Jabber for Windows outbound SIP calls are dropping after 15 minutes

[petark]

Hi All,

I have strange problem. Jabber for Windows outbound SIP calls to C series and MX series endpoints are dropping after exactly 15 minutes. Calls to MXP series endpoints are not affected. The default timer for SIP session is set to 1800 sec.

I have analyzed call logs but couldn't find anything relevant so I've come to the experts.

Any help would be greatly appreciated.

[aaron dempsey]

Try changing the SIP Session Expires Timer.  Had this problem on CIPC running SIP and Nortel SSL SIP configurations.

[petark]

I did change this timer before and did not help.

We have reproduced this issue today on C40. Call dropped after 15min17s.

[Wayne DeNardi]

I'm assuming your Jabber Client and room are on different sides of a firewall - you haven't clearly stated that in your posts.  Can you please also confirm which "Jabber" you're using "Jabber" or "Jabber Video for TelePresence".

From previous experience, this issue is 99% of the time, a firewall problem, where the firewall is closing the ports as it incorrectly thinks they are no longer in use or a session timer has expired.

Note: Your post would probably be better placed in the Jabber Clients section of the forums than in the Other Collaboration section you have posted in.

Wayne
--
Please remember to rate responses and to mark your question as answered if appropriate.

[petark]

Wayne, 

I was thinking the same thing, but there is no firewall between Jabber Client and Endpoints. If the firewall was the issue it would affect the MXP endpoints too, but calls to MXP series endpoints were not affected. 

Just to confirm, as I have already said in the title of this discussion, we are using Jabber for Windows. 

Is there a way to move this discussion to the right section? Since I am new on this Community, not really sure I know how to do this.

Regards

[Wayne DeNardi]

Regarding moving your discussion: If you press the big Edit button at the top of the page you should be able to change the section the discussion is in in the small section that appears just under that. 

But, further on your main question: About the only other time I've seen this issue occurring on a local network where there are no firewalls in between any of the equipment, is where there was interworking being done on the VCS back and forth between SIP and H.323.  Is there any difference in the way your devices are registered (MXPs vs the others)?

If you're using Jabber, then those clients will be SIP registered to your CUCM, so, if your VCS registered endpoints are registered using H.323, then some interworking may be occurring.

Same with Jabber Video, that'd be SIP registered to your VCS, so if your endpoints are H.323 registered, there will also be some interworking happening.

I've also seen it occur when some endpoints were dual registered SIP and H.323 where there were issues with the Search Rules on the VCS which caused some interworking to occur when it wasn't necessary due to incorrect configuration and search rule ordering - so that could also be a place to check on your VCS - which search rules you have, and which one is being matched.

Wayne
--
Please remember to rate responses and to mark your question as answered if appropriate.

[petark]

It looks like we solved the problem. After doing lot of tests and logs analysis, we came to a conclusion it was the VCS issue at the end. Specifically, the iX protocol.

We've changed "SIP UDP/IX filter mode" to "On" at the VCS advanced zone settings and everything works fine now.

Thank you for your time and assistance in this matter.

Petar