Re: Killing bug-buddy?

[Milan Crha <mcrha redhat com>]

On Fri, 2011-08-26 at 09:46 +0200, Olav Vitters wrote:
> > I guess you want to kill bug-buddy for regular users only, as you was
> > talking about bugzilla connection, because ABRT is unusable for
> > developers, as it silently ignores crashes in applications which are not
> > packaged, furthermore, if you enable to catch crashes of unpackaged
> 
> Do you know how difficult would it to change that behaviour in ABRT?
> Make it support jhbuild?

	Hi,
enabling abrt to catch unpackaged applications is easy, just edit
   /etc/abrt/abrt.conf
and change
   ProcessUnpackaged = no 
to
   ProcessUnpackaged = yes
I do not know whether it is doable with other than root user, though,
but I feel that this is not an option for regular jhbuild users.

> > applications, then it is not able to get the backtrace right. That's a
> > reason why I resurrected bug-buddy on my machine, also for gtk3
> > applications, thus I know why my application is crashing, and when.
> 
> Doesn't it just run gdb? I'd still think it is best to fix ABRT than to
> have multiple tools to report crashers.

I just gave this a try, just in case, with abrt 2.0.3, and it gathers
bactraces for unpackaged files correctly now, thus they fixed it since I
tried the last time, which, I confess, is some time ago. I suspect it
had been fixed together with an access to .xsession-errors file, to
which abrt daemon didn't have access in previous versions too.

Anyway, I'm not an ABRT developer, I only face its reports and try to
get something useful from them. If you've any questions, then it's
better to ask ABRT developers, as they know all about it.

One note I forgot to mention in the previous mail: From my point of
view, the all information brought in by ABRT is usually unnecessary, as
for me is backtrace enough, and when I would like to ask for more info
then I'll ask a reporter. Also, ABRT tends to attach backtraces instead
of pasting them in a comment, which results in unsearchable bugzilla for
duplicates. That all might be changeable, maybe it already is or with a
cooperation with them. As an example see this ABRT report:

   https://bugzilla.redhat.com/show_bug.cgi?id=733505

there is, from my point of view, tons of crap till you get to the only
valuable information, which is a backtrace (unfortunately attached), and
console output (the .xsession-errors). These are two things I usually
look at and silently ignore the rest. ABRT has some functionality to
disable unneeded output already, as is stated in:
   https://fedorahosted.org/abrt/ticket/226

How it is with forcing backtraces as comments, not as attachments, I do
not know. It's subject to check out, and as you said, some developing is
still required, then it'll be great once it's done.
	Bye,
	Milan