You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
$ lab ci trace --help
If a job is not specified the latest running job or last job in the pipeline is used
Usage:
lab ci trace [remote [[branch:]job]] [flags]
Aliases:
trace, logs
Flags:
-h, --help help for trace
the idea with branch:job was to emulate how git lets do stuff like git checkout origin branch:file -- A simple extension would just be to keep the colons going lab ci artifacts origin master:myjob:path/to/my/file, but mostly I'm wonder why colon separate at all and just do lab ci artifacts [remote [ branch [ job [ path ]]]
The idea of adjust the exit code for job pass/fail is appealing. I'm not sure the trace command is the best place for it. Since a failed exit code would still indicate lab ci trace successfully printed logs. But I think the feature makes a lot of since for lab ci waitlab ci wait #240
IMHO the exit code of a command should reflect if it succeeded or not, and in the lab ci trace it shouldn't have any relation to the job failing or not. For it, you could use grep for grepping a successful message and then get the exit code of grep itself, which relates to "text found" or "text not found".
@MartinDelille, considering the command line mentioned by @zaquestion on his comment and the above statement, would that solve your concerns raised in this issue?
What about adding a command and a job to
lab ci
?Example: in my CI I have to job
build
andtest
. Let's say I want to trace thetest
job:It would directly output the job trace and returning to the shell with a result of 0 or -1 if the job succeed or failed.
The text was updated successfully, but these errors were encountered: