On Wed, Jul 02, 2014 at 12:26:57AM +0200, Bastian Blank wrote: > Package: perl-base > Version: 5.18.2-4+b1 > Severity: grave > > On s390x perl fails to run debconf and segfaults later on:
> | $ gdb --args perl /usr/share/debconf/frontend [...] > | debconf: DbDriver "passwords" warning: could not open > /var/cache/debconf/passwords.dat: Permission denied > | Use of uninitialized value $ARGV[0] in string eq at > /usr/share/debconf/frontend line 17. > | Use of uninitialized value $ARGV[0] in pattern match (m//) at > /usr/share/debconf/frontend line 24. > | Use of uninitialized value $ARGV[0] in concatenation (.) or string at > /usr/share/debconf/frontend line 60. > | Use of uninitialized value $ARGV[0] in pattern match (m//) at > /usr/share/debconf/frontend line 61. > | Use of uninitialized value $ARGV[0] in pattern match (m//) at > /usr/share/debconf/frontend line 62. > | Use of uninitialized value $ARGV[0] in pattern match (m//) at > /usr/share/debconf/frontend line 72. > | Use of uninitialized value $r in <HANDLE> at > /usr/share/perl5/Debconf/ConfModule.pm line 72. > | readline() on unopened filehandle at /usr/share/perl5/Debconf/ConfModule.pm > line 72. FWIW these messages seem to be "normal" when running /usr/share/debconf/frontend by hand. I see them on amd64 too. > | Program received signal SIGSEGV, Segmentation fault. > | 0x000003fffded8838 in ?? () from /usr/lib/libperl.so.5.18 > | (gdb) bt > | #0 0x000003fffded8838 in ?? () from /usr/lib/libperl.so.5.18 > | #1 0x000003fffdeceba0 in ?? () from /usr/lib/libperl.so.5.18 > | #2 0x000003fffded8f60 in Perl_sv_clean_objs () from > /usr/lib/libperl.so.5.18 > | #3 0x000003fffde45634 in perl_destruct () from /usr/lib/libperl.so.5.18 > | #4 0x0000000080000f30 in main () > | (gdb) Unfortunately I can't reproduce the segfault on zelenka, in a sid or jessie s390x chroot. To make matters worse, even gdb doesn't seem to work there: (gdb) run Starting program: /usr/bin/perl /usr/share/debconf/frontend Couldn't write registers: Invalid argument. Is it possible to arrange access on a box where it happens? > This breaks installation of glibc binary packages like tzdata. As there > are no debugging infos available, no further information can be shown. Do you get a better backtrace after installing perl-debug? (Not that I expect that to help much.) I see 5.18.2-4+b1 is a rebuild against libc6 2.19, built with 2.19-1 on zani three weeks ago. Do you know the rebuild was necessary? Two builds failed on zemlinsky with libc6 2.19-1 and t/comp/parser FAILED--non-zero wait status: 11 which might be the same segfault issue. I guess it could be HW or kernel related... The zelenka chroot has libc6 2.19-4 installed. Which version do you have? If you download the perl_5.18.2-4 source and run perl t/comp/parser.t in the source directory, do you see a segfault too? #604128 / #679811 (tagged as unreproducible) might be the same issue. -- Niko Tyni nt...@debian.org -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
