Hello List, my syslog-ng "loghost" crash every 1-5 minute, so I did the update to 1.6.0rc1, but the problem still exist. The service died with a "Segmentation fault". A run in debug mode and/ore strace report: ----->8------schnipp-----8<----------------------------------- time(NULL) = 1047983908 time(NULL) = 1047983908 --- SIGSEGV (Segmentation fault) @ 0 (0) --- +++ killed by SIGSEGV +++ ----->8------schnapp-----8<----------------------------------- after the last logging entry. The os is a debian unstable on a Siemens Nixdorf 2x333Mhz server. Anybody with an idea? Bye Bye, Chriss mailto:syslog-ng@kordesee.de ---- Turn your Pentium into a Gameboy: Type WIN at C:\>
Hi List, had anyone seen something like the listing below? ---------------- # /sbin/syslog-ng -d -v -f /etc/syslog-ng.conf binding fd 3, unixaddr: /dev/log io.c: Preparing fd 3 for reading io.c: Preparing fd 4 for reading syslog-ng version 1.6.0rc1 starting io.c: sockaddr2info(): Unsupported address family 29336 (addrlen=256). io.c: Preparing fd 5 for writing [...] Segmentation fault # -------- every syslog message creates one error message from the io.c. C> my syslog-ng "loghost" crash every 1-5 minute, so I did the update to C> 1.6.0rc1, but the problem still exist. The service died with a C> "Segmentation fault". A run in debug mode and/ore strace report: ----->>8------schnipp-----8<----------------------------------- C> time(NULL) = 1047983908 C> time(NULL) = 1047983908 C> --- SIGSEGV (Segmentation fault) @ 0 (0) --- C> +++ killed by SIGSEGV +++ ----->>8------schnapp-----8<----------------------------------- C> after the last logging entry. C> The os is a debian unstable on a Siemens Nixdorf C> 2x333Mhz server. C> Anybody with an idea? Adios, Chriss mailto:syslog-ng@kordesee.de ----
On Wednesday 19 March 2003 11:49, Chriss wrote:
Hi List,
had anyone seen something like the listing below?
No! But can you give us some more information: - core file or gdb output when loading the core file - syslog-ng config file - libol version Achim
Hi All! On 2003 Mar 19, Achim Gsell wrote:
- core file or gdb output when loading the core file - syslog-ng config file - libol version
- kernel version (guess: 2.4.20) -- PGP ID 0x8D143771, /C5 95 43 F8 6F 19 E8 29 53 5E 96 61 05 63 42 D0 GPG ID ABA0E8B2, 45CF B559 8281 8091 8469 CACD DB71 AEFC ABA0 E8B2 I refused to attend his funeral. But I wrote a very nice letter explaining that I approved of it. -- Mark Twain
Holla *, SA> - kernel version (guess: 2.4.20) Yes, you are right: Linux fry 2.4.20-686-smp #1 SMP Mon Jan 13 23:06:41 EST 2003 i686 Pentium II (Klamath) Chriss mailto:syslog-ng@kordesee.de
Servus syslog-ng, AG> On Wednesday 19 March 2003 11:49, Chriss wrote:
Hi List,
had anyone seen something like the listing below?
AG> No! But can you give us some more information: AG> - core file or gdb output when loading the core file There is no core. :( AG> - syslog-ng config file options { long_hostnames(off); sync(0); }; source src { unix-dgram("/dev/log"); internal(); file("/proc/kmsg" log_prefix("kernel: ")); udp(); tcp(); }; destination fry_all { file("/log/syslog/fry/all" owner("root") group("root") perm(0600)); }; destination fels_all { file("/log/syslog/fels/all" owner("root") group("root") perm(0600)); }; destination leela_all { file("/log/syslog/leela/all" owner("root") group("root") perm(0600)); }; destination leela_cron { file("/log/syslog/leela/cron" owner("root") group("root") perm(0600)); }; filter f_fry_all {host(fry); }; filter f_fels_all {host(fels); }; filter f_leela_all {host(leela) and not facility(cron); }; filter f_cron { facility(cron); }; log { source(src); filter(f_fry_all); destination(fry_all); }; log { source(src); filter(f_fels_all); destination(fels_all); }; log { source(src); filter(f_cron); destination(leela_cron); }; log { source(src); filter(f_leela_all); destination(leela_all); }; Fry is the "localhost", "leela" is a second syslog-ng box and "fels" is a sol8 box with the syslog from Sun. AG> - libol version libol-0.3.9 Chriss mailto:syslog-ng@kordesee.de ---- Der Mensch steht im Mittelpunkt, und deshalb im Weg.
On Wednesday 19 March 2003 15:15, Chriss wrote:
AG> - core file or gdb output when loading the core file There is no core. :(
To enable core file generation you have to do (with root privileges) something like cd /tmp ulimit -c unlimited /etc/init.d/syslog-ng stop /etc/init.d/syslog-ng start When syslog-ng crashes, there will be a new core file in "/tmp". Achim
Guten Tag syslog-ng, AG> On Wednesday 19 March 2003 15:15, Chriss wrote:
AG> - core file or gdb output when loading the core file There is no core. :(
AG> When syslog-ng crashes, there will be a new core file in "/tmp". still no core file. :( Chriss mailto:syslog-ng@kordesee.de
participants (3)
-
Achim Gsell
-
Chriss
-
SZALAY Attila