Automated client failover

Edit online

If you backed up data to the IBM Storage Protect server, Data Protection for VMware can automatically fail over to the specified failover server for data recovery when there is an outage on the IBM Storage Protect server.

The server that the IBM Storage Protect data mover node connects to during normal production processes is called the primary server. When the primary server and data mover node are set up for node replication, the client data on the primary server can be replicated to another IBM Storage Protect server, which is the target replication server. This server also functions as the failover server.

When the data mover node logs on during normal operations, connection information for the failover server is automatically sent to the data mover node from the primary server. The failover server information is automatically saved to the client options file on the data mover node. No manual intervention is required by you to add the information for the failover server.

Each time the data mover node logs on to the server, the data mover node attempts to contact the primary server. If the primary server is unavailable, the data mover node automatically fails over to the failover server, according to the failover server information in the client options file. In failover mode, you can restore any replicated client data. When the primary server is online again, the data mover node automatically fails back to the primary server the next time the data mover node connects to a server.

There can be multiple target replication servers. If the client is not able to connect to the source replication server, the client will try to connect to all the target replication servers that have been defined, depending on availability.

Requirements: Before the connection information for the failover server is sent to the client options file, the following requirements must be met:
  • The primary server, failover server, and data mover node must be at version 7 (any level) or version 8 (any level).
  • The primary and failover servers must be set up for node replication, and all the client nodes must be configured for node replication on the server.
  • You must back up VMware data at least once to the primary server.
  • Client data on the primary server must be replicated to the failover server at least once.
Restriction: The following restrictions apply to Data Protection for VMware during failover operation:
  • Any operations that require data to be stored on the server, such as backup operations, are not available.
  • Schedules are not replicated to the failover server. Therefore, schedules are not run while the primary server is unavailable.
  • Validation of virtual machine backups is not available.
  • All client failover scenarios that use the Data Protection for VMware GUI and the IBM Storage Protect file restore interface are not operational. As a workaround, you can use the dsmc command line on the data mover node to restore only data from the failover server (except for VM file level restore, which is not available with the dsmc command). You can run a full VM restore, instant restore, or instant access operation by using the restore vm command with the vmrestoretype option. For more information, see Vmrestoretype.