[Qa-jenkins-dev] Bug#900630: Bug#900630: jenkins.debian.org: reproducible Debian: unhelpful mails when diffoscope crashed

Mattia Rizzolo mattia at debian.org
Sat Jun 2 17:11:30 BST 2018


On Sat, Jun 02, 2018 at 12:31:12PM +0200, Holger Levsen wrote:
> < Myon> fwiw, I'm pretty confused by all the mail notifications that link to pages that have no diffoscope output
> < h01ger> | can you share such a mail on paste.d.n please?
> < Myon> 2018-06-01 08:51 https://tests.reproducible-builds.org/debian/unstable/armhf/postgresql-pgmp
> < Myon> changed from reproducible -> unreproducible
> < Myon> 2018-06-02 00:27 https://tests.reproducible-builds.org/debian/unstable/arm64/orafce changed from
> < Myon> reproducible -> unreproducible
> < Myon> 2018-06-01 23:05 https://tests.reproducible-builds.org/debian/unstable/amd64/pgq-node changed
> < Myon> from reproducible -> unreproducible
> < h01ger> | well, those packages (i assume, only checked the first) have notes, so the notes is displayed. if 
> 	    there wasnt a note, the diffoscope output would be displayed. so its a click away
> < h01ger> | IOW: works as designed, not sure what to do
> < Myon> it's not a click away, there is no link
> < h01ger> | point
> < h01ger> | guess diffoscope crashed then
> < Myon> (on a sidenote, thanks for the Subject update in the mails, much more useful now)
> < h01ger> | yup, postgresql-pgmp is on https://tests.reproducible-builds.org/debian/index_breakages.html
> < Myon> I got about 6 mails this morning/night, none of them worked
> < h01ger> | mapreri rescheduled all packages where diffoscope crashed or ran into timeout…

Yes, all builds from yesterday were marked as unreproducible because for
some reason python3-distutils was not installed in the unstable
diffoscope schroot (but apt was happy?!) and so all diffoscope calls
crashed.  Original reason is a dak issue that prevented
python3-distutils from being built.

> < h01ger> | i guess what we should do is to create a page "diffoscope crashed, linking to index_breakages.html" 
>   	    for such cases. then at least you can click and see what happened
> < h01ger> | or not send such notification mails if this happens
> < h01ger> | or both
> < Myon> good idea, yes
> < h01ger> | Myon: i'll put everything since 11:49 into a bug report. ok?
> < Myon> and not call that status "unreproducible"
> < Myon> ci.d.n calls it "tmpfail"
> < Myon> ok
> < h01ger> | also a good idea (though that will require db changes)

The only sane idea here is the other status, everything else is IMHO
even harder to implement in our current setup.
(incidentally, I had even started to do that a couple of months ago…)

-- 
regards,
                        Mattia Rizzolo

GPG Key: 66AE 2B4A FCCF 3F52 DA18  4D18 4B04 3FCD B944 4540      .''`.
more about me:  https://mapreri.org                             : :'  :
Launchpad user: https://launchpad.net/~mapreri                  `. `'`
Debian QA page: https://qa.debian.org/developer.php?login=mattia  `-
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/qa-jenkins-dev/attachments/20180602/b05a4193/attachment.sig>


More information about the Qa-jenkins-dev mailing list