Problems integrating VCS and Lync via B2BUA

[megafonmsk]

Hello!

Could you please help me with correct B2BUA settings?

Now I am trying to migrate from direct SIP trunk to Lync in VCS X6.1 to B2BUA application on VCS 7.3 fully following "Cisco_VCS_Microsoft_Lync_2010_Deployment_Guide_X7-2" and ran into some problems.

My configuration:

• Video network being controlled by the same Cisco VCS that is the interface to Lync server
• Cisco VCS is a cluster with 2 peers
• No transforms being applied before search rules
• Search rule to B2BUA zone is right behind "LocalZoneMath" rule with alias pattern match ".+@mydomain\.com.*"
• Integration to Lync server based on TCP(not TLS).
• Lync server has a trusted application with 2 application hosts (2 peers of VCS-C cluster) on port 65072.
• Lync server has a static route to shared SIP domain to one of peers of VCS-C cluster on port 65072.
• FindMe aliases is to be registered on Lync.
• We use AMGW to convert H.264 - Microsoft RTV. AMGW is in the same VLAN with VCS-C cluster.

Presence of FindMe aliases works fully correct - Lync immediately changes presence status based on activity of FindMe devices.

When I am calling from Lync to FindMe alias, call works correct, signalling and media(across AMGW) to any FindMe device establishes without any problems.

When calling in the opposite direction (from VCS devices to Lync associated FindMe alias), I experience the following problem:

Lync client correctly changes presence status of caller's alias, I respond to the call in Lync, see the remote video, but i cannot start my video, and the call breaks after 30 seconds.Seems this is a problem with media negotiation...

I will be grateful for any advice or suggestion.

[awinter2]

Hi,

for a shared-domain integration with FindMe, it is not recommended to create a static route for the shared domain from Lync towards the VCS. Since the B2BUA will register FindMe accounts onto OCS/Lync, no static route is needed to reach these FindMe accounts. Also, having the static route in place for the OCS/Lync domain could cause OCS/Lync to send a lot of SIP "garbage" messages towards the VCS, which is not ideal.

Now for the problem you are describing, you would have to collect a diagnostics log from the VCS (Log levels set to DEBUG) so that we can see what's going on on the signaling level, since this is where the media negotiation is done.

For this it would probably be good to open a TAC case, since you probably shouldn't post any VCS logs on this forum.

[megafonmsk]

Hello, Andreas!

Thank you for answer!

I am aware of the negative impacts of static routs, but it's more convenient to me for the moment.

I will open TAC case and provide debug logs.

Best regards, Evgeny.