[Asterisk-Users] Tracing a crash with CAPI calls

Andrew Gough AGough at gcdtech.com
Thu Dec 22 16:32:31 MST 2005


> -----Original Message-----
> From: asterisk-users-bounces at lists.digium.com [mailto:asterisk-users-
> bounces at lists.digium.com] On Behalf Of Armin Schindler
> Sent: 22 December 2005 08:12
> To: Asterisk Users Mailing List - Non-Commercial Discussion
> Subject: RE: [Asterisk-Users] Tracing a crash with CAPI calls
> 
> On Wed, 21 Dec 2005, Andrew Gough wrote:
> > > On Wed, 21 Dec 2005, Andrew Gough wrote:
> > > > I have just setup asterisk on a debian sarge box. I am running
> > Asterisk
> > > > 1.21 with AMP and chan_capi_cm 0.6.1  using a BT Speedway (AVM
> > Fritz)
> > > > ISDN card, connected to a BT ISDN2e line. Currently we have 6
> > extensions
> > > > configured all using CounterPath(Xten) eyebeam softphone.
> > > >
> > > > After many hours of Googling I have finally got it all setup and
> > > > working. We can transfer calls internally and make and receive
> > external
> > > > calls. Its all great except for stability issues!!
> > > >
> > > > Essentially  every now and again, in the middle of a call (so
far
> > only
> > > > external CAPI calls) asterisk simply dies. No warning, no error,
> > just my
> > > > console session outputs a disconnected from console message.
> > > >
> > > > The server is a brand new AMD 3400+ with 512Mb RAM.
> > > > The other issue experienced is occasional break up on inbound
sound
> > > > quality.
> > > >
> > > > I don't expect anyone to be able to solve this one straight away
but
> > I
> > > > am at a loss where to look, I have tried /var/logs/messages
> > >
> > > Please try latest CVS sources of chan_capi from sourceforge. There
are
> > > some
> > > bugfixes which may help you.
> > > If this doesn't help, please provide a log (set verbose 5 , capi
> > debug) of
> > > the error.
> > >
> > > Armin
> >
> >
> > Ok got the lastest sources from cvs for chan_capi and installed it.
Left
> > console seesion running with setting suggested, Crashed again twice
this
> > afternoon. Below is the output, not sure if it tells you anything.
> 
> The CAPI details look good, there is no error and the call is released
> correctly. It looks like the crash happens after the other lines when
the
> SIP call is released:
> >     -- Executing Hangup("SIP/113-a753", "") in new stack
> >   == Spawn extension (macro-hangupcall, s, 4) exited non-zero on
> > 'SIP/113-a753' in macro 'hangupcall'
> >   == Spawn extension (from-internal, h, 1) exited non-zero on
> > 'SIP/113-a753'
> 
> So at this point it doesn't look like a chan_capi problem. Can you
create
> a backtrace (with a core dump)?
> 
> Armin
> 
> > -----------------------------------------------------
> >
> >
> >   == Spawn extension (macro-dialout-trunk, s, 21) exited non-zero on
> > 'SIP/113-a753' in macro 'dialout-trunk'
> >   == Spawn extension (from-internal, 907950846621, 1) exited
non-zero on
> > 'SIP/113-a753'
> >     -- Executing Macro("SIP/113-a753", "hangupcall") in new stack
> >     -- Executing ResetCDR("SIP/113-a753", "w") in new stack
> > INFO_IND ID=001 #0x3473 LEN=0017
> >   Controller/PLCI/NCCI            = 0x101
> >   InfoNumber                      = 0x8
> >   InfoElement                     = <80 90>
> >
> > INFO_RESP ID=001 #0x3473 LEN=0012
> >   Controller/PLCI/NCCI            = 0x101
> >
> >     -- ISDN1: info element CAUSE 80 90
> > DISCONNECT_B3_CONF ID=001 #0x2927 LEN=0014
> >   Controller/PLCI/NCCI            = 0x10101
> >   Info                            = 0x0
> >
> > DISCONNECT_B3_IND ID=001 #0x3474 LEN=0015
> >   Controller/PLCI/NCCI            = 0x10101
> >   Reason_B3                       = 0x3301
> >   NCPI                            = default
> >
> > DISCONNECT_B3_RESP ID=001 #0x3474 LEN=0012
> >   Controller/PLCI/NCCI            = 0x10101
> >
> > DISCONNECT_REQ ID=001 #0x2928 LEN=0017
> >   Controller/PLCI/NCCI            = 0x101
> >   AdditionalInfo
> >    BChannelinformation            = default
> >    Keypadfacility                 = default
> >    Useruserdata                   = default
> >    Facilitydataarray              = default
> >
> > DISCONNECT_CONF ID=001 #0x2928 LEN=0014
> >   Controller/PLCI/NCCI            = 0x101
> >   Info                            = 0x0
> >
> >     -- Executing NoCDR("SIP/113-a753", "") in new stack
> >     -- Executing Wait("SIP/113-a753", "5") in new stack
> > DISCONNECT_IND ID=001 #0x3475 LEN=0014
> >   Controller/PLCI/NCCI            = 0x101
> >   Reason                          = 0x3490
> >
> > DISCONNECT_RESP ID=001 #0x3475 LEN=0012
> >   Controller/PLCI/NCCI            = 0x101
> >
> >        > CAPI INFO 0x3490: Normal call clearing
> >   == ISDN1: Interface cleanup PLCI=0x101
> >     -- Executing Hangup("SIP/113-a753", "") in new stack
> >   == Spawn extension (macro-hangupcall, s, 4) exited non-zero on
> > 'SIP/113-a753' in macro 'hangupcall'
> >   == Spawn extension (from-internal, h, 1) exited non-zero on
> > 'SIP/113-a753'
> > rapid*CLI>
> > Disconnected from Asterisk server
> >
> > Regards
> >
> > Andrew Gough
> >


OK this here is the backtrace, this crash happened when there was noone
in the office, I don't know whether there was an inbound call at the
time or not. I hope this is what you need, Let me know if you need
anything else. There does not seem to be any pattern to the crashes.
Today I was on an outbound call over CAPI ISDN for 25 minutes, during
which several other inbound, outbound calls were made, and suddenly it
crashed, ending my call. Then the crash BTed below happened when there
was noone in the office. Any help would be much appreciated, as I will
have to regretfully give up on Asterisk if I can't get a solution. If
you need the actual core dump it is at
http://dev.gcdtech.com/core.tar.gz 

Andrew Gough

#0  0x400268b7 in pthread_mutex_trylock () from /lib/tls/libpthread.so.0
No symbol table info available.
#1  0x0806c146 in ast_mutex_trylock (pmutex=0x672e33fc) at lock.h:597
No locals.
#2  0x0806175a in ast_queue_hangup (chan=0x672e3330) at channel.c:671
        f = {frametype = 4, subclass = 1, datalen = 0, samples = 0,
  mallocd = 0, offset = 0, src = 0x0, data = 0x0, delivery = {tv_sec =
0,
    tv_usec = 0}, prev = 0x0, next = 0x0}
#3  0x408fc2d9 in __sip_autodestruct (data=0x81be208) at chan_sip.c:1315
        p = (struct sip_pvt *) 0x81be208
#4  0x08056c3e in ast_sched_runq (con=0x8172f28) at sched.c:373
        current = (struct sched *) 0x8174868
        tv = {tv_sec = 1135275568, tv_usec = 989877}
        x = 0
        res = 1083432672
#5  0x40927e28 in do_monitor (data=0x0) at chan_sip.c:11253
        res = 0
        sip = (struct sip_pvt *) 0x0
        peer = (struct sip_peer *) 0x0
        t = 1135275568
        fastrestart = 0
        lastpeernum = -1
        curpeernum = 6
        reloading = 0
#6  0x40024b63 in start_thread () from /lib/tls/libpthread.so.0
No symbol table info available.
#7  0x401ac18a in clone () from /lib/tls/libc.so.6
No symbol table info available.




More information about the asterisk-users mailing list