[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: FTBFS on ppc64el

On Thu, Oct 25, 2018 at 04:55:54PM -0300, Breno Leitao wrote:
> On 10/23/2018 09:00 AM, Erwan Prioul wrote:
> > On Thu, 16 Feb 2017 16:49, Breno Leitao <brenohl@br.ibm.com> wrote:
> >> I would like to announce a small page we created to track the current
> >> packages that FTBFS on ppc64el.
> >>
> >> It also shows if the package only FTBFS on ppc64el, or if it FTBFS on
> >> other arches as well.
> >>
> >> As a bonus, we have a historical view of FTBFS in ppc64el.
> >>
> >> http://ftp.unicamp.br/pub/ppc64el/debian/packages/debian/
> > 
> > Since a couple of days, this has been available as a UDD cgi script:
> > 
> >     https://udd.debian.org/cgi-bin/ftbfs.cgi
> > 
> > This means it's quite faster and also gives results for any supported architecture> 
> Very nice to see it fully integrated! Thanks!
> Paul and Helmut had some suggestions about improving this tool even better.
> Paul, Helmut, would you mind sharing them in this public email?

I have two suggestions:

Document the use of usertags more prominently. For instance, the form
where you select the architecture could link to a wiki page explaining
which usertags it uses.

Since a while we have a tag "ftbfs". When filed against a source
package, the package fails to build. When filed against a binary
package, the affected source packages fail to build. Maybe you could
add reported FTBFS bugs to the view? Here is an example udd query:

SELECT id, package, affected_packages, title FROM all_bugs WHERE EXISTS (SELECT 1 FROM bugs_tags WHERE bugs_tags.id = all_bugs.id AND tag = 'ftbfs') AND severity IN ('serious', 'critical', 'grave') AND affects_unstable = 't';

It needs further post processing in another language though. The
affected_packages column yields space separated packages. That's
non-trivial to dissect in sql.

The presently linked bugs are few, because few are usertagged. Mirroring
the general ftbfs tag into the view would allow checking that all issues
are properly reported.


Reply to: