[RTPproxy Users] Bug in bridging mode with NAT?

[Christian Koch]

Thanks!!! With that flag it works without modifying the rtpproxy code. I 
simply missed that flag. Sorry for the confusion.

Ovidiu Sas schrieb:
> OpenSER has a flag 'w' that forces support symmetric RTP.
> http://www.openser.org/docs/modules/1.3.x/nathelper.html#AEN316
>
> Regards,
> Ovidiu Sas
>
> On Wed, May 21, 2008 at 8:50 AM, Christian Koch
> <chri.koch.vier at googlemail.com> wrote:
>   
>> Hi,
>>
>> I've been facing a problem with rtpproxy in bridging mode where one
>> client is behind a NAT (for details please see
>> http://lists.openser.org/pipermail/users/2008-May/017388.html). The
>> natted client didn't get the RTP data, because it wasn't sent to the NAT
>> hole. It seems the solution is to remove the following line from the
>> rtpproxy code:
>>
>>    asymmetric = (bmode != 0) ? 1 : 0;
>>
>> After recompiling everything works fine. Now I'm wondering why rtpproxy
>> assumes clients are asymmetric in bridging mode? Will I get problems
>> when running rtpproxy without above code line?
>> Also I think assuming asymmetric clients shouldn't be the default
>> behaviour. If clients are really asymmetric this can be set with the
>> "a"-flag in force_rtp_proxy(). But if clients are symmetric it can only
>> be fixed by changing the code. At least there should be a flag to force
>> rtpproxy to handle symmetric clients. Or am I missing the point?
>>
>> Thanks for your comments,
>> Christian
>> _______________________________________________
>> Users mailing list
>> Users at rtpproxy.org
>> http://lists.rtpproxy.org/mailman/listinfo/users
>>
>>     
> _______________________________________________
> Users mailing list
> Users at rtpproxy.org
> http://lists.rtpproxy.org/mailman/listinfo/users
>