Identify product issues #223
Conversation
perlpunk
force-pushed
the
product-issues
branch
2 times, most recently
from
02e83ce
to
d727b23
Compare
perlpunk
force-pushed
the
product-issues
branch
from
d727b23
to
c992c51
Compare
perlpunk
force-pushed
the
product-issues
branch
3 times, most recently
from
3f645cf
to
e543a1a
Compare
|
I updated the screenshot with a new comment I just created on osd. |
There was a problem hiding this comment.
Can you explain in more detail why you wait for the retry-comment and extend that instead of it's open comment? If you wanted to reduce the amount of comments and have found a good way to edit existing comments how about always updating the existing initial comment instead?
openqa-investigate
Outdated
| is-product-issue "$origin_job_id" || return $? | ||
|
|
||
| if "$product_issue"; then | ||
| comment+="."$'\n'"Test(s) with last good build pass, likely a product issue" |
There was a problem hiding this comment.
We use the past tense and the actual result for the tests job, I would use the same here, i.e. "passed" or "soft-failed". As alternative use "ok", as we also use within openQA where we define "ok" as "passed or soft-failed"
There was a problem hiding this comment.
"Jobs including the last good build are ok, likely a product issue"
I don't understand that sentence at all. Can you rephrase? What waiting do you mean?
Yes, that would be possible. |
Sorry. I meant: Please explain why you add a message to the comment from the "retry" investigation job instead of creating a new comment with just your information about potential product regression |
Because I'm doing that at the same time anyway. Posting two different comments didn't even occur to me. At the time I know that the simple retry test failed, I also investigate the other investigation tests, and then I post a comment with that combined information. I have this information at the same time. It wouldn't make sense to me to just split these two sentences into two comments. We also don't split the first investigation test list into four separate comments. Maybe we can clarify this tomorrow in a chat. |
Use the retry job to fetch the results of the other investigate jobs by parsing the investigation comment. Issues: * https://progress.opensuse.org/issues/126527 * https://progress.opensuse.org/issues/109920
Spike solution which uses the retry job to fetch the results of the other investigate jobs by parsing the investigation comment.
Use the retry job to fetch the results of the other investigate jobs by
parsing the investigation comment.
Issues:
Demo:
