Latest silc server segfaults

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi

we are running a silc-sever running latest version connected to a
silc-router (also latest version, likely on freebsd but not within our
control). However from time to time (sometimes every 6 hour, sometimes
only every second week) our silc server segfaults.

This is what we could record so far:

16:40:29 silc_server_packet_parse_type:328: Received CHANNEL KEY packet
[flags 0] from X.Y.Z.A:706
16:40:29 silc_server_save_channel_key:4074: Saving new channel blob key

Program received signal SIGSEGV, Segmentation fault.
0x0000003feec08ca0 in pthread_mutex_lock () from /lib64/libpthread.so.0

Backtrace:
- ----------

(gdb) backtrace
#0  0x0000003feec08ca0 in pthread_mutex_lock () from /lib64/libpthread.so.0
#1  0x00000036e002c47e in silc_mutex_lock () from
/usr/lib64/libsilc-1.1.so.2
#2  0x00000036e00209ba in silc_packet_get_context () from
/usr/lib64/libsilc-1.1.so.2
#3  0x0000000000423bff in silc_server_packet_send (server=0x85e0e0,
sock=0xdeece0, type=24 '\030', flags=0 '\000',
    data=0x0, data_len=0) at packet_send.c:37
#4  0x0000000000423cd2 in silc_server_send_heartbeat
(server=0x200000005, sock=0xdeece0) at packet_send.c:1580
#5  0x0000000000415a51 in silc_server_do_heartbeat (schedule=<value
optimized out>, app_context=0x200000005, type=24,
    fd=0, context=0x0) at server.c:2910
#6  0x00000036e00319e4 in ?? () from /usr/lib64/libsilc-1.1.so.2
#7  0x00000036e0031c2a in ?? () from /usr/lib64/libsilc-1.1.so.2
#8  0x00000036e0031dbc in silc_schedule () from /usr/lib64/libsilc-1.1.so.2
#9  0x0000000000416163 in silc_server_run (server=0x85e0e0) at server.c:1220
#10 0x000000000040d1b5 in main (argc=<value optimized out>,
argv=0x85fac0) at silcd.c:784


We are currently installing -debuginfo for libsilc so we could get also
information about the libsilc parts.

The system is a CentOS 5.5 x86_64, both libsilc and silc-server are
handcrafted rpm packages with the following configure options:

libsilc:
- --------
%configure --libdir=%{_libdir} --enable-shared --without-libtoolfix \
           --includedir=%{_includedir}/silc
- --with-simdir=%{_libdir}/silc/modules \
           --docdir=%{_docdir}/%{name}-%{version} CFLAGS="$RPM_OPT_FLAGS"

silc-server:
- ------------
%configure \
        --with-silcd-config-file=%{_silcetcdir}/silcd.conf \
        --with-logsdir=%{_var}/log/silc \
        --mandir=%{_mandir} \
        --libdir=%{_silclibdir} \
        --enable-ipv6 \
        --enable-debug \
        --with-silcd-pid-file=%{_var}/run/silcd.pid CFLAGS="$RPM_OPT_FLAGS"


Any ideas why the server is segfaulting? Do you need any additional
information? What else could we do to provide you with sufficient
information about the problem to be able to fix the issue?

I'll repost detailed information about the next segfault with -debuginfo
for libsilc when it happended.

Thanks!

~pete
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkz7kYIACgkQbwltcAfKi3/WlQCcD/Bz91CTLWMG2lZ6CJN/UUba
+VUAn2eK+Aev1t5tV6slJMeCHrEUdBbt
=2B/l
-----END PGP SIGNATURE-----
_______________________________________________________________________
Info:    https://lists.silcnet.org/mailman/listinfo/silc-announce
Archive: https://lists.silcnet.org/pipermail/silc-announce
FAQ:     http://silcnet.org/support/faq/