Oracle RMAN duplicate ignoring NSR_CLIENT

Hi all,

I am trying to do a remote restore /  RMAN duplicate using NMDA.

We have an Oracle 11.2.0.3 database "PRODDB" running on an Oracle Linux 5 x86_64 server called "proddbserver".

The fqdn for this server is proddbserver.domain.local.

On this server we have a working backup/restore using NMDA going.

The following packages are installed:

lgtoman-8.1.1.1-1

lgtoclnt-8.1.1.1-1

lgtonmda-1.6.0.2-1

The backups are made to networker server "nwserver.networker" with automatic channel allocation from RMAN and the following NSR options are used:

NSR_SERVER=nwserver.networker

NSR_CLIENT=proddbserver.abc.networker

Next, we have an Oracle 11.2.0.3 database called "TESTDB" also running an Oracle Linux 5 x86_64 server called "testdbserver".

The fqdn for this server is testdbserver.domain.local.

Again the following packages are installed:

lgtoman-8.1.1.1-1

lgtoclnt-8.1.1.1-1

lgtonmda-1.6.0.2-1

The testdbserver has been granted remote access to the proddbserver.

On the testdbserver I can successfully query the networker server for volume info using nsrorainfo:

For example last night a controlfile and spfile autobackup for the PRODDB was created with file name "c-3942296553-20140812-01".

oracle@testdbserver ~]$ nsrorainfo -s nwserver.networker -c proddbserver.abc.networker c-3942296553-20140812-01

This successfully lists the volume networker intends to use, so the privileges seem to be correct.

Now when I run the RMAN duplicate statement on the testdbserver I have to use the NSR_CLIENT setting which was used to create the backup, so in my RMAN run block I have to use NSR_SERVER=proddbserver.abc.networker in order to be able to perform the duplicate.

Defining the required NSR_CLIENT and NSR_SERVER variables for an Oracle restore

For some reason however RMAN / Networker does not use the NSR_CLIENT which I have included in the run block and ignores it.

When I start the duplicate I get this message:

RMAN> connect target sys@PRODDB

target database Password:

connected to target database: PRODDB (DBID=123456789)

RMAN> connect auxiliary /

connected to auxiliary database: TESTDB (not mounted)

RMAN>

run

{

allocate auxiliary channel ch1 DEVICE TYPE 'SBT_TAPE' PARMS 'ENV=(NSR_SERVER=nwserver.networker,NSR_CLIENT=proddbserver.abc.networker)';

duplicate target database to 'TESTDB' 

UNTIL TIME "to_date('08/11/14 16:00:00','mm/dd/yy hh24:mi:ss')";

}

Starting restore at 12-AUG-14

channel ch1: starting datafile backup set restore

channel ch1: restoring control file

channel ch1: reading from backup piece c-3942296553-20140812-01

channel ch1: ORA-19870: error while restoring backup piece c-3942296553-20140812-01

ORA-19507: failed to retrieve sequential file, handle="c-3942296553-20140812-01", parms=""

ORA-27029: skgfrtrv: sbtrestore returned error

ORA-19511: Error received from media manager layer, error text:

Could not locate the backup piece 'c-3942296553-20140812-01' on the NetWorker server 'nwserver.networker' for the client 'testdbserver.domain.local'. Notify the backup administator. (2:9:2

failover to previous backup

etc...

This then continues to look for an older backup piece which it also cannot find until it has tried all relevant backup pieces.

It makes sense that it cannot locate the backup piece for client 'testdbserver.domain.local', but it should have been looking for client 'proddbserver.abc.networker' instead as that was what I have configured in the allocate channel.

I have tried several things but can't get it to work.

I've tried exporting NSR_CLIENT on OS.

I've created a file /nsr/apps/res/nmda_oracle_defaults:

oracle@testdbserver ~]$ cat /nsr/apps/res/nmda_oracle_defaults

NSR_CLIENT=proddbserver.abc.networker

NSR_SERVER=nwserver.networker

But it keeps ignoring the NSR_CLIENT setting and using the 'localhost' name instead.

Do any of you know what could be wrong here and why the NSR_CLIENT value I have used is being ignored?

Thanks!

Lauris