ABRT duplicate detection

[Christoph Höger]

Hi all,

I was just wondering: How mature is ABRTs duplicate detection?

You might all know the MS "search for a solution to this problem" window
popping up after an application crashes. I've never seen it doing
something usefull, though. But if ABRT could detect duplicate crashes
one could use this to display a workaround or even propose an update via
packagekit. All we'd need to do would be to tag (or maybe link from, if
a wiki would be better) an bugzilla entry as containing either a
workaround or a fix. ABRT could then display the solution just after the
crash happened. Great, wouldn't it be?

comments?

Christoph

[Alexander Boström]

lör 2010-04-24 klockan 10:37 +0200 skrev Christoph Höger:
> But if ABRT could detect duplicate crashes
> one could use this to display a workaround or even propose an update via
> packagekit.

It could look in bodhi for packages which claim to fix the bug.
 "Would you like to test a proposed fix for this problem?"
If the update is already installed or obsoleted then it just comments on
the bug as usual.

Also, a week later, if the same user logs in again:
 "Have you experienced any problems with the update of package foo which
  was installed last Tuesday to fix the problem 'Foo borkborks on bar
  when baz'?"

/Alexander

[Christoph Höger]

Am Sonntag, den 25.04.2010, 19:56 +0200 schrieb Alexander Boström:
> lör 2010-04-24 klockan 10:37 +0200 skrev Christoph Höger:
> > But if ABRT could detect duplicate crashes
> > one could use this to display a workaround or even propose an update via
> > packagekit.
>
> It could look in bodhi for packages which claim to fix the bug.
>  "Would you like to test a proposed fix for this problem?"
> If the update is already installed or obsoleted then it just comments on
> the bug as usual.
>
> Also, a week later, if the same user logs in again:
>  "Have you experienced any problems with the update of package foo which
>   was installed last Tuesday to fix the problem 'Foo borkborks on bar
>   when baz'?"

Yeah that would be the "easy" part. The more difficult would handle the
workarounds.

Nice to see some interest in this.

[Radek Vokál]

2010/4/24 Christoph Höger :
> Hi all,
>
> I was just wondering: How mature is ABRTs duplicate detection?
>
> You might all know the MS "search for a solution to this problem" window
> popping up after an application crashes. I've never seen it doing
> something usefull, though. But if ABRT could detect duplicate crashes
> one could use this to display a workaround or even propose an update via
> packagekit. All we'd need to do would be to tag (or maybe link from, if
> a wiki would be better) an bugzilla entry as containing either a
> workaround or a fix. ABRT could then display the solution just after the
> crash happened. Great, wouldn't it be?
>
> comments?
>
> Christoph

I've cced the crash-catcher mailing list. Probably the developers can
comment more about this idea, but in general I really - especially the
part -known bug & known workaround, which might be linked to a
wiki-like database of  user problems and solutions.

Radek

[Crash-catcher] ABRT duplicate detection

[Jiri Moskovcak]

On 04/25/2010 09:33 PM, Radek Vokal wrote:
> 2010/4/24 Christoph Höger:
> > Hi all,
> >
> > I was just wondering: How mature is ABRTs duplicate detection?
> >
> > You might all know the MS "search for a solution to this problem" window
> > popping up after an application crashes. I've never seen it doing
> > something usefull, though. But if ABRT could detect duplicate crashes
> > one could use this to display a workaround or even propose an update via
> > packagekit. All we'd need to do would be to tag (or maybe link from, if
> > a wiki would be better) an bugzilla entry as containing either a
> > workaround or a fix. ABRT could then display the solution just after the
> > crash happened. Great, wouldn't it be?
> >
> > comments?
> >
> > Christoph
>
> I've cced the crash-catcher mailing list. Probably the developers can
> comment more about this idea, but in general I really - especially the
> part -known bug&  known workaround, which might be linked to a
> wiki-like database of  user problems and solutions.
>
> Radek

If ABRT finds a duplicate it can ask bodhi if an update fixing this BZ 
exists and if so, then it can provide a way to run pkgkit to install 
this update.

J.
> Crash-catcher mailing list
> [e-mail]
> https://fedorahosted.org/mailman/listinfo/crash-catcher

[Pekka Pietikäinen]

On Sun, Apr 25, 2010 at 09:33:40PM +0200, Radek Vokal wrote:
> > You might all know the MS "search for a solution to this problem" window
> > popping up after an application crashes. I've never seen it doing
> > something usefull, though. But if ABRT could detect duplicate crashes
> > one could use this to display a workaround or even propose an update via
>
> I've cced the crash-catcher mailing list. Probably the developers can
> comment more about this idea, but in general I really - especially the
> part -known bug & known workaround, which might be linked to a
> wiki-like database of  user problems and solutions.

Also pretty relevant (and implemented by MS & Apple) is whether the crash
has potential security implications.  Doing it "well" is somewhat involved,
but a rough estimate shouldn't be that difficult, look at the backtrace
and do a little-lots of heuristics.

http://msecdbg.codeplex.com/
http://seclists.org/dailydave/2009/q3/11

We may be doing some work in this area soon btw. :)