On 6/22/07, Bryan Donlan <[EMAIL PROTECTED]> wrote:
On 6/22/07, Daniel Burrows <[EMAIL PROTECTED]> wrote:
> On Fri, Jun 22, 2007 at 12:28:32AM -0400, Bryan Donlan <[EMAIL PROTECTED]>
was heard to say:
> > Aptitude in experimental crashes on startup. Backtrace:
> > Program terminated with signal 11, Segmentation fault.
> > #0 0x081f0867 in aptitudeDepCache::cleanup_after_change (this=0x8324a38,
> > undo=0x0, alter_stickies=true) at aptcache.cc:756
> > 756
if(PkgState[pkg->ID].Mode!=backup_state.PkgState[pkg->ID].Mode ||
> > (gdb) bt
> > #0 0x081f0867 in aptitudeDepCache::cleanup_after_change (this=0x8324a38,
> > undo=0x0, alter_stickies=true) at aptcache.cc:756
>
> Interesting, that looks a lot like the backtrace I get for #430050...
> but that one is a real pain to catch, since it just happens for an instant.
> Would it be possible for you to upload the contents of /etc/apt and
> /var/lib/apt to the same location as the other files?
Uploaded at http://fushizen.net/dpkg-state-2.tbz2 (this includes new
snapshots of the other dirs) - but note that this segfault has
mysteriously stopped. Here's a outline of the situation:
* Took dpkg-state.tbz2
* apt-get build-dep aptitude
* Took the coredump from the (still-crashing) debug build of aptitude
* ???
* Tried aptitude again, didn't crash
* Took dpkg-state-2.tbz2
Not entirely sure what changed in-between. Anyway, maybe you can find
a difference in the state tarball above.
Also, I've uploaded the actual debug binary (or will, in about 60
seconds) at http://fushizen.net/aptitude.bz2 - as it just now occurred
to me that perhaps a core without the symbols won't be particularly
useful :)
It started crashing again today. I grabbed some new state tarballs:
http://fushizen.net/aptitude-state-3.tbz2
http://fushizen.net/core.17966.bz2
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
