chan_sip.c:3245 auto_congest: Auto-congesting

[kamirie]

Good Day,
I've been trying to figure out what's wrong in this setting , whenever i tried to dial out , I'm ending with auto-congestion , I already tried setting qualify=(specific number ) or no but still having problems. This same trunk is being used on a FreePBX and working , my problem was making it work on our vicidial dialer. I already specify ulaw as codec on soft phone and IP phone

Code: Select all

[trunkname]
username=username
type=peer
secret=password
reinvite=yes
realm=voice.mibroadband.com.au
qualify=yes
insecure=invite,port
host=trunk.engin.com.au
fromdomain=trunk.engin.com.au
fromuser=xxxxxxxx
disallow=all
allow=ulaw
dtmfmode=rfc2833
auth=md5
canreinvite=no


Code: Select all

register => username:password@trunk.engin.com.au/xxxxxxx

Code: Select all

engin2 = SIP/trunkname

Code: Select all

Dial Plan
exten => _80XXXXXXXXX,1,AGI(agi://127.0.0.1:4577/call_log)
exten => _80XXXXXXXXX,2,Dial(SIP/engin2/${EXTEN:1},,tTor)
exten => _80XXXXXXXXX,3,Hangup


Code: Select all

[Aug 16 20:49:05] NOTICE[18523]: chan_sip.c:13822 handle_response_peerpoke: Peer 'trunkname' is now Reachable. (341ms / 2000ms) 

SIP show peers
trunkname/username      203.161.160.71       N      5060     OK (341 ms)

SIP show registry
trunk.engin.com.au:5060         username        3585 Registered           Fri, 16 Aug 2013 20:49:01  [color=#FF0000]<< does this mean something? Refresh is 3585[/color]


Code: Select all

[b]CLI LOG[/b]

 -- Executing [80XXXXXXXXX@default:1] AGI("SIP/2010-00000002", "agi://127.0.0.1:4577/call_log") in new stack                           
    -- AGI Script agi://127.0.0.1:4577/call_log completed, returning 0                                                                   
    -- Executing [80XXXXXXXXX@default:2] Dial("SIP/2010-00000002", "SIP/engin2/0XXXXXXXXX||tTor") in new stack                           
    -- Called engin2/0734145060                                                                                                                                                                                                                           
[Aug 16 20:37:42] NOTICE[14437]: chan_sip.c:3245 auto_congest: Auto-congesting SIP/engin2-00000003                                       
    -- SIP/engin2-00000003 is circuit-busy                                                                                               
  == Everyone is busy/congested at this time (1:0/1/0)                                                                                   
    -- Executing [80XXXXXXXXX@default:3] Hangup("SIP/2010-00000002", "") in new stack                                                     
  == Spawn extension (default, 80XXXXXXXXX, 3) exited non-zero on 'SIP/2010-00000002'                                                     
    -- Executing [h@default:1] DeadAGI("SIP/2010-00000002", "agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----16-----CONGESTION----------") in new stack                                                                                                               
    -- AGI Script agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----16-----CONGESTION---------- completed, returning 0 


[williamconley]

http://hotfile.com/dl/139441882/70b7cf9 ... 1.pdf.html <- I don't go to hosted solutions that don't just provide the file based on clicking the link. Put it on your wordpress page or something. LOL (Millions of free hosting sites available that don't try to extract money from your downloaders!).

autocongesting means asterisk cannot get a roundtrip packet to/from the sip account in the pre-defined time. It may also be an indication that it never got the packet back at all. set qualify=no on the account in question to disable this "feature", but be aware that if the packet never returns, your call will still fail. This feature is designed to allow auto-failover to a functioning trunk when a trunk becomes unreliable. If you don't have failover ... obviously this is not as helpful.

note also that your "sip show peers" should not show "ok" during this moment. However, your supplied samples show a fake name for the trunkname but a real name in the sip account. why would you block it in one but not the other? possible that they are different. IF you don't show us the same sip account for sip show peers as you are trying to send a call to ... that could be related to your problem.

you show Dial("SIP/2010-00000002", "SIP/engin2/0XXXXXXXXX||tTor"), but your sip account shows "[trunkname]" not "engin2". Perhaps you really named your sip account [trunkname]? in which case you should change that to [engin2] to match the SIP/engin2 in the dial command.

happy hunting.

and ... you should reinstall with vicibox 4.0.3 (or try the new 5.0.2, which we aren't using yet but hear is very good). you'll find that the installation goes smoother and getting the system online is much smoother. you are using a Very Old version of Vicidial

[kamirie]

sorry william , I edited the sig , from a scratch install I mostly followed the guide in the link. everything was working fine until the client changed providers.

I had this in the carrier Global String filed : engin2 = SIP/trunkname

before , if i removed the entry to global String i will have this error

Code: Select all

   -- Executing [8XXXXXXXXX@defaultlog:1] AGI("SIP/2000-00000000", "agi-NVA_recording.agi|BOTH------Y---Y---Y") in new stack 
    -- Launched AGI Script /var/lib/asterisk/agi-bin/agi-NVA_recording.agi                                                     
    -- AGI Script Executing Application: (Monitor) Options: (wav|/var/spool/asterisk/monitor/MIX/2013082092205_2000_XXXXXXXXX)
    -- AGI Script agi-NVA_recording.agi completed, returning 0                                                                 
    -- Executing [8XXXXXXXXX@defaultlog:2] Goto("SIP/2000-00000000", "default|8XXXXXXXXX|1") in new stack                     
    -- Goto (default,8XXXXXXXXX,1)                                                                                             
    -- Executing [8XXXXXXXXX@default:1] AGI("SIP/2000-00000000", "agi://127.0.0.1:4577/call_log") in new stack                 
    -- AGI Script agi://127.0.0.1:4577/call_log completed, returning 0                                                         
    -- Executing [8XXXXXXXXX@default:2] Dial("SIP/2000-00000000", "SIP/engin2/XXXXXXXXX||tTor") in new stack                 
    -- Called engin2/XXXXXXXXX                                                                                                 
    -- Got SIP response 604 "Does not exist anywhere" back from 203.161.160.71                                                 
  == Everyone is busy/congested at this time (1:0/0/1)                                                                         
    -- Executing [8XXXXXXXXX@default:3] Hangup("SIP/2000-00000000", "") in new stack                                           
  == Spawn extension (default, 8XXXXXXXXX, 3) exited non-zero on 'SIP/2000-00000000'                                           
    -- Executing [h@default:1] DeadAGI("SIP/2000-00000000", "agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----1-----CHANUNAVAIL----------") in new stack                                                                                           
    -- AGI Script agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----1-----CHANUNAVAIL---------- completed, returning 0


then I tried to change the entry in global settings to engin2 = SIP/user:pass@trunk.engin.com.au

and now even if I removed the entry of global string in carrier settings , I'm no longer being prompted with the 604 "Does not exist error" but the auto-congest error. I already tried setting qualify=no but it didn't made any difference.

I will try to download vicibox and do test install. We are using this server still becuase my manager wanted to stay with Debian boxes.

[williamconley]

when making such a shift, it should have bothered you that the old entry was simple:

engin2 = SIP/trunkname

but the new entry was more complex and not likely suited to the same purpose:

engin2 = SIP/user:pass@trunk.engin.com.au

The reason that the first one does not have a user/pass and an "@" symbol is because those things are supplied within the SIP Context of [trunkname].

So ... you would modify the [trunkname] content to have the new user/pass and domain name, and not modify the global variable "engine2=SIP/trunkname" at all. Or you create a whole new sip context like [trunk2name] and put all the necessary information inside it just like the other account information was in [trunkname], which requires creation of a whole new global variable. Then you'd have both of them and could switch "engin2" to point to either easily.

ordinarily, you don't modify the old one at all, you merely make a whole new one and deactivate the old one (so you can go back by changing which one is active, one at a time only).

our preferred method is to have a different dialplan entry for each so they can both be active at the same time. in this situation, you would have one that starts with "9" and one that starts with "8" and the dial prefix in the campaign would change to choose providers. both active at the same time. and you can have two different campaigns on two providers both running at the same time without even interfering with one another.

Lots of choices. Pick one!

[kamirie]

I stick with the first one engin2=SIP/trunkname and had the trunkname details in the account entry in vicidial , But I can't still find the cause of auto-congestion while doing outbound calls

[williamconley]

the cause of autocongestion is lack of contact with the sip account's ip address. that's what autocongestion is for

[kamirie]

Ok , william , I'll double check with the provider if our end is indeed reaching their servers .. Thanks

[williamconley]

it's not a question of whether they think they got it. it's a question of whether your asterisk server gets a response.

asterisk sends the packet and gets no response = autocongestion.

at the moment of autocongestion you should also have "unreachable" as the status for the peer

and asterisk's sip debug should complain at some point before the attempt that this "peer" has become unreachable.

of course, it may also never have been reachable in the first place, in which case there would be no such complaint anywhere near the attempt because it may have been that way for quite some time.

qualify=no may resolve this, but not if the peer refuses traffic. i've only had one instance in 6 years where qualify=no actually fixed the problem (but ... that means I did have ONE, LOL). usually qualify=no just shows "unmonitored" and the system actually attempts the call ... which fails. but not due to auto-congestion!

[Rubal]

Hi,

I am getting exactly the same error/issue with Engine Outbound on ViciDial.

-- Got SIP response 604 "Does not exist anywhere" back from 203.161.160.71
== Everyone is busy/congested at this time (1:0/0/1)

Any ideas how this was resolved?

[williamconley]

auto-congesting does not include "response 604". That error is your carrier refusing to complete the call and telling you Why. The reason in this case is usually that the number is not valid (it includes a country code, but not the international 011 prefix ... or it has the wrong number of digits ... or it's to an area code that is not valid).

Auto-congesting is when the asterisk server determines that the sip connection is not "viable" and refuses to try. This happens if the sip packets are never returned during a "qualify" check. In that situation there is no "response 604" in fact there is not any response because asterisk never bothered to make the call (because the qualify packet never got back or took too long to return, thus DISqualifying the sip connection).