Follow-up from "[Backend MVP] Root-cause analysis of failing pipeline job"
The following discussions from !121081 (merged) should be addressed. They're just qualitative refactorings for some tests, but the feature is still flagged and experimental so we don't need to hold up the original MR for them.
-
@drew started a discussion: Just thinking about that now, do we have any kind of itegration-y style test that sort of runs through "a job failing and having it's log analyzed with the response content saved to somewhere"? It might not be in this particular MR or anything but a good thing for me to use to sanity check the whole workflow.
-
@drew started a discussion: I'm still not completely sure what's meant by this test. It does the same thing twice with two different jobs - do we mean to assert that the analysis for
job
has content while the analysis forjob2
doesn't, on line 18 before we save it for the second job? Or if we just want make sure the analyses are different, we might want to have the contents not be the same?I think this MR is fine so I'm going to go ahead and merge it, but open a follow-up for this where we can chat about it
🏃