Cisco Cisco IP Contact Center Release 4.6.1 Leaflet

Unified CCE can also be overlaid with the Unified CM design model for clustering over the WAN, which 
allows for high availability of Unified CM resources across multiple locations and data center locations. 
There are a number of specific design requirements for Unified CM to support this deployment model, 
and Unified CCE adds its own specific requirements and new failover considerations to the model.

Specific testing has been performed to identify the design requirements and failover scenarios. The 
success of this design model relies on specific network configuration and setup, and the network must 
be monitored and maintained. The component failure scenarios noted previously (see 

) are still valid in this model, and the additional failure scenarios for this 

model include:

The terms public network and visible network are used interchangeably throughout this document.

In clustering over the WAN with Unified CCE, there should be a separate private network connection 
between the geographically distributed Central Controller (Call Router/Logger) and the split Peripheral 
Gateway pair to maintain state and synchronization between the sides of the system.

To understand this scenario fully, a brief review of the ICM Fault Tolerant architecture is warranted. On 
each call router, there is a process known as the Message Delivery Service (MDS), which delivers 
messages to and from local processes such as router.exe and which handles synchronization of messages 
to both call routers. For example, if a route request comes from the carrier or any routing client to side A, 
MDS ensures that both call routers receive the request. MDS also handles the duplicate output messages.

The MDS process ensures that duplex ICM sides are functioning in a synchronized execution, fault 
tolerance method. Both routers are executing everything in lockstep, based on input the router receives 
from MDS. Because of this synchronized execution method, the MDS processes must always be in 
communication with each other over the private network. They use TCP keep-alive messages generated 
every 100 ms to ensure the health of the redundant mate or the other side. Missing five consecutive TCP 
keep-alive messages indicates to Unified ICM that the link or the remote partner system might have 
failed.

When running duplexed ICM sides as recommended for all production system, one MDS will be the 
enabled synchronizer and will be in a paired-enabled state. Its partner will be the disabled synchronizer 
and is said to be paired-disabled. Whenever the sides are running synchronized, the side A MDS will be 
the enabled synchronizer in paired-enabled state. Its partner, side B, will be the disabled synchronizer 
and paired-disabled state. The enabled synchronizer sets the ordering of input messages to the router and 
also maintains the master clock for the ICM system.

If the private network fails between the Unified ICM Central Controllers, the following conditions 
apply:

The Call Routers detects the failure by missing five consecutive TCP keep-alive messages. The 
currently enabled side (side A in most cases) transitions to an isolated-enabled state and continues 
to function as long as it is in communication with at least half of the PGs configured in the system.