[SR-Users] RTPProxy issue?

Thu Mar 19 15:19:33 CET 2015

Hello Maxim,

Just to let you know that I quick-fixed this issue by increasing the value compared into "ts_less" in rtp.c. Instead of the bitwise calculation, I set the maximum value of an unsigned integer.

I will schedule to test, qualify and replace all my instances with the new version. I will let you know if I reproduce the issue with the new version.

Regards,

Igor.

De : sr-users [mailto:sr-users-bounces at lists.sip-router.org] De la part de Maxim Sobolev
Envoyé : lundi 9 mars 2015 19:09
À : Kamailio (SER) - Users Mailing List
Objet : Re: [SR-Users] RTPProxy issue?

Igor, yes, I'd say give 2.0 a try and see if the problem is still there. There were tons of changes, particularly in the rtp_resize subsystem.

Thanks!

On Sun, Mar 8, 2015 at 7:31 AM, Igor Potjevlesch <igor.potjevlesch at gmail.com <mailto:igor.potjevlesch at gmail.com> > wrote:

Hello Maxim,

I'm running legacy 1.2 or 1.4, not sure.

I see in the latest code that the function is still there. Do you suggest to upgrade or there's a patch to make?

Regards,

Igor.

De : sr-users [mailto:sr-users-bounces at lists.sip-router.org <mailto:sr-users-bounces at lists.sip-router.org> ] De la part de Maxim Sobolev
Envoyé : samedi 7 mars 2015 09:14

À : Kamailio (SER) - Users Mailing List
Objet : Re: [SR-Users] RTPProxy issue?

Ah, ok, I see now. I did not realize you guys are using resizer. Which version of the software are you actually using? I.e. is it latest rel_2_0 / master, or some legacy 1.x code? We've done quite some revamping down there, so that it needs to be checked against the very latest code to make sure. Let us know.

Thanks!

On Mar 6, 2015 12:31 AM, "Igor Potjevlesch" <igor.potjevlesch at gmail.com <mailto:igor.potjevlesch at gmail.com> > wrote:

Hi Maxim,

Hard to do because it's in production.

I have a serious finding since yesterday on how this happened. 

My understanding is that the function "ts_less" returns FALSE into "rtp_resizer.c" because the timestamp between the two packets is > (1 << 31) [for example: 3740425320].

That's result in a drop of any following packets as I can see it into a capture.

Regards,

Igor.

De : sr-users [mailto:sr-users-bounces at lists.sip-router.org <mailto:sr-users-bounces at lists.sip-router.org> ] De la part de Maxim Sobolev
Envoyé : vendredi 6 mars 2015 07:44
À : Kamailio (SER) - Users Mailing List
Objet : Re: [SR-Users] RTPProxy issue?

Hi Igor, that's bit strange, since the rtpproxy is not checking any of the rtp flags including marker bit. It would help if you can post a tcpdump capture of the streams in question along with the log output of the rtpproxy running at the "dbug" level. Thanks!

On Mar 5, 2015 5:54 AM, "Igor Potjevlesch" <igor.potjevlesch at gmail.com <mailto:igor.potjevlesch at gmail.com> > wrote:

I reviewed again a call trace and I can be more precise: a RTP packet comes with a new SSRC and the Marker bit set to "True". This packet is properly forwarded.

Then, just after this packet, another RTP packet containing a new SSRC with the huge timestamp and the Marker bit set to "True" is coming from the UA. 

The RTPProxy stops forward since this packet.

Regards,

Igor.

De : Igor Potjevlesch [mailto:igor.potjevlesch at gmail.com <mailto:igor.potjevlesch at gmail.com> ] 
Envoyé : jeudi 5 mars 2015 11:34
À : miconda at gmail.com <mailto:miconda at gmail.com> ; 'Kamailio (SER) - Users Mailing List'
Objet : RE: [SR-Users] RTPProxy issue?

Hello,

Thank you.

Just to let you know, the RTPProxy is running in bridging mode.

Regards,

Igor.

De : sr-users [mailto:sr-users-bounces at lists.sip-router.org] De la part de Daniel-Constantin Mierla
Envoyé : jeudi 5 mars 2015 09:33
À : Kamailio (SER) - Users Mailing List
Objet : Re: [SR-Users] RTPProxy issue?

Hello,

maybe Maxim (cc-ed) will be able to provide more insights.

Cheers,
DAniel

On 04/03/15 16:59, Igor Potjevlesch wrote:

Hello,

I discovered an issue related to the handling of "timestamp" and/or "Marker bit" with rtpproxy (I use the latest Extension 20081224).

The call-flow is the following: one UA places a call to A and put this call on hold. Then, the same UA call another number B. Individual streams are ok.

When the UA tries to transfer A with B, the RTPProxy receive a RTP packet with a huge timestamp and the Marker bit set to "True".

Just after this RTP packet, RTPProxy stop forward the RTP packets from A to B. B to C is still working.

Anyone have an idea?

Regards,

Igor.

_______________________________________________
SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
sr-users at lists.sip-router.org <mailto:sr-users at lists.sip-router.org> 
http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users

-- 
Daniel-Constantin Mierla
http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Kamailio World Conference, May 27-29, 2015
Berlin, Germany - http://www.kamailioworld.com

_______________________________________________
SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
sr-users at lists.sip-router.org <mailto:sr-users at lists.sip-router.org> 
http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users

_______________________________________________
SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
sr-users at lists.sip-router.org <mailto:sr-users at lists.sip-router.org> 
http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users

_______________________________________________
SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
sr-users at lists.sip-router.org <mailto:sr-users at lists.sip-router.org> 
http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users

-- 

Maksym Sobolyev
Sippy Software, Inc.
Internet Telephony (VoIP) Experts
Tel (Canada): +1-778-783-0474
Tel (Toll-Free): +1-855-747-7779
Fax: +1-866-857-6942
Web: http://www.sippysoft.com
MSN: sales at sippysoft.com <mailto:sales at sippysoft.com> 
Skype: SippySoft

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-users/attachments/20150319/166a09d7/attachment.html>