sipgate dropped calls and asterisk ID question

[bigbloke]

Hi All,

Firstly, in case it makes a difference, I am located in the UK.

I have a regular PSTN line that is "divert on busy" to my Sipgate.co.uk DDI.

This is forwarding calls to my PIAF box which routes to
ZAP1/1 for answering.

Asterisk is behind a NAT boundary, mapped with a dyndns_custom account.

when a call comes in from sipgate:

zap1/1 rings, CLI is presented;

I answer the call;

Two way speech is established;

after 17 seconds The call is released without warning.
Zap1/1 gets "congestion" the PSTN caller gets silence.

this is also the case if I dial in directly to my sipgate DDI (i.e. not via call divert)

I notice in all the logs that asterisk is transmitting its name as [email protected] sipgate also picks this up as

<[email protected]>

I have tried several ways to set this to my dns name, (e.g. externhost = pbx.my_own_tld.com in sip_nat.conf) but have have failed to resolve this.

IIRC Asterisk doesnt interwork with STUN which is how I assume that Sipgate's own client resolves the naming problem. It reports <mySIPGATEddi@my_public_ip>

but whilst it seems that asterisk is incorrect, it nevertheless somehow manages to route calls to the right IP and setup RTP voice streams.

I am kind of assuming that this is some kind of timer related issue whereby 1/1either party is not accepting / sending the answer notification, and 17 seconds after the call sets up the "circuit" is released.

all other inbound trunks (voipuser.org/voipcheap.com) work 100% just sipgate

It appears that zap1/1 releases the call (I see that in the asterisk CLI logs)

I'm reluctant to flood the forum with yards of logs straight off, prefering someone better qualified than myself to tell me specifically what part is of interest.

I'm planning to escalate this to sipgate (had one problem their end already with inbound calls resulting in busy tone and no sip reads directed to asterisk) but would feel much happer doing so knowing my own house was in order first.

grateful for any assistance resolving this

MTIA

Regards

BB

 

[bigbloke]

read many, reply once ;-))

solved the problem...re-discovered another

Kudos to Mr Roper's plus others reply in the externhost thread

I didnt appreciate you could set externip=myhost.mytld.com

It certainly seems FAR more effective than externhost!

so here's the solution:

vi /etc/hosts

look at your hostname in there

if the full entry is:

127.0.0.1 myhost.mytld.com

edit it to read:

127.0.0.1 myhost

then save the change ([esc]:wq[enter])

cd /etc/asterisk
vi sip_nat.conf

my sip_nat.conf now consists of:

externip=myhost.mytld.com
externrefresh=120
localnet=192.168.1.0/255.255.255.0

save the change again ([esc]:wq[enter])

and that seems to have fixed it !!

but I'm pushing calls into the "from-trunk" context and FOP shows no calls ?? very wierd !!

regards

BB