RCC Redundancy Configurations (Local and Inter-Site)

Route Cache Controller (RCC) configurations are required for high availability (HA) to work. This is required for RCC services running at one site (local redundancy) and running at different sites (inter-site redundancy).

The RCC service handles route cache processing through a route cache messaging interface called the Route Cache Agent (RCA). At least one RCA instance is active on each RCC service for main processing. Also, additional RCA instances may be required for local and inter-site redundancy as follows:

• Local redundancy — At a site where two RCC services (an active/standby HA pair) run locally, each RCC instance in the RCC HA pair must have an RCA [standby] entry in its RCA configuration to point to its local peer.
• Inter-site redundancy — The engine chain that implements a customer sub-domain can be distributed across sites. For example, Engine 1 and Engine 2 form the engine chain running customer sub-domain A where Engine 1 is located at site 1 and Engine 2 is located at site 2. When an engine chain that implements a customer sub-domain is distributed across sites, each RCC instance in the MATRIXX environment must have an RCA [intersite] entry in its RCA configuration to point to its inter-site peer. This enables a Route Cache inter-site connection.
  Tip: The inter-site peer is typically the RCC-service VIP provisioned for the RCC HA pair on the other site.

  The Route Cache inter-site connection is used for:

  • Sending route-cache update messages to the inter-site peer RCC every time the local RCC receives a route-cache update message from a local (same site) processing blade.

    The Route Cache at each site must be able to receive updates to its data set for the latest processing (the activity, such as sessions, being processed by the active engine) no matter if the active engine is at the local site or the remote site. In this way, when an engine failover occurs, the RCC service at the site where the active engine was shut down can still provide sub-domain look-up (to the local TRA-RT-(SI/DR)) based on activity being processed by the newly active engine at the remote (backup) site.

  • Sending domain resolution requests to all engine processing blades in the environment (to both sites) every time the local RCC receives a cache-miss resolve request from the local TRA-RT-(SI/DR) or Network Enabler.

    If a cache miss occurs, the RCC service must be able to send domain resolution requests to engine processing blades and receive domain resolution responses from engine processing blades no matter at which site the processing blades are located.

For more information about the [standby] and [intersite] RCA configuration entries, see the discussion about Route Cache Controller configuration properties.