Hi all,

I seem to have fallen off the list!  No mail since 7/26/2013.  Have I made someone mad??  J

I have also been blissfully running with ISSUES what so ever since that time.

I have a NATed client with PERIODIC one way audio.  The client end has a Cisco 220W router with SIP ALG disabled.  The phones are CISCO 504Gs and Polycom 330s(?), basciallt the same as everywhere else we have phones.

I did a sofia global sip trace on for a while and captured the output.  After writing an app to sort those sip packets into logical stream files, I dug in to the SIP conversations.  This has left me with som questions for the people who know these things:

1.   I found all my phones are REGISTERing every 23 seconds!   That seems absurd TO ME, so where have I screwed that up/where do I set it and what is a reasonable number?

2.   The only difference I see in the SIP packets from my working phone and these non-working ones is of course NAT related.  The working phone’s REGISTER looks like this:

recv 757 bytes from udp/[WW.XX.YY.93]:5063 at 17:53:12.164203:
------------------------------------------------------------------------
REGISTER sip: <MYDOMAIN.COM> SIP/2.0
Via: SIP/2.0/UDP WW.XX.YY.93:5063;branch=z9hG4bK-238fcc8b;rport

But the failing phones look like this:

recv 801 bytes from udp/[AA.BB.CC.38]:5104 at 17:53:40.693691:

------------------------------------------------------------------------

REGISTER sip: <MYDOMAIN.COM> SIP/2.0

Via: SIP/2.0/UDP 10.2.2.239:5104;branch=z9hG4bKb374d6507894B431

The response is still sent to [AA.BB.CC.38]:5104

I guess first, IS THAT A PROBLEM?

So, can anyone share any insight into the CISCO 220W config that I might be missing? 

Why would it only be a periodic problem?  Could this be a problem where one end is using a wider range of rports than the other supports?

Thanks,

Sean