[asterisk-users] Re: Wrong account code from iax_buddies

[Tim Panton]

On 15 Jul 2006, at 19:26, voiplist wrote:

> On 7/15/06, Tim Panton <tim at mexuar.com> wrote:
>>
>> On 15 Jul 2006, at 17:24, voiplist wrote:
>>
>> > After further testing, here is what we found..
>> >
>> > The account code was actually right after all, what made us  
>> think it
>> > was incorrect was the fact that * was reporting the wrong  
>> channel for
>> > the call.
>> >
>>
>> >
>> > Test 2:
>> >
>> > We turned off the softphones mentioned above. We then setup a test
>> > Asterisk box with the same IAX accounts mentioned above. Each  
>> account
>> > registered to our server remotely just as the softphones did. We  
>> then
>> > made 10 phone calls with each account dialing totally different  
>> phone
>> > numbers so we could identify the records from each IAX account.
>> >
>> >
>> > Results (Test 2):
>> >
>> > We found that the majority of calls now had wrong/mismatched  
>> channels
>> > in the CDR. Basically we would show a channel like "IAX2/user2"  
>> for a
>> > call that we are certain came from the IAX user "user1".
>> >
>> >
>> > And... That's my story and I'm stickin' to it :)
>>
>> Hmm, I'd like to know more about this, as it is likely to bite me
>> too :-(
>>
>> Can you describe the setup of the 'originating' asterisk in your
>> second test?
>> If possible please share the relevant bits of extensions.conf and
>> iax.conf.
>>
>> Thanks.
>> Tim.
>>
>>
>> Tim Panton
>> tim at mexuar.com
>
> Sure it was TrixBox 1.1 running on a VmWare virtual machine (great for
> testing). We basically setup two "Trunks" as they are known is TrixBox
> (Asterisk at home), one with each of the IAX accounts we setup on the
> server end. Just to keep the confusion to a minimum, the TrixBox was
> the client which was sending the calls to the Asterisk Server which
> was terminating the calls to the PSTN.
>
> If you are familiar with TrixBox Trunk Configs I can send you the
> basic details of the settings in the GUI Trunk screen.

Ah, no, I am using the base asterisk facilities, plus a bit
of AGI magic.
You would need to strip the test case down to just asterisk
before you could be sure where the problem is.


Tim Panton

www.mexuar.com