Ive been having a massive problem with CW segfaulting for no apparent
reason.
Its the latest trunk of cw, the latest SpanDSP, no special hardware; only
SIP, and about
2 dozen users. It runs on [k]Ubuntu with kernel 2.6.20-16 [lowlatency].
Here is a log showing some of the erratic behavior:
2007-07-18 14:07:38 - CW died!
2007-07-18 15:12:12 - CW died!
2007-07-18 15:22:24 - CW died!
2007-07-18 16:22:38 - CW died!
The worst part isnt cw dying (it just automatically restarts) but that it
drops call anywhere from
immediately to a few minutes into the call, and it seems to do that to
almost 40% of the calls.
It also seems to just cut in and out of the conversation and have brief
periods of heavy echo.
Some users have also reported that as soon as the call is dropped MOH starts
playing, but usually
they just get a dialtone or a 'busy' beep.
Here is a stack-backtrace:
Core was generated by `/usr/local/sbin/callweaver -vvvdgc'.
Program terminated with signal 11, Segmentation fault.
(gdb) bt full
#0 0xb7e87460 in pthread_mutex_lock () from
/lib/tls/i686/cmov/libpthread.so.0
No symbol table info available.
#1 0xb67aacb0 in retrans_pkt (data=0x811a1a0)
at ../include/callweaver/lock.h:585
cur = <value optimized out>
iabuf = "\b\016w¶\001\000\000\000\000\000\000\000¤Õ÷·"
reschedule = <value optimized out>
__PRETTY_FUNCTION__ = "retrans_pkt"
#2 0xb7edb24c in opbx_sched_runq (con=0x8086e50) at sched.c:498
x = 0
res = <value optimized out>
#3 0xb67b7424 in do_monitor (data=0x0) at chan_sip.c:14542
res = 0
sip = <value optimized out>
t = 1184789530
fastrestart = 0
lastpeernum = -1
curpeernum = 24
reloading = <value optimized out>
__PRETTY_FUNCTION__ = "do_monitor"
#4 0xb7e8531b in start_thread () from /lib/tls/i686/cmov/libpthread.so.0
No symbol table info available.
#5 0xb7d5b57e in clone () from /lib/tls/i686/cmov/libc.so.6
No symbol table info available.
It seems that rc4 runs more stable, so Im going to switch to that and see
how much
of my problems it corrects.
RoyK has been looking into this for me, you checked out all my conf files,
my setup,
he even built me a new kernel with high-res timers. Still no go.
Also, with RoyK's changes (enabling lots of debugging, etc), I managed to
isolate this message whenever a call was dropped:
Jul 18 13:38:10 WARNING[3060685712]: chan_sip.c:1468 retrans_pkt: Maximum
retries exceeded on transmission [EMAIL PROTECTED] for seqno
19250 (Critical Response)
Jul 18 13:38:10 WARNING[3060685712]: chan_sip.c:1490 retrans_pkt: Hanging up
call [EMAIL PROTECTED] - no reply to our critical packet.
== Spawn extension (procycle, 221, 2) exited non-zero on
'SIP/procycle-b5f070a8'
The part that doesnt make sense is that all the equipment are on the same
LAN, just a few switches apart. Im using a GXW4104 4-port FXO gateway and a
GXW4008 8-port FXS gateway for my connection with the analog lines.
Someone please help me out :(
_________________________________________________________________
Don't just search. Find. Check out the new MSN Search!
http://search.msn.com/
_______________________________________________
Callweaver-users mailing list
[email protected]
http://lists.callweaver.org/mailman/listinfo/callweaver-users
