[Asterisk-Users] chan_skinny core dump

Jeremy McNamara jj at nufone.net
Wed Oct 15 16:27:21 MST 2003


Try latest cvs, this is nuking inside the Asterisk RTP stack.

Jeremy McNamara


CW_ASN wrote:

> Hi all:
>  
> I've got some core dumps with chan_skinny. The client is ATA186 with 
> v2.16.1.ms ata18x (Build 030814a). The * version is CVS-10/05/03-16:03:26.
> When I make a call, the phone connected with ATA rings only 1 time and 
> * dies. Maybe I have some errores in ATA config. If someone has proven 
> configs for ATA, please send me the details.
>  
> Thanks in advance,
>  
> Gus
>  
>  
> The logs:
>  
> *CLI> Version Request
> Received CapabilitiesRes
> Buttontemplate requested
> Recieved SoftKey Template Request
> Received SoftKeySetReq
> Received LineStateReq
> Received Time/Date Request
>     -- Executing Dial("SIP/2050-e4c0", "Skinny/2040 at 172.16.254.62 
> <mailto:Skinny/2040 at 172.16.254.62>") in new stack
> Found device: 172.16.254.62
>     -- skinny_request(2040 at 172.16.254.62 <mailto:2040 at 172.16.254.62>)
>     -- Skinny cw: 0, dnd: 0, so: 0, sno: 0
>     -- skinny_call(Skinny/2040 at 172.16.254.62-2 
> <mailto:Skinny/2040 at 172.16.254.62-2>)
>     -- Called 2040 at 172.16.254.62 <mailto:2040 at 172.16.254.62>
>     -- Skinny/2040 at 172.16.254.62-2 
> <mailto:Skinny/2040 at 172.16.254.62-2> is ringing
> Recieved Open Recieve Channel Ack
> Ouch ... error while writing audio data: : Broken pipe
>  
> (gdb) bt
> #0  0x0807989d in ast_rtp_set_peer (rtp=0x0, them=0x4913201c) at rtp.c:773
> #1  0x41d958de in handle_message (req=0x49200480, s=0x81303f0) at 
> chan_skinny.c:2169
> #2  0x41d93ab0 in skinny_session (data=0x81303f0) at chan_skinny.c:2295
> #3  0x4003b2b6 in start_thread () from /lib/tls/libpthread.so.0
> ; Skinny config
> ;
> [172.16.254.62]
> device=ata0008a3962bab
> ;device=ATA0008A3962BAB
> nat=0
>  
> callerid=2040
> mailbox=2040
> context=default
> linelabel="2040"
> line => 2040
>  
> callerid=2041
> mailbox=2041
> context=default
> linelabel="2041"
> line => 2041






More information about the asterisk-users mailing list