ABRT frustrating for users and developers

[Nils Philippsen]

On Mon, 2010-01-18 at 11:17 +0100, Jan Kratochvil wrote:
> On Mon, 18 Jan 2010 09:18:11 +0100, Jiri Moskovcak wrote:
> > On 01/17/2010 05:57 PM, Christoph Wickert wrote:
> > >      5. Instead of hashes the missing debuginfo packages should be
> > >         listed with n-v-r, so people can install them manually.
> >
> > This could be a problem. ABRT determines the required debuginfo
> > package using build_id. It calls yum --enablerepo=*debuginfo*
> > --quiet provides /usr/lib/debug/.build-id/bb/11528d59940983f495e9cb099cafb0cb206051.debug
> > and I don't know any other way how to map build_id to package name
> > if yum fails.
> 
> yum fails commonly because rpm knows about build-id hashes only in the
> debuginfo.rpm archives (not the main .rpm archive). And debuginfo.rpm exists
> in repositories only for the single latest release (this is a distro bug!).
> 
> Therefore if the crashed program has been updated in repositories since it
> started there is no way to map build-id -> nvr.

Chiming in a bit late, but nevertheless: IMO, rather than somehow
getting debuginfo packages for older versions, abrt could e.g. ask
people to try to reproduce their problem with an up to date set of
packages. This will often be the right thing to do: somebody else didn't
dawdle and reported the bug right away, the maintainer fixed it and
pushed an update. Abrt could then even tell the user "your problem has
already been reported and apparently fixed" if it finds the reported
Bugzilla ticked is CLOSED/CURRENTRELEASE or .../ERRATA. No need to clog
up Bugzilla for that.

I agree that sometimes it should generate a backtrace from old debuginfo
packages and open a ticket with that, e.g. when the issue is hard to
reproduce. However, in that case it even more shouldn't attach an
incomplete backtrace, because it's likely missing essential information
for the developer: "Hmm, your backtrace is incomplete, can you reproduce
it with this and that debuginfo RPM installed?" -- "No." -- "Sorry
then." --> CLOSED/INSUFFICIENT_DATA.

Nils
-- 
Nils Philippsen      "Those who would give up Essential Liberty to purchase 
Red Hat               a little Temporary Safety, deserve neither Liberty
nils at redhat.com       nor Safety."  --  Benjamin Franklin, 1759
PGP fingerprint:      C4A8 9474 5C4C ADE3 2B8F  656D 47D8 9B65 6951 3011