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
danvet has quit [Ping timeout: 480 seconds]
alanc has quit [Remote host closed the connection]
Leopold_ has quit []
genpaku has quit [Remote host closed the connection]
genpaku has joined #freedesktop
kattupoochi has joined #freedesktop
<nirbheek> Is there some outage going on? I get this when I try to do a git fetch:
<nirbheek> > rpc error: code = Unavailable desc = The git server, Gitaly, is not available at this time. Please contact your administrator.
<nirbheek> Also job pages load and then disappear with an error
<nirbheek> > An error occurred while fetching the job.
<SanchayanMaity> I see the same here when using SSH urls. HTTPs urls seem to work.
kattupoochi has left #freedesktop [#freedesktop]
ximion has quit []
itoral has joined #freedesktop
<bilboed> and getting 500 when trying to post an issue also
<bentiss> we have one server down, not sure what happened, rebooting it now
<bentiss> (and on that server was one gitaly pod, which explains the git issues)
<bentiss> should be back-ish
SanchayanMaity has quit []
SanchayanMaity has joined #freedesktop
pjakobsson has joined #freedesktop
Satan3 has quit [Ping timeout: 480 seconds]
miracolix has quit []
Satan3 has joined #freedesktop
ofourdan has joined #freedesktop
danvet has joined #freedesktop
rgallaispou has joined #freedesktop
<bentiss> FWIW, I still have a few services not properly binding to their IPs (not in gitlab AFAICT), so I'm going to upgrade and reboot all servers, it won't hurt to do this
scrumplex_ has quit []
scrumplex has joined #freedesktop
<nirbheek> Just FYI getting this now, please ignore if this is expected :)
<nirbheek> > Unable to load the merge request widget. Try reloading the page.
<bentiss> nirbheek: yeah, I am currently upgrading/rebooting all servers one by one, so kind of expected
<nirbheek> 👍️
<Venemo> good morning, gitlab gives me a HTTP 502
<bentiss> Venemo: yep, known, see above, I am upgrading/rebooting all servers and it takes a little bit of time
<Venemo> it works now
<Venemo> bentiss: thanks. I wasn't sure if this was the same issue you had a few hours ago or a new problem
<bentiss> Venemo: technically, it's the continuation, because I have at least one service on ceph not happy, so I figured I should flush all servers
<bentiss> (this was way overdue FWIW)
<Venemo> okay
<Venemo> thanks for taking care of it :)
<bentiss> no worries
ximion has joined #freedesktop
<eric_engestrom> bentiss: are you doing something on gitlab? it's not loading anymore :'(
<bentiss> eric_engestrom: yes, rebooting the last server of the flock
<eric_engestrom> ok, thanks :)
<bentiss> and gitaly-3 is not happy, as it was on that server and the mount poiont had issues migrating to other servers
<eric_engestrom> ah, so possibly not just a couple of minutes of down time?
<bentiss> roughly
<bentiss> roughly a couple of minutes I would say
<eric_engestrom> ok
<bentiss> the machine is still in the BIOS
<bentiss> gitaly-3 is starting up
<bentiss> eric_engestrom: should be OK for you now
<eric_engestrom> yeah, it loaded but it also logged me out, not sure if that's a normal side effect of what you did
<bentiss> eric_engestrom: probably redis being wiped out
<eric_engestrom> logged back in all fine, everything works
<eric_engestrom> thanks!
<bentiss> great :)
<eric_engestrom> oh btw for the CI runner tags I mentioned a couple of days ago, we decided we don't need to filter like that anymore, so we don't need those tags anymore
<daniels> eric_engestrom: oh that's easy, I can close that tab then
Haaninjo has joined #freedesktop
<eric_engestrom> daniels: not sure what you mean ^^
<daniels> eric_engestrom: I meant that I was about to go through and add the tags to all the different devices, but now I don't have to
Haaninjo has quit [Quit: Ex-Chat]
Haaninjo has joined #freedesktop
<eric_engestrom> ah, thanks for that!
<eric_engestrom> and indeed it's not needed anymore
itoral has quit []
MajorBiscuit has joined #freedesktop
mvlad has joined #freedesktop
ybogdano has joined #freedesktop
Leopold_ has joined #freedesktop
pjakobsson has quit [Ping timeout: 480 seconds]
pjakobsson has joined #freedesktop
<tomeu> anholt: something seems to be wrong with the swrast runners other than #2
<tomeu> venus-lavapipe jobs time out
<tomeu> I think daniels had to blacklist some vmware-related kernel modules to prevent vsock (or vhost?) from stopping working properly after some time
<tomeu> sergi: ^
<eric_engestrom> DavidHeidelberg[m], daniels: what tool produces reports like https://gitlab.freedesktop.org/mesa/mesa/-/issues/7180 ?
<eric_engestrom> we might be interested in running that internally to track the health of our own CI
<eric_engestrom> (it's really heavy with everything at once)
<tomeu> see the pipelines in that repo
<daniels> tomeu: iirc that was terminal to vsock working at all, not just introducing failures - the vmw stuff greedily occupied vsock and the normal one didn't load
<eric_engestrom> tomeu: thanks!
<tomeu> hrm
<tomeu> wonder if something similar about vmware stuff interfering with normal work of upstream vhost stuff
<tomeu> as that is a host resource that all VMs have in common
<tomeu> could reproduce like that
<daniels> looking at https://gitlab.freedesktop.org/sergi/virglrenderer/-/jobs/31623002 I'm seeing a ton of venus tests failing with a crash, and only towards the end do we start getting prints about guessing the syslog fd failing - I wonder if there's something like an FD leak or space exhaustion after enough tests have failed
MrCooper has quit [Remote host closed the connection]
MrCooper has joined #freedesktop
<robclark> bentiss: I tried to update minio server for drm-ci.. but I guess I'm missing something? https://gitlab.freedesktop.org/drm/msm/-/jobs/31636346#L11400
<bentiss> robclark: yes, you should strip out any `ci-fairy minio login` and replace `ci-fairy minio cp minio://` by `ci-fairy s3cp https://`
<robclark> fwiw, this is what I have, adapted from mesa commit.. but maybe missing something.. https://gitlab.freedesktop.org/robclark/drm-ci/-/commit/3c9efb5b7c4ea2c83b058898cb5eb386085f1341
<robclark> oh, I guess I needed more than just the toplevel .gitlab-ci.yml
<bentiss> robclark: seems correct, but your job is trying to pull minio login
<robclark> bentiss: hmm, fixed up the missing conversions.. but it doesn't know what s3cp is?? https://gitlab.freedesktop.org/drm/msm/-/jobs/31637754#L1779
<bentiss> robclark: you need to use the latest ci-fairy image, and remove the `minio` part in your command. It's not minio anymore
<robclark> ahh
MajorBiscuit has quit [Ping timeout: 480 seconds]
AbleBacon has joined #freedesktop
alanc has joined #freedesktop
miracolix has joined #freedesktop
<robclark> is this the right thing to get a suitably new enough version of ci-fairy? Because it doesn't seem to work.. https://gitlab.freedesktop.org/drm/msm/-/jobs/31640335#L1726
jarthur has quit [Ping timeout: 480 seconds]
<bentiss> robclark: it should be enough yeah, wondering if you have another installed ci-fairy in the current image
<robclark> it's possible.. keep in mind I don't know what I'm doing ;-)
<bentiss> robclark: also s/minio:/https:/
<bentiss> (in your s3cp command)
<robclark> oh, I missed that in trying to port over the mesa change
jarthur has joined #freedesktop
miracolix has quit []
ybogdano has quit [Ping timeout: 480 seconds]
ybogdano has joined #freedesktop
<eric_engestrom> tomeu: mesa_ci_status.py is eating almost 4.5 GB of ram and counting, and it printed "Processing pipelines pages 1..4" at the start but hasn't printed anything in over half an hour
miracolix has joined #freedesktop
<eric_engestrom> (I ran it with no arguments, default everything)
<eric_engestrom> I'm assuming this isn't normal, right? ^^'
Leopold_ has quit []
<robclark> bentiss: hmm, I guess /usr/local isn't where the pip installed ci-fairy would go? https://gitlab.freedesktop.org/drm/msm/-/jobs/31644647#L1751
Leopold_ has joined #freedesktop
<bentiss> robclark: I would have said /usr/local is the target path... maybe check if there is any ci-fairy installed before running pip install
<bentiss> (I'll be afk a bit for diner)
<anholt> tomeu: sorry, is there something specific you want me to go look for?
<robclark> bentiss: ahh, yes, there is a version installed prior..
<robclark> from *somewhere*.. maybe I didn't bump some container label or something like that
rsripada has joined #freedesktop
Kayden has quit [Quit: to office & lunch?]
danvet has quit [Ping timeout: 480 seconds]
Leopold_ has quit []
ybogdano has quit [Ping timeout: 480 seconds]
ybogdano has joined #freedesktop
Leopold_ has joined #freedesktop
mvlad has quit [Remote host closed the connection]
Leopold_ has quit [Remote host closed the connection]
Leopold_ has joined #freedesktop
Kayden has joined #freedesktop
Haaninjo has quit [Quit: Ex-Chat]
dakr has quit [Quit: ZNC 1.8.2+deb2 - https://znc.in]
dakr has joined #freedesktop
pzanoni has joined #freedesktop