Summary:	ASTERISK-07033: Dial timers fail on SIP-SIP calls
Reporter:	John Todd (jtodd)	Labels:
Date Opened:	2006-05-25 01:22:37	Date Closed:	2006-09-06 12:26:57
Priority:	Major	Regression?	No
Status:	Closed/Complete	Components:	Core/General
Versions:		Frequency of Occurrence
Related Issues:
Environment:		Attachments:	( 0) sip-sip-timers2.txt
Description:	I use timers for creating cascading rings through my dialplan for users.  All users are on SIP phones (varying device types.)  Calls from PRI Zap channels can reach SIP devices, and if no answer, the cascading timers allow calls to fall through to voicemail, alternate devices, etc. Currently, calls from one extension (SIP) to another extension (SIP) do not trigger the timers.  The Dial statement that is first executed continues forever or until answered.  This obviously makes voicemail and other functions of the dialplan fail entirely. OS: RH9.0 I have a quad-port PRI card installed (not that it should matter.) I am running SVN-HEAD as of 40 minutes ago. ****** ADDITIONAL INFORMATION ****** Dialplan example: ; 2405 = John Todd exten => 2405,1,Set(TIMEOUT(absolute)=10800) exten => 2405,n(dial),Dial(SIP/2405&SIP/2598&SIP/2505&IAX2/2585&SIP/jhtodd@2588,9,tr) exten => 2405,n,NoOp( ${LEN(${CALLERID(num)})} ) exten => 2405,n,GotoIf($[${LEN(${CALLERID(num)})} < 10]?setcid:dial-out)   exten => 2405,n(setcid),Set(CALLERID(num)=650581${CALLERID(num)}) exten => 2405,n(dial-out),Dial(SIP/2405&SIP/2598&IAX2/2585&${PRIOUT}/13015551212&SIP/jhtodd@2588,20,tr) exten => 2405,n,Goto(voicemail,${EXTEN},1) exten => 2405,dial+101,Goto(voicemail,${EXTEN},1) I have several SIP phones that I use, many are not logged in.  While there is a lot of junk in this particular dialplan example, even the most simple Dial statement fails.  This is an old, sloppy example (still showing outmoded conditional jumps) but that shouldn't matter... right?    -- Executing [from-desks:1] Set("SIP/2588-fdf6", "TIMEOUT(absolute)=21600") in new stack    -- Channel will hangup at 2006-05-25 12:12:31 UTC.    -- Executing [from-desks:2] Goto("SIP/2588-fdf6", "from-desks2|2405|1") in new stack    -- Goto (from-desks2,2405,1)    -- Executing [from-desks2:1] Set("SIP/2588-fdf6", "TIMEOUT(absolute)=10800") in new stack    -- Channel will hangup at 2006-05-25 09:12:31 UTC.    -- Executing [from-desks2:2] Dial("SIP/2588-fdf6", "SIP/2405&SIP/2598&SIP/2505&IAX2/2585&SIP/jhtodd@2588|9|tr") in new stack 2006-05-24 23:12:31 WARNING[7184]: app_dial.c:1046 dial_exec_full: Unable to create channel of type 'SIP' (cause 3 - No route to destination)    -- Called 2598 2006-05-24 23:12:31 WARNING[7184]: app_dial.c:1046 dial_exec_full: Unable to create channel of type 'SIP' (cause 3 - No route to destination) 2006-05-24 23:12:31 WARNING[7184]: app_dial.c:1046 dial_exec_full: Unable to create channel of type 'IAX2' (cause 3 - No route to destination)    -- Called jhtodd@2588    -- SIP/2598-5c10 is ringing    -- SIP/2588-38ef is ringing  [ring, ring, ring... for a LONG time, then I hang up]  == Spawn extension (from-desks2, 2405, 2) exited non-zero on 'SIP/2588-fdf6'
Comments:	By: Serge Vecher (serge-v) 2006-05-25 08:21:40 jtodd: do you have a call logs with timeouts working to compare what's going on? By: John Todd (jtodd) 2006-05-29 00:10:21 Yes, I downgraded to 1.2 stable and I have added the output as an attachment of the same dial path.  Works fine with 1.2 but fails with SVN-HEAD. By: Serge Vecher (serge-v) 2006-06-28 09:36:25 jtodd: I've just had an idea. Can you please enable [debug] logging and post one log from 1.2 and another from trunk, so they can be analyzed for differences? Thanks. By: Serge Vecher (serge-v) 2006-06-28 09:40:13 Also, will the timeout function work if you are calling a SIP->Zap? I wonder if the problem is actually in the timeout function rather than chan_sip per se. By: Serge Vecher (serge-v) 2006-07-12 13:32:32 jtodd: need your response ... By: John Todd (jtodd) 2006-07-12 14:09:30 Sorry, haven't been able to keep up.  My Asterisk test box with Zap channels is currently too old (RH9) to compile Asterisk, and I've not yet upgraded it to test this problem again.  It's still on my to-do list; possibly this weekend will see results. By: Serge Vecher (serge-v) 2006-08-08 13:26:06 *ping* By: Serge Vecher (serge-v) 2006-08-21 14:52:30 ping again By: Serge Vecher (serge-v) 2006-09-06 12:26:57 ok, given that nobody else has confirmed this bug in 3 month timeframe, it's presumed that this was a problem in early trunk version. Please feel free to reopen if this is still reproducible with trunk r>42000.