Within an IM and Presence redundancy group, the Cisco Server Recovery Manager (SRM) service continually polls the peer nodes. If a node or a critical service on that node becomes unreachable, SRM initiates the appropriate failover actions (for example, activating the user’s backup home node) and advertises the new state to all cluster members. Because SRM alone performs both detection and orchestration of failover/state-change activities, it is the service that “handles failover and state changes in the cluster.”

A. XCP Sync Agent – Replicates user/buddy-list data between home and backup nodes; does not detect or trigger failover.
C. Cisco XCP Connection Manager – Manages client (e.g., Jabber) sessions; no cluster failover logic.
D. XCP Router – Routes XCP messages between services/nodes; relies on SRM for failover decisions.

1. Cisco IM and Presence Serviceability Administration Guide

Release 14 (Cisco

2021)

Table 2-1 “Critical IM and Presence Services”

p. 29: “Cisco Server Recovery Manager – Provides failover recovery and state change management across the IM and Presence cluster.”

2. Cisco Unified Communications SRND for IM & Presence

Release 12.x

Ch. 9 “High Availability”

p. 9-6: “SRM monitors all nodes in a redundancy group and initiates user failover when a node or service is lost.”

3. Cisco IM and Presence Redundancy Design Guide

Rev 11.5

§3.2 “Server Recovery Manager Operation”

pp. 17-18.

(Only official Cisco documentation—no commercial prep material—has been cited.)