daniels 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]
AbleBacon has joined #freedesktop
martink has quit [Remote host closed the connection]
thelounge1473 has quit []
martink has joined #freedesktop
thelounge1473 has joined #freedesktop
itaipu has quit [Ping timeout: 480 seconds]
itaipu has joined #freedesktop
peelz has joined #freedesktop
peelz is now known as Guest9010
Guest9010 is now known as peelz
sentriz has quit [Read error: Connection reset by peer]
sooc has joined #freedesktop
sentriz has joined #freedesktop
konstantin_ has joined #freedesktop
konstantin has quit [Ping timeout: 480 seconds]
CrEddy has joined #freedesktop
konstantin has joined #freedesktop
konstantin_ has quit [Ping timeout: 480 seconds]
bmodem has joined #freedesktop
ximion has quit [Quit: Detached from the Matrix]
damian has quit [Ping timeout: 480 seconds]
damian has joined #freedesktop
privacy has quit [Quit: Leaving]
tzimmermann has joined #freedesktop
sima has joined #freedesktop
<pinchartl>
I've just seen a CI job failing with
<pinchartl>
ERROR: Job failed: failed to pull image "registry.freedesktop.org/pinchartl/libcamera/debian/bookworm:2023-12-04.1" with specified policies [if-not-present]: writing blob: adding layer with blob "sha256:af3565a9b15738ee019e0f2586e32cdb813ce6df3e84991ffe361c3371081d6e": layer not known (manager.go:237:19s)
<pinchartl>
retrying it worked
<pinchartl>
is it normal ?
thaller has joined #freedesktop
<MrCooper>
pinchartl: does that job depend on the job which makes sure the image tag exists?
<pinchartl>
MrCooper: yes, it does
<pinchartl>
I have multiple jobs that depend on the job which builds the container, they all failed with the same error
<pinchartl>
it's the first time I see this
mripard has joined #freedesktop
bochecha has joined #freedesktop
AbleBacon has quit [Read error: Connection reset by peer]
mvlad has joined #freedesktop
i509vcb has quit [Quit: Connection closed for inactivity]
MrCooper has quit [Remote host closed the connection]
MrCooper has joined #freedesktop
alpernebbi has quit [Remote host closed the connection]
alpernebbi has joined #freedesktop
Haaninjo has joined #freedesktop
lynxis has quit [Quit: No Ping reply in 180 seconds.]
lynxis has joined #freedesktop
lynxis has quit []
lynxis has joined #freedesktop
lynxis has quit []
lynxis has joined #freedesktop
lynxis has quit [Quit: No Ping reply in 180 seconds.]
lynxis has joined #freedesktop
lynxis has quit [Quit: No Ping reply in 180 seconds.]
lynxis has joined #freedesktop
privacy has joined #freedesktop
lynxis has quit [Quit: No Ping reply in 180 seconds.]
lynxis has joined #freedesktop
ximion has joined #freedesktop
lynxis has quit []
lynxis has joined #freedesktop
lynxis has quit [Quit: No Ping reply in 180 seconds.]
lynxis has joined #freedesktop
ximion has quit [Quit: Detached from the Matrix]
blatant has joined #freedesktop
blatant has quit [Ping timeout: 480 seconds]
mripard has quit [Remote host closed the connection]
blatant has joined #freedesktop
<bochecha>
hi, I'm having a small issue in gitlab.fd.o, sometimes my test suite fails anc thetest runner tells me "See ./test-suite.log" except that file doesn't exist :-/
<bochecha>
if I run the same tests manually on my host I do have that file, but they only fail in some of the CI envs
<bochecha>
how can I get access to the test suite logs to get some more info on the actual failures?
<pinchartl>
daniels: is there a way to have different artifacts for the failure and success cases ? it doesn't seem possible, just wanted to cinform
<daniels>
pinchartl: not ttbomk - you'd have to use `when: always` and then rm the bits you don't want to keep within the execution script
blatant has quit [Quit: WeeChat 4.1.1]
<bochecha>
@daniels, thank you, I'm already capturing '*.log' in my artifacts declaration :-/
<pinchartl>
daniels: thanks
<pinchartl>
I was surprised to see that generating multiple artificats from one job wasn't supported. it seems useful, and quite frankly it's a basic features I would have expected gitlab to support from the beginning
<pinchartl>
(same with the ability to comment on commit messages during review...)
<bochecha>
commenting on commit messages during a review, there's an issue for that that never went anywhere :(
<pinchartl>
yes, I know
<pinchartl>
and github doesn't have the feature either
<pinchartl>
almost as if most people didn't care about commit messages
<bochecha>
@daniels, in my case the CI captured `config.log`, which is not at all what I wanted, it didn't caapture the test-suite.log, which might be deleted by the time the artifacts thing happens :(
<pinchartl>
that can't be the case, can it ? :-)
<bochecha>
lot's of people do care, just not the devs in charge of those forges I guess
<daniels>
bochecha: maybe copy test-suite.log to something which won't be deleted then?
<pinchartl>
I have a theory that the lack of that feature is partly responsible for the abysmal quality of commit messages in most projects. just a theory
<bochecha>
how do I do that? :-/
<bochecha>
I'm running the tests as follows: - make distcheck || cat ./test-suite.log
<daniels>
bochecha: without knowing your project, I don't know, but it sounds like your script does something which generates test-suite log, followed by something that deletes test-suite.log, so it can't be archived at the end because it's gone. so the options seem to be either a) make it not delete test-suite.log, or b) insert something between the generation step and the deletion step to copy it to a path which isn't deleted?
<bochecha>
so that when the tests fail I print the contents of the file, but nothing ever gets printed :(
<bochecha>
well, all I know is "cat: ./test-suite.log: No such file or directory"
<bochecha>
so that might indicate that when the "make distcheck" finiches it deletes the log file :(
<bochecha>
but when I run distcheck on my local computer, that file still exists at the end :(
<pinchartl>
bochecha: I don't know if it's an option in this specific case, but you can download and run the CI container locally to run the job manually, step by step
<bochecha>
can I? :o
<bochecha>
that would be amazing!!
<pinchartl>
copy the link to the container from the gitlab container registry page
<pinchartl>
and run
<pinchartl>
podman container run -it $container_registry_url bash
<pinchartl>
that will give you an interactive shell in the container
<pinchartl>
then you can run the job's script by hand, typing the commands manually
<pinchartl>
and see what happens
<bochecha>
there is nothing in the container registry page :(
<bochecha>
ah, there are commands, should I run those?
<bochecha>
doesn't work, when I run the second command listed on that page: docker build -t registry.freedesktop.org/cangjie/ibus-cangjie .
<bochecha>
I get:
<bochecha>
base name (${BASE_IMAGE}) should not be blank
flom84 has joined #freedesktop
<bochecha>
yeah, it still says base name should not be blanked, even after I exported all those env vars in the current shell :(