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
sima has quit [Ping timeout: 480 seconds]
soaffine has joined #freedesktop
affinetyped has joined #freedesktop
soaffine has quit []
affinetyped has quit []
soaffine has joined #freedesktop
scrumplex_ has joined #freedesktop
scrumplex has quit [Ping timeout: 480 seconds]
dri-logger has joined #freedesktop
marcheu_ has joined #freedesktop
dri-logg1r has quit [Ping timeout: 480 seconds]
marcheu has quit [Ping timeout: 480 seconds]
dri-logg1r has joined #freedesktop
dri-logger has quit [Read error: Connection reset by peer]
marcheu_ has quit [Ping timeout: 480 seconds]
dri-logger has joined #freedesktop
zerozerozero has joined #freedesktop
dri-logg1r has quit [Ping timeout: 480 seconds]
marcheu has joined #freedesktop
zerozero1 has quit [Ping timeout: 480 seconds]
soaffine has quit [Ping timeout: 480 seconds]
strugee has quit [Remote host closed the connection]
kode547 has joined #freedesktop
kode54 has quit [Ping timeout: 480 seconds]
strugee has joined #freedesktop
eluks has quit [Remote host closed the connection]
eluks has joined #freedesktop
jturney has quit [Ping timeout: 480 seconds]
ity has quit [Remote host closed the connection]
ity has joined #freedesktop
swatish2 has joined #freedesktop
kode547 has quit []
kode54 has joined #freedesktop
<kode54> the lack of SASL is really shitty
<kode54> then again, so is my client that replaces the concept of an alternate nickname field or two with just appending a random number digit and calling it a day
jsa1 has joined #freedesktop
<dwfreed> kode54: fixing the sasl issue is on the roadmap
<kode54> sorry
<dwfreed> unfortunately we're short on people available to do the work
<kode54> thanks, I hope whoever gets it working doesn't have too hard a time
<kode54> I understand
<dwfreed> well, getting it working is "moving to solanum"
<dwfreed> it's a whole process
<kode54> aaaaaa
<kode54> good luck
<dwfreed> basically rebuilding an IRC network without anyone noticing is a lot harder when you don't have a Korean prince to sell out to
<kode54> indeed
<kode54> but the result is still a lot better, because it didn't involve selling out to a Korean prince
<dwfreed> hehe, yeah
<kode54> one could say he had musk beaten to the punch on acquiring and destroying a social network
<dwfreed> rofl
swatish21 has joined #freedesktop
swatish2 is now known as Guest12248
swatish21 is now known as swatish2
jturney has joined #freedesktop
Guest12248 has quit [Ping timeout: 480 seconds]
sghuge has quit [Remote host closed the connection]
sghuge has joined #freedesktop
jsa1 has quit [Ping timeout: 480 seconds]
zerozerozero has quit [Ping timeout: 480 seconds]
TrinitronX has quit [Ping timeout: 480 seconds]
swatish21 has joined #freedesktop
TrinitronX has joined #freedesktop
swatish2 has quit [Ping timeout: 480 seconds]
tzimmermann has joined #freedesktop
Hoolootwo has quit [Quit: ZNC 1.8.2+deb2+deb11u1 - https://znc.in]
sima has joined #freedesktop
jsa1 has joined #freedesktop
Hooloovoo has joined #freedesktop
<alatiera> bentiss re: plugins-rs volume size. I am almost sure the issue is with gitlab not cleaning up the volumes even with cache volumes disabled in the runners. I will add an afterscript nuking build artifacts as a workaround
<bentiss> alatiera: I've enabled the config I mentioned above, and I hadn't had to restart the runners yet, so the config is working
<bentiss> it's just that by default, gitlab runner keeps all containers states, and this doesn't scale unless you have TBs of disks
<bentiss> alatiera: TL;DR: it should be fixed
<alatiera> oh cool thanks
<alatiera> I've also seen disable_cache not work in the past as well fyi
<bentiss> k, well, it seems to be OK now :)
<alatiera> I've often cleanup volume by hand at the end of jobs just to be sure
<alatiera> anyhow feel free to shout if you need anything
<bentiss> alatiera: thanks!
Hooloovoo has quit [Ping timeout: 480 seconds]
jsa1 has quit [Ping timeout: 480 seconds]
Hooloovoo has joined #freedesktop
Kayden has quit [Quit: Leaving]
Kayden has joined #freedesktop
jsa1 has joined #freedesktop
swatish2 has joined #freedesktop
sailus has quit [Ping timeout: 480 seconds]
swatish21 has quit [Ping timeout: 480 seconds]
sailus has joined #freedesktop
AbleBacon has quit [Read error: Connection reset by peer]
swatish2 has quit [Ping timeout: 480 seconds]
cadubentzenbackThu2703[m] is now known as cadubentzen[m]
swatish2 has joined #freedesktop
swatish21 has joined #freedesktop
swatish2 has quit [Ping timeout: 480 seconds]
Hazematman has quit [Quit: WeeChat 4.5.1]
shagant[m] has joined #freedesktop
Hazematman has joined #freedesktop
Hazematman has quit []
Hazematman has joined #freedesktop
larunbe has quit []
alarumbe has joined #freedesktop
shagant[m] has left #freedesktop [User left]
swatish2 has joined #freedesktop
swatish21 has quit [Ping timeout: 480 seconds]
guludo has joined #freedesktop
swatish21 has joined #freedesktop
swatish2 has quit [Ping timeout: 480 seconds]
ity has quit [Remote host closed the connection]
dcunit3d has joined #freedesktop
ity has joined #freedesktop
ity1 has joined #freedesktop
ity has quit [Ping timeout: 480 seconds]
swatish21 has quit [Ping timeout: 480 seconds]
swatish2 has joined #freedesktop
ity1 has quit [Remote host closed the connection]
alarumbe has quit [Remote host closed the connection]
alarumbe has joined #freedesktop
ity has joined #freedesktop
<bentiss> eric_engestrom, daniels, mupuf: regarding gitlab_runner_priority.py, I just realized that if we want to implement it that way, we need per server 12 * (priority count) registered runners :/
swatish2 has quit [Ping timeout: 480 seconds]
<mupuf> why 12?
<bentiss> we split the runner by cpu_sets to present only 8 cpus
<bentiss> actually... if the `run-single` command takes args, we could make the cpu_sets dynamic
<bentiss> so only priority_count runners
<bentiss> eric_engestrom: I have a couple of questions on the gitlab_runner_priority script: 1. does it actually run untagged jobs? and 2. why ThreadPoolExecutor(len(duts_queues)) and not ThreadPoolExecutor(config["concurrent"]) ?
<mupuf> it may have been designed with bare metal in mind
<bentiss> well, I'm on bare metal ATM :)
<mupuf> lol
<bentiss> eric_engestrom: ok, I think I got my answers: 1. every registered runners is also running untagged job (which can be fixed by changing the parameters FWIW), and 2. well, that should work :)
airlied_ has joined #freedesktop
airlied has quit [Ping timeout: 480 seconds]
<eric_engestrom> bentiss: I haven't looked at that code in forever, let me dive back into it
<bentiss> eric_engestrom: that's fine, I'll probably amend what I feel is necesssary
<eric_engestrom> for the question about untagged jobs:
<eric_engestrom> --run-untagged Register to run untagged builds; defaults to 'true' when 'tag-list' is empty [$REGISTER_RUN_UNTAGGED]
<eric_engestrom> since we always set the priority tag and don't pass that flag, that means untagged jobs don't get picked up
<bentiss> yeah, I've added a "none" priority in my deploy script, and this will run untagged jobs
<eric_engestrom> (I don't remember whether I tested to verify though)
<eric_engestrom> ack
<bentiss> so we'll have "high", "medium", "none" <- untagged, "low"
<eric_engestrom> sounds good
<bentiss> thanks a lot for that because it'll simplify the registration of the runners as well. As before I had to register one per cpuset, no I can dynamically allocate them :)
<eric_engestrom> :)
<zmike> are enough runners active to effectively merge things today or no
<eric_engestrom> I assume you're copying the relevant bits out of my script and into yours, right? can you link to your script?
<eric_engestrom> zmike: I think it should work, yeah
<bentiss> I copied/paste the file in my tree
<eric_engestrom> zmike: it's been working fine for the last few hours
<bentiss> then I'll amend, and send back a MR I guess
<eric_engestrom> bentiss: ack, thanks!
<zmike> okay, will try
airlied has joined #freedesktop
airlied_ has quit [Ping timeout: 480 seconds]
zerozerozero has joined #freedesktop
jstntime[m] has joined #freedesktop
jsa1 has quit [Ping timeout: 480 seconds]
haaninjo has joined #freedesktop
tzimmermann has quit [Quit: Leaving]
guludo has quit [Quit: WeeChat 4.5.2]
<bentiss> \o/ all the equinix x86 runners are now paused and replaced by Hetzner ones
<bentiss> we can try experimenting with the priority:high / medium / low also :)
guludo has joined #freedesktop
slavka has joined #freedesktop
<bentiss> FWIW, unless there is an issue over the night, I'll release the dedicated runners at equinix tomorrow I think
<bentiss> I guess dropping the tmpfs parameter was not a good idea
Guest12276 has joined #freedesktop
Guest12276 has quit []
<f_> \o/
Kayden has quit [Quit: -> JF]
<__tim> I'm seeing lots of weird failures on runners with `priority: none` suffix, e.g. https://gitlab.freedesktop.org/slomo/gst-plugins-rs/-/jobs/73472383
<__tim> usually works when retrying and it ends up on a "fdo-gitlab-runner-coordinator-x86-X" runner
<__tim> (just for information, sounds like you're already aware of issues)
<bentiss> __tim: yeah, see above :(
alanc has quit [Remote host closed the connection]
<bentiss> tmpfs re-enabled, but I wonder if there are no other issues...
alanc has joined #freedesktop
<DemiMarie> Why did FreeDesktop choose to go with Hetzner instead of colocation in a datacenter somewhere?
<mupuf> DemiMarie: admin time is valuable
<alanc> DemiMarie: https://gitlab.freedesktop.org/freedesktop/freedesktop/-/issues/2011 tracked the decision making process
jsa1 has joined #freedesktop
todi1 has quit [Remote host closed the connection]
bozo16 has joined #freedesktop
zerozero1 has joined #freedesktop
zerozero1 has quit []
zerozero1 has joined #freedesktop
zerozero1 has quit []
zerozero1 has joined #freedesktop
zerozero1 has quit []
zerozero1 has joined #freedesktop
zerozerozero has quit [Ping timeout: 480 seconds]
zerozero1 has quit [Quit: WeeChat 4.5.2]
zerozerozero has joined #freedesktop
ofourdan has quit [Remote host closed the connection]
ofourdan has joined #freedesktop
zerozero1 has joined #freedesktop
zerozero1 has quit []
zerozero1 has joined #freedesktop
zerozero1 has quit []
zerozerozero has quit [Ping timeout: 481 seconds]
<DemiMarie> alanc mupuf: makes sense
AbleBacon has joined #freedesktop
jsa1 has quit [Ping timeout: 480 seconds]
todi has joined #freedesktop
gshumway_ has quit [Ping timeout: 480 seconds]
gshumway has joined #freedesktop
jsa1 has joined #freedesktop
sima has quit [Ping timeout: 480 seconds]
<daniels> bentiss: how do I push new config for runners? I've got a change which disables coredumps since they're crazy slow, but I'm not 100% sure how I should get that into the fleet
<daniels> I've pushed that to the hetzner branch now
haaninjo has quit [Quit: Ex-Chat]
<bentiss> daniels: fleeting or bare metal?
jljusten has left #freedesktop [WeeChat 4.5.1]
<bentiss> for fleeting, the easiest for now is to ssh to the coordinator, do the config change manually and systemctl restart gitlab-runner
<bentiss> yep, for your change, the easiest is the manual change
<bentiss> I can do that if you want
<bentiss> all 3 updated and in the process of restarting
zerozerozero has joined #freedesktop
jsa1 has quit [Ping timeout: 480 seconds]
guludo has quit [Ping timeout: 480 seconds]
zerozero1 has joined #freedesktop
zerozero1 has quit []
zerozerozero has quit [Ping timeout: 480 seconds]