On Tue, Jul 31, 2012 at 9:42 PM, Kevin Kofler <kevin.kofler(a)chello.at> wrote:
I'm looking into these:
Bill Nottingham wrote:
> Package komparator (fails to build)
> Package krecipes (fails to build)
> Package qalculate-kde (fails to build)
> Package tesseract (fails to build)
but since the build.log files are no longer available, I need to run new
builds first, so it's going to take a while. :-(
Are those build.log files really so large that they cannot be kept for
longer? :-( It's so annoying to see those build failures and to have no idea
why the builds failed.
There's 3 per build (src,i686,x64) and they do get pretty big, just
look at the size of a kernel, qt or libreoffice build.
I also really dislike the way FTBFS are handled now. In the past,
filed for the build failures and they were all tracked on a tracking bug in
Bugzilla, so it was always possible to see what needed fixing, and it was
possible for provenpackagers to work on that in a coordinated fashion. These
days, we get surprised with the failures at the last moment when they have
actually been happening for 2 or 3 releases (depending on whether they would
have built on F16 had there been a mass rebuild there)! We really need to
track those failures in Bugzilla again.
That's being worked on as discussed in one of these threads,
previously it was done by Matt Domsch with access large amounts of
enterprise HW, there was a request for someone to step up to expedite
the process if you have time. That said every build gets failure
emails sent out so you could setup some form of rules to highlight the
problem or write a script to query koji on occasion for your failed
builds to email you a list of current non builds.