OPCAE server-level failover

To ensure that data continues to flow from the OPCAE server to PI Interface for OPC AE, the interface can be configured to switch to another OPCAE server under the following conditions:

• Loss of connectivity to current OPCAE server

• The OPCAE server state changes to a value other than "RUNNING"

OPCAE servers can be fully redundant (multiple servers active at the same time) or configured for failover, where one server at a time is active. Note that OPCAE servers vary widely in their approach to tracking and reporting status, so consult your OPCAE server documentation to determine what options are supported.

Configure OPCAE server failover

Configure PI Interface for OPC AE to fail over to another OPCAE server in the event it loses connection to the original OPCAE server.

1. In PI ICU, go to the OPCAE > Failover tab and enter the node and name of the other OPCAE server. This basic configuration triggers failover only when PI Interface for OPC AE loses connectivity to the OPCAE server.

2. Optionally, create a string PI point to track the active OPCAE server. Assign the PI point to an unused point source.

   In PI ICU, go to the OPCAE > Failover tab and enter the PI point name in the Current Active Server Tag field. To display the value of the point, launch PI SMT and use the Data > Current Values feature.

   When failover occurs, the value of this point changes to the name of the currently connected OPCAE server. The archive of these changes enables you to view failover history.

3. To verify that failover occurs when connectivity is lost:

   • Start both OPCAE servers, then start the interface

   • Use PI SMT or the pigetmsg utility to check the PI SDK log on the PI Data Archive node for messages that verify successful startup.

   • Stop the currently active OPCAE server and check the SDK log to confirm that the interface has switched to the other OPCAE server.

   • To switch back to the first OPCAE server, restart it, stop the second server, and check the SDK log or the value of the PI active server point, if defined, to verify that the PI OPC AE interface has switched back to the first OPCAE server.

Failover timing

When failover is triggered, PI Interface for OPC AE must quickly recognize that the current OPCAE server is no longer available and determine whether the backup OPCAE server is available. To configure failover efficiency in the PI ICU, you can adjust the following settings on the OPCAE > Failover tab:

• Number of seconds to try to connect before switching to backup server (/FT=#)

  The number of seconds to try to connect, before switching to the backup OPC AE Server

• Number of seconds to wait for RUNNING State before switching to backup server (/SW=#)

  The number of seconds to wait for RUNNING status, before switching to the backup OPC AE Server