Application failover

When the IBM Spectrum Protect server becomes unavailable because of an outage, applications that use the API can automatically fail over to a secondary server for data recovery.

The IBM Spectrum Protect server that the client and API connects to during normal production processes is called the primary server. When the primary server is set up for node replication, that server is also known as the source replication server. The client node data on the source replication server can be replicated to the target replication server. This server is also known as the secondary server, and is the server that the client automatically fails over to when the primary server fails.

The client and API must be configured for automated client failover, and must connect to a version 7.1 (or later) server that replicates client node data. The configuration for the API is the same as the configuration for the backup-archive client.

During normal operations, connection information for the secondary server is automatically sent to the client from the primary server during the logon process. The secondary server information is automatically saved to the client options file.

Each time the client application logs on to the IBM Spectrum Protect server, it attempts to contact the primary server. If the primary server is unavailable, the application automatically fails over to the secondary server by using the secondary server information in the client options file. In failover mode, the application can query the secondary server and restore or retrieve replicated data.

You must back up the application at least one time to the primary server. The API can fail over to the secondary server to recover data only if the data from the client node was replicated from the primary server to the secondary server.