VCS-C/VCS-E - How to prevent some registered codecs to connect to Video bridge through VCS traversal?

[ysoulieres1]

Hi everyone! 

I'm wondering if anyone has enough experience/knowledge with the VCS platform to help me figure out how to get out of my current problem... 

Basically we have a VCS-C/VCS-E in place. Codecs in the main office register to VCS-C, coders in distant locations register to the VCS-E. 

I have a bridge internally that handles multi-protocol calls, and there is a zone set in the VCS so that the calls are forwarded to the internal conference nodes. 

That being said, we just recently deployed a new conference node in the APAC region, and if I de-register codecs in that region, through DNS, they will connect to that conference node, which optimizes the connection. 

The issue is that when the codecs are registered to the VCS-C and I try to connect to the bridge, the VCS knows there's a zone for it internally and will always forward the calls (traversal) to the internal conference node in the head office instead of telling the codec to actually do its dns resolution locally to connect to the nearest conference node. 

Is there (I'm sure there is) a way to make a conditional rule that would say something like: 

- All these codecs will go connect internally to that zone

- All those codecs will not have their media handled through the VCS and will instead use local (codec) DNS resolution to find the closest conference node

Thanks in advance for your help!

[Patrick Sparkman]

If I'm understanding you correct, your VCS-C (HQ) and -E (APAC) each have their own endpoints and conference bridge, and you want each to only be able to connect to their own local resources, be it on the VCS-C or -E?
You could accomplish this using search rules, create a search rule on your VCS-C/E for the local conference bridge that has a higher priority than your traversal zone.  If you have multiple conference bridges on the same VCS, you could register the endpoints to a Subzone and use a CPL script to allow only calls originating from that Subzone to connect to the conference bridge. There is a CPL References chapter in the VCS Admin Guides that has an example on this.

[ysoulieres1]

Hi Patrick!
I was afraid my description might be confusing to be honest. It's a bit different than what you're describing. I'll try to be clearer, sorry for that!

Basically, we have our DMZ at HQ where the VCS-E is located, then the internal network zone where the VCS-C and VC Bridge are both located. Both VCS allow for firewall traversal mainly.
Some endpoints (Cisco SX series) are located in offices around the world, and many are registered on the VCS-E in the DMZ. Many endpoints in our HQ building are registering directly to the VCS-C.
The thing is, international locations where we have endpoints, which are registered on the VCS-E, should take advantage of DNS geolocation to find the best conference node from the VC bridge (main infrastructure located in HQ, but Conference node are spread around the world).
At the moment, because of current configurations, both VCS-E and VCS-C know where to find the VC bridge, so they send all traffic locally at the HQ instead of sending the traffic to the closest node because the VCS does internal DNS resolution.
Does that make more sense?
Thanks!