Step-by-Step RMAN Point-in-Time Database Restore Using Veeam Plug-in for Oracle RMAN on ExaCC

Performing an RMAN Point-in-Time Recovery (PITR) on Oracle Exadata Cloud@Customer (ExaCC) using the Veeam Plug-in for Oracle RMAN (sbt_tape) requires careful orchestration—especially when working with an external RMAN Recovery Catalog and managing database incarnations.

Here is a complete, production-ready guide based on real-world steps and common troubleshooting scenarios.

Prerequisites & Environment Setup

Before starting the restore operation, make sure:

1. The target database instance is down or stopped via Clusterware/Server Control (srvctl stop database -db CALVIN).

2. You have environment variable access to connect to both the Target Database and the Recovery Catalog.

3. The Veeam Plug-in shared library file exists at /opt/veeam/VeeamPluginforOracleRMAN/liboracleRMANPlugin.so across cluster nodes.

4. The database instance is started from the primary ExaCC cluster. 

5. The spfile is already on the dbs directory.

Step 1: Connect to the target database and RMAN Catalog.  

 Connect to the target instance and recovery catalog using RMAN:

$ srvctl stop database -db CALVIN 

$ rman target / catalog rcat/${KEY}@RCAT_XXX

Step 2:  Explicitly Set the DBID

Set the database ID (`DBID`) explicitly, which is required when restoring control files from autobackups while the database is not mounted.

Step 3:  Restore Control File from Veeam Autobackup

Allocate an 'sbt_tape' channel using the Veeam RMAN Plug-in library and pass the Veeam source backup parameter (`srcBackup`)

Step 4:  Mount the Database

RMAN> ALTER DATABASE MOUNT;

Step 5:  Initial Restore & Recover Attempt (And Handling Incarnation Mismatch)

Run the database restore and recovery script allocating multiple Veeam channels for parallel streams.

I got the below error due to database incarnation mismatch. 

RMAN-03002: failure of recover command at 09/24/2026 18:41:51 

RMAN-06004: Oracle error from recovery catalog database: 

RMAN-20011: target database incarnation is not current in recovery catalog

Step 6: Query all recorded database incarnations from the catalog.

 RMAN> LIST INCARNATION;

Step 7: Set the database back to the active parent/current incarnation that matches the target point-in-time backup.

 RMAN> RESET DATABASE TO INCARNATION 1410755;

Step 8: Re-execute The Complete Restore & Recovery.

Re-run the restore and recovery block after setting the correct incarnation.

The restore and recovery executes successfully without any error.