete) to avoid confusion, and sent https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34878 to Marge to fix the fail from the second pipeline
<hakzsam> okay, makes more sense now, thanks :)
<daniels> np
<sentry> Hey, I'm planning to do a hard-fork of a project hosted on freedesktop that has since become unmaintained and wanted to stay on freedesktop, what would be my course of action here? I already have an approve account and a repo fork but I can't run the CI with that.
swatish2 has quit [Ping timeout: 480 seconds]
<daniels> sentry: which project is it?
<sentry> switcheroo-control
<sentry> hadess had to step away and the appointed maintainer hasn't been active on the project or gotten back to my emails
<daniels> sentry: tbh, looking at it I can see one MR which had a lull in activity until last week, and the maintainer says he hasn't got any emails from you directly and doesn't see everything that hits fedora-devel@
jsa1 has quit [Ping timeout: 480 seconds]
<dj-death> in the runners
<dj-death> the artifact URL doesn't show up :
<dj-death> $ curl -L --retry 4 -f --retry-connrefused --retry-delay 30 -s "https://${JOB_RESULTS_PATH}" | tar --warning=no-timestamp --zstd -x
<dj-death> ah... there is a button :)
<Consolatis> how long is a anubis ticket valid and is it restricted to an ip address? it took my system 44 seconds to pass which is fine if it doesn't happen that often but the host part of my ipv6 address changes regularly due to privacy extension so I'd rather not have the anubis check every hour or so
<karolherbst> Consolatis: for whatever reason it's super slow only on Firefox
<karolherbst> but it works based on cookies stored in your browser
<vyivel> takes 10 seconds on my phone on average šŸ™‚ which is enough time to rethink whether i actually want to look at whatever
<Consolatis> the cookie itself seems to be valid for a week
ximion has joined #freedesktop
<emersion> it takes 30s on my phone
Kayden has quit [Quit: -> JF]
<sentry> daniels: I keep maintaining the MR so it makes sense that it is active however there is little to no interaction from others including the maintainer
<sentry> bit confused about the email thing, I double checked that it was the email I found on their github
<sentry> we are both talking about Jonas, right?
guludo has quit [Quit: WeeChat 4.6.1]
guludo has joined #freedesktop
<sentry> triple checked, its also the email they sign commits off with, unsure where it got lost
guludo has quit [Ping timeout: 480 seconds]
guludo has joined #freedesktop
AbleBacon has joined #freedesktop
tzimmermann has quit [Quit: Leaving]
alanc_work has quit []
alanc has joined #freedesktop
scrumplex_ has quit [Quit: Quassel - Signing Off]
scrumplex has joined #freedesktop
<daniels> yeah, Jonas
jsa1 has joined #freedesktop
JanC has quit [Remote host closed the connection]
<sentry> do you know where I could best reach them? email didn't work and I don't think sending a message on every avenue avilable would be nice
JanC has joined #freedesktop
ximion has quit [Remote host closed the connection]
ximion has joined #freedesktop
swatish2 has joined #freedesktop
ximion has quit [Remote host closed the connection]
ximion has joined #freedesktop
ximion has quit [Remote host closed the connection]
nephyrin has quit [Quit: ... besides, it was hot]
ximion has joined #freedesktop
ximion1 has joined #freedesktop
bozo16 has quit []
Kayden has joined #freedesktop
ximion has quit [Ping timeout: 480 seconds]
JanC is now known as Guest15436
JanC has joined #freedesktop
Guest15436 has quit [Ping timeout: 480 seconds]
<sentry> also I think I misread your message, there was no activity in the MR because I consider it functionally complete (minus bug fixes) since the beginning
guludo has quit [Ping timeout: 480 seconds]
<__tim> not really sure what to do about the large artifact upload
<__tim> feel a bit powerless here tbh :)
<__tim> and like perhaps there's something else going on, seeing that there's no sign of a major outage, and we're getting 503 errors
<__tim> do we have someone who can investigate why we're getting these errors?
<jadahl> sentry: I think you're talking about me, and I don't think I ever saw an e-mail from you. feel free to ping me here on IRC if you want though
guludo has joined #freedesktop
jsa2 has joined #freedesktop
jsa1 has quit [Ping timeout: 480 seconds]
<sentry> jadahl: I'll just summarize the email: whats the plan going forward, multi-GPU support is broken in KDE, Gnome and more, they all rely on switcheroo-control, my MR is already being shipped in Nobara Linux and Bazzite to ensure it doesn't break for users
<__tim> also, the large artifact upload failures started happening right after anubis was enabled I think, so wonder if that could be connected? Someone mentioned a 40s timeout or so earlier - that would probably be a problem if the s3 uploads go through that
guludo has quit [Ping timeout: 480 seconds]
guludo has joined #freedesktop
<__tim> it's also always the same jobs/architectures that succeed and the same that fail
<daniels> __tim: I can try to look properly next week - Iā€™m currently only half working whilst Iā€™m away and still need to set up a bunch of things to be able to look into Fastly
<__tim> thanks
<__tim> yeah I know thursday evening is not the greatest time to ask for live support :)
<jadahl> sentry: i'll try to prioritize getting it merged and into a release, sorry for all the delays. I agree things are a bit broken
nephyrin has joined #freedesktop
nephyrin has quit []
<__tim> (but thought maybe there's something easy we could try like increasing the timeout or such to see if it's that :))
mvlad has quit [Remote host closed the connection]
Traneptora has joined #freedesktop
guludo has quit [Ping timeout: 480 seconds]
johnny0 has joined #freedesktop
guludo has joined #freedesktop
sima has quit [Ping timeout: 480 seconds]
nephyrin has joined #freedesktop
nephyrin has quit [Quit: ... besides, it was hot]
nephyrin has joined #freedesktop
nephyrin has quit [Quit: ... besides, it was hot]
nephyrin has joined #freedesktop
Kayden has quit [Quit: home]
jsa2 has quit [Ping timeout: 480 seconds]
haaninjo has quit [Quit: Ex-Chat]
nephyrin has quit [Quit: ... besides, it was hot]
nephyrin has joined #freedesktop