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
anholt has quit [Ping timeout: 480 seconds]
anholt has joined #freedesktop
mohamexiety has quit []
Haaninjo has quit [Quit: Ex-Chat]
columbarius has joined #freedesktop
co1umbarius has quit [Ping timeout: 481 seconds]
BlkPoohba has quit []
BlkPoohba has joined #freedesktop
shbrngdo has quit [Ping timeout: 480 seconds]
BlkPoohba has quit []
BlkPoohba 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]
feto_bastardo has quit [Read error: Connection reset by peer]
feto_bastardo has joined #freedesktop
jarthur has quit [Ping timeout: 480 seconds]
BlkPoohba has quit []
sewn has joined #freedesktop
chip_x has joined #freedesktop
danvet has joined #freedesktop
chipxxx has quit [Ping timeout: 480 seconds]
mvlad has joined #freedesktop
<bentiss>
DavidHeidelberg[m]: still working on a hopefully better solution
chip_x has quit [Remote host closed the connection]
Haaninjo has joined #freedesktop
MajorBiscuit has joined #freedesktop
Major_Biscuit has joined #freedesktop
MajorBiscuit has quit [Ping timeout: 480 seconds]
scrumplex has joined #freedesktop
therickestrick001[m] has quit []
hasebastian[m] has quit []
inpursuitofsilence[m] has quit [Quit: Idle for 30+ days]
scrumplex_ has quit [Ping timeout: 480 seconds]
scrumplex_ has joined #freedesktop
bilboed has quit [Ping timeout: 480 seconds]
scrumplex has quit [Ping timeout: 480 seconds]
bilboed has joined #freedesktop
halfline[m] has quit []
bilboed has quit [Ping timeout: 480 seconds]
bilboed has joined #freedesktop
DodoGTA has quit [Quit: DodoGTA]
DodoGTA has joined #freedesktop
chipxxx has joined #freedesktop
<eric_engestrom>
bentiss: could you disable the #438 banner? I just realized it's still on
<eric_engestrom>
or maybe reword it as a future restriction, not one that has been applied
<bentiss>
eric_engestrom: removed
<bentiss>
eric_engestrom: and FWIW, the CI-OK group is way better.
<eric_engestrom>
thanks, confirmed
<eric_engestrom>
and that's good to hear to the ci-ok group
<bentiss>
eric_engestrom: the new script is now using that group, but is still not enforced and still allow MR pipelines to succeed. We can turn that ability off anytime
<eric_engestrom>
bentiss: for the ci-ok fetch you forgot to use $GITLAB_API, and the reason I was looking at these lines is that I think you should check .state=="active"
<eric_engestrom>
(and the id check is not necessarily useful, as it will always be true unless gitlab has a bug making it return something other than what was requested, but it's harmless so why not)
<bentiss>
eric_engestrom: thanks for the GITLAB_API
<bentiss>
.state==active -> the user you mean?
<eric_engestrom>
yeah
<bentiss>
re id check: given that we can overwrite a bunch of predefined variables, all the only guarantee gitlab gives us is "this might make your pipeline behave in an unexpectecd way", I'd rather trust the job token's job output
<bentiss>
(what a terrible name "job's token job")
<eric_engestrom>
oh yeah I agree with that
<eric_engestrom>
it's the `if [ x"$CI_OK_USER_ID" = x"$GATING_USER_ID" ]` line I was taking about
<eric_engestrom>
since you're fetching `?user_ids=${GATING_USER_ID}`, it shouldn't be possible to get anything else
<bentiss>
for users not part of the group, we'll get an empty string
<eric_engestrom>
because of the .[]
Haaninjo has quit [Quit: Ex-Chat]
<bentiss>
yeah, we get eith '[]' or '[lots-of-json-stuff]'
<bentiss>
so '.[].id' returns the ID we queried if it was in the group
<eric_engestrom>
it can only be of length 0 or 1, right? doesn't [0] raise an error if it's empty?
* eric_engestrom
hasn't used jq in a while
<bentiss>
by default, nope, it returns null,
<eric_engestrom>
just tested locally, indeed
<bentiss>
but we should be able to change that
<eric_engestrom>
ok then, lgtm
<bentiss>
eric_engestrom: jq -e '.[0]' works as intended
<eric_engestrom>
ah yes, I thought I did use that [0] -> error thing at some point in the past
<eric_engestrom>
so it is possible, just not the default behaviour
<eric_engestrom>
oh also, maybe we should enforce a minimum .access_level>=30 (developer)?
<bentiss>
not a big fan of access level currently
<bentiss>
well, you can always ask for somebody else to trigger the pipeline
<bentiss>
but we might want to take it slow...
<eric_engestrom>
yeah you're right, let's not add too many restrictions until it's necessary
AbleBacon has quit [Read error: Connection reset by peer]
<mupuf>
bentiss: looks good! I highlighted the "get yourself added as a guest" part
<mupuf>
actually, let me reword this section, it is too confusing
chip_x has joined #freedesktop
chip_x has quit [Max SendQ exceeded]
chip_x has joined #freedesktop
<daniels>
bentiss: looks good here too, thanks!
<mupuf>
actually, no, I can't edit it because some information seems contradictory :s
<mupuf>
it likely is a question of timeline
<mupuf>
bentiss: can you clarify in which cases adding someone as guest would be needed?
<mupuf>
and when?
chipxxx has quit [Ping timeout: 480 seconds]
<bentiss>
mupuf, daniels: thanks
<bentiss>
mupuf: hard to clarify when it's still not clear here too :)
<mupuf>
Right
chipxxx has joined #freedesktop
<bentiss>
basically, we should have "I can't run CI, can you add me as a guest?" -> if still not working "hello dear admins, there is a problem here"
<bentiss>
I haven't added all groups to ci-ok, and I already know there are issues with people directly added to projects in the group, so that's why it's fuzzy
<bentiss>
but I think we might want to have subgroups like wyaland-ci-ok without projects that would be maintained by group maintainers
chipxxx has quit [Max SendQ exceeded]
<bentiss>
once we have that in place, I'll remove the MR pipeline workaround I think
<emersion>
eric_engestrom: maybe time to remove the GitHub repos?
<ishitatsuyuki>
sad to hear since I was relying on github code search, but ok
<emersion>
ishitatsuyuki: there is GitLab code search
<ishitatsuyuki>
I'm aware, it's not a perfect substitution for me
<emersion>
fwiw, in my experience, GitHub code search has always been inferior
<emersion>
it's hiding results when there are multiple in a file, for instance
tintou has joined #freedesktop
<tintou>
Not in mine really, not being able to filter by file type is useless for me
<eric_engestrom>
emersion: yeah, it's on my to-do list and way past the 1 month deadline from the email; I just haven't had time to to do it yet (and also I'm not admin on github but that can be fixed when needed)
<emersion>
i'll delete the wayland ones
<eric_engestrom>
thanks!
<eric_engestrom>
fyi the mesa one needs to stay (and I'll set up mirroring from gitlab), but I haven't heard anything back about all the other ones
<eric_engestrom>
that is, gh/mesa3d/mesa needs to stay, but I think there are also mirrors of mesa under gh/freedesktop/...
<eric_engestrom>
daniels: are you also an owner of https://github.com/orgs/freedesktop ? if so, can you add me so that I can go through all the projects and delete the ones that have no activity (just in case someone does use github but didn't reply to the email)
strugee has quit [Ping timeout: 480 seconds]
BlkPoohba has quit []
chipxxx has quit [Remote host closed the connection]
Leopold_ has quit []
Leopold has joined #freedesktop
damian has joined #freedesktop
<daniels>
eric_engestrom: what’s your username?
<eric_engestrom>
daniels: 1ace
<eric_engestrom>
btw I probably won't get around to doing it until next week, so no need to rush
<bentiss>
daniels: FWIW I accepted the invite but do not plan on doing anything on github ;)
<eric_engestrom>
also, I just realized that from github's point of view, I'll look like a defacer: just after I'm added to an org, I proceed to go around and delete everything 🙃
<MrCooper>
maybe you should rename your account to something like grim_reaper1 :P
agd5f_ has joined #freedesktop
cazzacarna has quit [Remote host closed the connection]
agd5f has quit [Ping timeout: 480 seconds]
agd5f has joined #freedesktop
BlkPoohba has joined #freedesktop
agd5f_ has quit [Ping timeout: 480 seconds]
MrCooper has quit [Remote host closed the connection]
MrCooper has joined #freedesktop
vkareh has joined #freedesktop
DodoGTA has left #freedesktop [#freedesktop]
mohamexiety has joined #freedesktop
DodoGTA has joined #freedesktop
<eric_engestrom>
haha
<daniels>
eric_engestrom: heh! anyway, done
Kayden has quit [Quit: office]
<eric_engestrom>
daniels: ta
DodoGTA has left #freedesktop [#freedesktop]
thaller has quit [Ping timeout: 480 seconds]
Kayden has joined #freedesktop
DodoGTA has joined #freedesktop
DodoGTA has left #freedesktop [#freedesktop]
miracolix has joined #freedesktop
DodoGTA_ has joined #freedesktop
gallo72 has quit []
gallo72 has joined #freedesktop
hadess has quit [Remote host closed the connection]
hadess_ has joined #freedesktop
__tim has quit [Read error: Connection reset by peer]
__tim has joined #freedesktop
jarthur has joined #freedesktop
ryanpavlik has quit [Remote host closed the connection]
ryanpavlik has joined #freedesktop
JoshuaAs- has quit [Ping timeout: 480 seconds]
JoshuaAshton has joined #freedesktop
scrumplex has joined #freedesktop
scrumplex_ has quit [Ping timeout: 480 seconds]
AbleBacon has joined #freedesktop
Major_Biscuit has quit [Ping timeout: 480 seconds]
Major_Biscuit has joined #freedesktop
Major_Biscuit has quit [Ping timeout: 480 seconds]
alanc has quit [Remote host closed the connection]
alanc has joined #freedesktop
<alatiera>
could I have another shared runner token heh
<alatiera>
the previous one complains now
<alatiera>
on the positive side I think I got podman in user mode working
<DavidHeidelberg[m]>
gitlab frontend bit overloaded?
GNUmoon has quit [Read error: Connection reset by peer]
GNUmoon has joined #freedesktop
Haaninjo has joined #freedesktop
___nick___ has joined #freedesktop
vkareh has quit [Quit: WeeChat 3.6]
mvlad has quit [Remote host closed the connection]
DodoGTA_ has quit [Remote host closed the connection]
DodoGTA has joined #freedesktop
___nick___ has quit []
___nick___ has joined #freedesktop
<i-garrison>
hope not overloaded by guys spamming pulseaudio each 5 min approx ))
___nick___ has quit []
___nick___ has joined #freedesktop
<alatiera>
i-garrison if that was all that our "spammers" did we would be good 😆
<alatiera>
umm
<alatiera>
I may have accidentally nuked the config file of gst-2 instead of appending to it
<alatiera>
(once an admin is awake tmr, I will need a shared runner token to re register, maybe send it to __tim too for good measure and avoid bus factors)
* bentiss
forwarded a runner token to alatiera, so that should be it for that
<alatiera>
life saver
___nick___ has quit [Ping timeout: 480 seconds]
danvet has quit [Ping timeout: 480 seconds]
BlkPoohba has quit []
Kayden has quit [Quit: -> home]
Kayden has joined #freedesktop
emersion has quit [Remote host closed the connection]
emersion has joined #freedesktop
<alatiera>
I was looking at the runner config docs in the toml and I have a couple of interesting finds
<alatiera>
(will play with them tmr, but worth sharing already)
<alatiera>
cap_add/cap_drop/security_opt are useful for getting read of the privileged mode while also allowing a handful of syscalls to make certain things work
<alatiera>
helper_image_flavor might be useful if we port the gate script into python, we can switch to something that has py3 by default in case alpine doesn't
<alatiera>
user and userns might be exactly what I was looking for to fix the root_by_default docker images
<alatiera>
though no idea what happens if the user declared doesn't exist already in the container, but maybe we could require it from the images or add it at runtime with some pre-script
<alatiera>
container_labels can add labels to each container the runner will create, maybe we can use it for metrics or more aggressive garbage collecting on the runners
<alatiera>
allowed_services we should probably set to [] so nobody tries to use them, I think they require dind anyway
<alatiera>
and I am also curious to see how cpus_ keys behave, I recall we looked at them before when trying to reduce the impact of jobs on the runners but I don't recall what the issues were
<alatiera>
there's a sysctl options key as well, it could be useful but I can't think of anything we could use it for atm
mohamexiety has quit []
emersion has quit [Remote host closed the connection]