Strange things are going on.
Next time, please send only one dump. under Linux, you can capture from
both interfaces at the same time ("any").
To me, it looks like the call is beeing forked - one branch goues to the
second cisco (correct) and one goes to the proxy again. Than, the
looped request is again forked.
So, I think in your location database for sip:1006@10.0.0.54 are two
entries:
sip:1006@10.0.0.54 and
sip:1006@10.0.0.54@lima.dsi.crc.ca
the second one is looped to the same account and causes the forking like
this:
-- sip:1006@10.0.0.54
/
- < - sip:1006@10.0.0.54
\ /
--sip:1006@10.0.0.54@lima...ca-<
\
-sip:1006@10.0.0.54@lima.....ca
solution: clean the user location entries.
regards,
Klaus
Ticknor.Scott(a)ic.gc.ca wrote:
>
>>-----Original Message-----
>>From: Klaus Darilion [mailto:klaus.mailinglists@pernau.at]
>>Sent: Wednesday, February 18, 2004 12:53 pm
>>To: Ticknor, Scott: DGSE
>>Cc: serusers(a)lists.iptel.org
>>Subject: Re: [Serusers] eliminating 483 errors
>>
>>
>>Strange. You should watch the network traffic also on the loopback
>>interface. (ethereal supports this). you can send me the dump
>>file from
>>ethereal.
>>
>>klaus
>
>
> thanks Klaus, here are the dumps.
>