On Wed, Jul 04, 2012 at 08:48:59AM +0200, Harald Dunkel wrote: > On 07/03/12 23:57, Yann Dirson wrote: > > > > This seems to show the debconf "frontend" script not dealing handling > > the termination of the memtest86+.postinst script. If any, it seems > > more plausible that the bug would be in debconf. Which debconf > > version is this ? Are you running squeeze or sid ? > > > > Testing. debconf is version 1.5.45 > > > If I run "dpkg-reconfigure memtest86+" here (wheezy/amd64), however, > > things go perfectly fine: > > > > | Generating grub.cfg ... > > | Found background image: /usr/share/images/desktop-base/desktop-grub.png > > | Found linux image: /boot/vmlinuz-3.2.0-2-amd64 > > | Found initrd image: /boot/initrd.img-3.2.0-2-amd64 > > [...] > > | Found memtest86 image: /memtest86.bin > > | Found memtest86+ image: /memtest86+.bin > > | Found memtest86+ multiboot image: /memtest86+_multiboot.bin > > | done > > | root# > > > > It would be interesting to know what the frontend process is doing > > when that happens. Can you retry, and show > > > > 1. the output of "ps l <pid>" > > root@ppcl007:~# ps l 3648 > F UID PID PPID PRI NI VSZ RSS WCHAN STAT TTY TIME COMMAND > 0 0 3648 3636 20 0 50532 11132 - S+ pts/1 0:00 > /usr/bin/perl -w /usr/share/debconf/frontend > /var/lib/dpkg/info/memtest86+.postinst configure > > > 2. the output of "strace -p <pid>" > > > > root@ppcl007:~# strace -p 3648 > Process 3648 attached - interrupt to quit > read(11,
In that case, we'll need to know what's hidden behind file descriptor 11. You can use "ls -l /proc/<pid>/fd/" to find out. You can also get info from debconf itself. Setting DEBCONF_DEBUG=developer while running dpkg can help. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
