Answer

When performing the "DUPLICATE" command thru "RMAN" to duplicate a database to an alternate machine, the script will initially run and access the source machine. However, the duplicate process will actually run on the auxiliary (alternate) machine. Therefore, the Data Protection for Oracle must be configured on this alternate machine, not just the machine where the backup was performed.

If the alternate machine has not been "setup/configured" for the Data Protection Oracle client, then a failure will be seen when the restore attempts to run on this target/auxiliary machine.

Additionally, the source Tivoli Storage Manager nodename would need to be specified in the "TDPO_OPTFILE" on this alternate machine. For example if the "RMAN" script for the duplicate is:

connect target sys/change_on_install@prod1
connect auxiliary sys/sysdba@dupdb
run
{
allocate auxiliary channel t1 type 'sbt_tape' parms
'ENV=(TDPO_OPTFILE=/usr/tivoli/tsm/client/oracle/bin/tdpo.opt)';
set newname for datafile 1 TO '$ORACLE_HOME/dbs/tbs_01.f';
set newname for datafile 2 TO '$ORACLE_HOME/dbs/tbs_02.f';
set newname for datafile 3 TO '$ORACLE_HOME/dbs/tbs_03.f';
duplicate target database to dupdb logfile
group 1 ('$ORACLE_HOME/dbs/log_1_1.f', '$ORACLE_HOME/dbs/log_1_2.f') size 200K reuse;
}

In the case where a "duplicate" of a database is being made, the script is launched on the Backup "prod1" machine, but actually runs on the Target "dupdb" machine. Thus the "TDPO_OPTFILE" "/usr/tivoli/tsm/client/oracle/bin/tdpo.opt" that is specified in the "ENV parms" is going to be looked for on the Target "dupdb" machine. The "TDPO_NODE" in this "tdpo.opt" file must be the Tivoli Storage Manager node that performed the backup.