[OpenSIPS-Users] rtpproxy wrong source interface
jpyle at fidelityvoice.com
Tue Aug 6 00:16:39 CEST 2013
Ok, this is humbling.
I didn't *change* anything, but I did reboot the system. It seems I never
added my private routes to the system startup scripts. It had been up for
so long previously it didn't even cross my mind.
OSI model schmo-SI model. Thanks for the reality check! :)
On Mon, Aug 5, 2013 at 5:51 PM, Laszlo <laszlo at voipfreak.net> wrote:
> Hi Jeff,
> Did you change anything in the routing inside linux?
> If you try to connect to an internal host (like ping, ssh or anything),
> the connection attempt is originated from the right ethernet interface?
> I saw this behaviour before on one of my customer's server, caused by a
> wrong ip route entry.
> 2013/8/5 Jeff Pyle <jpyle at fidelityvoice.com>
>> I'm using rtpproxy 1.2.1 with Opensips 1.9.0 r9980. The configuration is
>> multihomed, one public interface and one private. Calls that come from the
>> public side t_relay() to the private one and vice versa with rtpproxy in
>> the middle to handle the media.
>> Everything was working fine. And now it's not. As far as I can tell
>> nothing has changed. Obviously something has, but I can't determine what.
>> Inbound calls (public to private) invoke rtpproxy with
>> engage_rtp_proxy("feirocn"). These are fine. Outbound calls (private to
>> public) are invoked with engage_rtp_proxy("fierocn"). These are not fine.
>> When rtpproxy sends the RTP packets from its private interface to the
>> private host, it sends them from the wrong interface. The IPs are correct,
>> but the interface is not.
>> On debian the /etc/default/rtpproxy file contains:
>> EXTRA_OPTS="-l PRIVATE_IP/PUBLIC_IP -m 16384 -M 23552 -F -d WARN"
>> What could cause this odd behavior?
>> Users mailing list
>> Users at lists.opensips.org
> Kind regards,
> Laszlo Bekesi
> Users mailing list
> Users at lists.opensips.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Users