SIPing Aterisk to Asterisk

Asterisk Asterisk Support
1

Hi All,

We have an issue where we call from an Asterisk box to another via SIP. Calls normally go through fine and there are no issues, however at a random interval (with no pattern) the server making the call will timeout. The server receiving calls will not get anything come up, and if i try to run “core show channels” it comes up blank, to the point it doesn’t even show how many calls processed etc.

Dialer sip.conf

[general] disallow=all allow=alaw

Receivers sip.conf

[general] context=IncomingFromGateWay disallow=all allow=alaw allow=slin

The dial command (which is strange, but needed)

exten => s,n,Set(SIPADDRESS=SIP/${ASMIPADDRESS}/MV_${OTHER}_${DELIMITED}_${INFO})
exten => s,n,Dial(${SIPADDRESS},20,t)
;eg, Dial(SIP/192.168.42.1/MV_Other_Delimited_info,20,t)

As I said, this will work for the most of the time, until Asterisk crashes on the receiving end. We currently restart the Asterisk service to fix this.

If anyone knows why this might be happening, some advice would be great.

Asterisk is 1.8.5
Centos is 6.3
If any other info is need, please let me know

Thanks
Alex

2

You will need to provide logging. Start at verbose level 5 on the originating machine, but you may need to go both verbose and debug 5 and enable sip debugging on both machines.

3

Hi David, Thankyou for the quick reply.

I have set the verbosity to 5 on both and below are the output from the logs;

Sender:

[Apr 12 13:13:35] VERBOSE[11705] pbx.c: -- Executing [s@mv-SipAppSM:40] Dial("Local/s@mv-BalanceTest-8d05;1", "SIP/192.168.42.15/MV_CONFLEG_3255477_177814_0_1_0_B_0_0_0_dcTMp_177814^2106778,20,t") in new stack [Apr 12 13:13:35] VERBOSE[11705] netsock2.c: == Using SIP RTP CoS mark 5 [Apr 12 13:13:35] VERBOSE[11705] app_dial.c: -- Called SIP/192.168.42.15/MV_CONFLEG_3255477_177814_0_1_0_B_0_0_0_dcTMp_177814^2106778 [Apr 12 13:13:35] VERBOSE[11705] app_dial.c: -- SIP/192.168.42.15-000044b3 is circuit-busy [Apr 12 13:13:35] VERBOSE[11705] app_dial.c: == Everyone is busy/congested at this time (1:0/1/0) [Apr 12 13:13:35] VERBOSE[11705] pbx.c: -- Executing [s@mv-SipAppSM:41] GotoIf("Local/s@mv-BalanceTest-8d05;1", "0?siptestcontinue") in new stack [Apr 12 13:13:35] VERBOSE[11705] pbx.c: -- Executing [s@mv-SipAppSM:42] Hangup("Local/s@mv-BalanceTest-8d05;1", "") in new stack [Apr 12 13:13:35] VERBOSE[11705] pbx.c: == Spawn extension (mv-SipAppSM, s, 42) exited non-zero on 'Local/s@mv-BalanceTest-8d05;1'

Receiver:

[Apr 12 13:13:35] ERROR[2290] res_rtp_asterisk.c: Oh dear... we couldn't allocate a port for RTP instance '0x7f5e6a2fa9e8' [Apr 12 13:13:35] ERROR[2290] chan_sip.c: Got SDP but have no RTP session allocated.

This has already been of great help as the error on the Receiving Asterisk was not before coming up! That crazy verbosity!

Now to find out why an RTP instance couldnt be created

4

Looks like there may be a patch to this issue for a later version of Asterisk 1.8.8.1
issues.asterisk.org/jira/browse/ASTERISK-19192

Can anyone tell me if updating my versions of Asterisk to 1.8.21.0 would include this patch and so fix the issue?

5

It was marked as a blocker for 1.8.9.0, so 1.8.9.0 wouldn’t have been released without its having been fixed.