On 28. 08. 19 13:24, Pierre-Yves Chibon wrote:
On Wed, Aug 28, 2019 at 12:16:09PM +0200, Miro Hrončok wrote:
On 28. 08. 19 9:32, Clement Verna wrote:
On Tue, Aug 27, 2019, 15:35 Miro Hrončok <mhroncok@xxxxxxxxxx
<mailto:mhroncok@xxxxxxxxxx>> wrote:
If updates in rawhide gating fail the tests, what is the procedure? Should they
be unpuhsed, or left to rot forever?
There are a few options here, fix the package so that the tests pass,
disable the tests, or waive the results. If none of these are done then
yes the update will just stay there to rot.
I meant what to do with the broken updates that are actually broken. When a
new update is created, the older one doesn't seem to be obsoleted or
unpushed.
I'm not sure I understand what you mean. What do you mean with "Broken updates
that are actually broken"?
I mean when I push an update A and a test T realizes update A is broken, so I
fix it and push another update B. I.e. I have no interest in waiving the failed
test T in update A, because update A actually was broken.
Update B now does not obsolete update A (as you have described in the paragraph
below). As a packager, should I unpush update A or just do nothing?
We did think about obsoleting updates when a newer one is created but decided
not to (if for some reasons -1's tests take longer than -2's, the outcome of
-1's tests run may still be of interest for the packager and yum/dnf will anyway
prefer the higher version).
Is the issue about updates lingering in -testing?
Yes.
--
Miro Hrončok
--
Phone: +420777974800
IRC: mhroncok
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
