Hi!

On Fri, 2010-11-26 at 15:58:05 -0500, Michael Gilbert wrote:
> package: dpkg
> version: 1.15.8.5
> severity: important
> 
> Building a fakechroot on i386 fails due to a dpkg segfault.  The
> following executed on an i386 system should reproduce the issue:
> 
> $ export PATH=/sbin:/usr/sbin:/usr/local/sbin:/bin:/usr/bin:/usr/local/bin
> $ fakeroot fakechroot debootstrap --variant fakechroot --arch=i386
> squeeze ./chroot http://ftp.us.debian.org/debian
> W: Failure trying to run: chroot /home/zero79/chroot dpkg
> --force-depends --install
> /var/cache/apt/archives/base-files_5.10_i386.deb
> /var/cache/apt/archives/base-passwd_3.5.22_i386.deb
> 
> The output of ./chroot/debootstrap/debootstrap.log contains:
> 
>   /usr/share/debootstrap/functions: line 745: 12387 Segmentation fault
>     chroot "$TARGET" "$@"
> 
> Running the same command on an amd64 results in the same failure with
> following in deboostrap.log
> 
>   ERROR: ld.so: object 'libfakeroot-sysv.so' from LD_PRELOAD cannot be
> preloaded: ignored.
>   dpkg: requested operation requires superuser privilege
> 
> Running the command on amd64 without "--arch=i386" results in a
> successful fakechroot bootstrap.

Is it really dpkg segfaulting? Even if it is, my guess would be
fakechroot is at fault here, or some interaction problem between
fakeroot and fakechroot, as both do symbol overrides through linker
preloading. Anyway I'd recommend enabling core dumps and providing a
backtrace.

At least I'll not have time to look into this specific issue, so maybe
it would be better off being reassigned preemptively.

thanks,
guillem



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Reply via email to