Cisco Cisco Customer Voice Portal 8.0(1) Design Guide

A trunk configured with MTP required will not work with the LBCAC siteID feature. The reason is 
when MTP is inserted, the media is terminated between the end point and MTP resource, not 
between the two end points. 

If a MTP/Transcoder/TRP media resource is inserted by the UCM media layer, the incoming 
location information is not used. 

If the inter cluster call is not hair-pin/loop back to the same cluster, the former behavior of Location 
CAC logic will apply. 

Each site is uniquely identified by one siteID. Multiple gateways at the same site would need to align 
to the same siteID, but if two clusters happen to use the same location name, then two siteIDs can 
map to the same physical branch.

A second Unified CM cluster may have the same location as the first cluster, but will still be required 
to use a unique siteID on Unified CVP. You can define a route in the proxy server to send those 
cluster calls to the common VXML gateway at the same location, but used by both the clusters. 

Each cluster would manage the bandwidth for devices in its cluster. If two clusters happen to use the 
same physical location, then they would each separately manage the bandwidth for the phones that 
they manage. 

The following considerations apply when using LBCAC:

During the CAC failure, Unified CVP returns a failure code to Unified CCE that triggers router 
requery. 

If a branch doesn't have a VXML Gateway, then it is recommended to use the VXML Gateway at 
the Central data center.

The previous version of Unified CVP provided a method of configuring CAC. This method is superseded 
by the LBCAC method presented here. Both configuration methods are provided in the Configuration 
and Administration Guide for Cisco Unified Customer Voice Portal (CVP), available at: