Siebel Universal Queuing Administration Guide > Installing and Running Siebel Universal Queuing > Implementing Failure Recovery >

Heartbeat Protocol

The Siebel Universal Queuing routing engine can detect connection failures actively by implementing a heartbeat protocol with Siebel Communications Server.

In the heartbeat protocol, Siebel Communications Server sends a message with a sequence number to the routing engine on every request connection of each connection set. The message is sent at each interval as determined by the value of the Ping Interval parameter. The default value of this parameter is 30 (seconds). Upon receiving a heartbeat message, the routing engine echoes the sequence number back in the response and restarts the Ping Interval timer.

If the Ping Interval timer expires, it is automatically reset. If the Ping Interval timer expires two consecutive times, the routing engine declares the connection to have failed and initiates recovery as described in Siebel Universal Queuing Failure, below. If a heartbeat request is received with the sequence number out of order, the routing engine treats this the same as a timer expiration.

If Siebel Communications Server encounters a network error when issuing a heartbeat request, Siebel Communications Server will initiate failure recovery and reconnect as described in Siebel Communications Server Failure.

For more information about setting the Ping Interval parameter, see Table 7.