[Spce-user] Problems with NAT-Fukction and mediaproxy

[Andreas Granig]

Hi,

On 10/31/2011 09:17 PM, Klaus Peter v. Friedeburg wrote:
> The PST-Gateway that we used also work with loadbalancer and inserted some contact-header. All of the items of the gateway platform are reacheable over public IP's but ce interpreted this a NAT-DEVICE! The consequence is, that all calls to PST are handled by mediaproxy.
> How is the way to tell ce that the PST-Gateway is not a NAT-DEVICE?

Basically it detects "NAT" (or better: uses mediaproxy) if the address
in the SDP is different from the IP it received the SIP message from.

In the upcoming release (later in November), there's already a peer
preference to never use mediaproxy for that peer. If you can't wait that
long, you need to change proxy.cfg.tt2 to not call rtpproxy_offer/answer
for calls from/to this peer.

> The second problem is that mediaproya (CE is running as a vmware guest) generate very small packets, on average only 160 bytes. The PST-Gateway signals a Ptime=20 that means that this gateway use Packetszise of 200 bytes. So that jitter don’t work and after a few minutes (3 to 5 minutes) the stream is interrupted.
> Is this a knowing problem by using ce as VMWARE guest?

The CE doesn't usually generate any packets, but does transparent
pass-through of the payload. Do a trace on the CE system before and
after the mediaproxy to compare the size of what's received by
mediaproxy and what's sent out on the other side. Then also do a trace
on the host system to see what's being received and passed into the
guest system and what comes out and gets send back onto the wire.
Generally speaking, it's not a known issue, no.

Andreas


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 900 bytes
Desc: OpenPGP digital signature
URL: <http://lists.sipwise.com/pipermail/spce-user_lists.sipwise.com/attachments/20111031/5fa8bc0e/attachment-0001.asc>