Configuring failover for Integration Services

In a failover environment, your PATROL Agents can move from one Integration Service to any other existing Integration Service nodes. In such a case, the PATROL Agent gets disconnected from the first Integration Service (with which it was originally connected) and moves to some other Integration Service.

When this happens, the Knowledge Module (KM) provided by this product only detects that the PATROL Agent has got disconnected. It does not detect that after disconnecting from the first Integration Service node, the PATROL Agent has moved out to some other Integration Service node. In fact, if it detects that the PATROL Agent has remained in the disconnected state for more than the maximum allowed time (10 minutes, by default), then it generates an alarm.

The AgentConnectionStatus parameter monitors the status of all the PATROL Agents connected to or disconnected from the Integration Services and is responsible to generate alarms when the maximum allowed disconnect time threshold is breached. To avoid this alarm, you must configure failover nodes.

You can configure failover nodes by using the menu command in the PATROL Console or by using the PATROL Configuration Manager in the following way:

By default, the capability to handle failover is enabled in this product. You can disable this capability by modifying the value of the following PATROL Agent configuration variable from one to zero:

/PBP__P4PNetSelfMon/IS/0/Connectivity/failoverEnabled

For a list of all the PATROL Agent configuration variables available in this product, see PATROL-Agent-configuration-variables.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*