RMAN showing the data but the backup piece not available in TSM

sunny00

Active Newcomer
Joined
Apr 18, 2013
Messages
6
Reaction score
0
Points
0
Hello everyone,

I am new to TSM and RMAN. I am facing a issue; when DBA's tried to restore the database,
the backup piece was not available in the TSM, while the RMAN List Backup; showed that control file as available.

We have checked the backup logs, the backup got completed and within the retention period.
The RMAN settings are :-
RMAN> show all;


RMAN configuration parameters are:
CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 14 DAYS;
CONFIGURE BACKUP OPTIMIZATION OFF; # default
CONFIGURE DEFAULT DEVICE TYPE TO DISK; # default
CONFIGURE CONTROLFILE AUTOBACKUP ON;
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO '%F'; # default
CONFIGURE DEVICE TYPE DISK PARALLELISM 1; # default
CONFIGURE DATAFILE BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE ARCHIVELOG BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE MAXSETSIZE TO UNLIMITED; # default

As I have read that its the RMAN who can mark the backup as obsolete and then delete it after that that data is being expired from the TSM.
is their any other way the data is getting deleted or any other TSM/TDPO/RMAN settings are forcing to delete the data before the RECOVERY WINDOW OF 14 DAYS.
 
Ask your DBA to list the volumes that RMAN sees as the volumes where the data is located.

After you get the list, check if the volumes are in the library or online and available.
 
Also ask your DBA backup script. Backup name must be unique (option format).
Efim
 
I tried to collect the information but didn't received that from DBA's however I have checked all the volumes are online.
is there any way which can delete the data from the TSM without the intervention of RMAN?
Also the backup name is unique in the backup script of RMAN
 
Hi
If backup name is unique all Oracle backup objects are active object.
You can't delete active object from TSM using ANY TSM policy settings.
They will live forever :) until you use one of the following the command:
- delete filespace
- delete volume
- delete object
- RMAN commands and parameters.

Efim
 
Hi,
I have checked the delete filespace;- delete volume;- delete object in the actlog nothing has triggered to delete the backup.
Is there any way to check some RMAN commands and parameters.which could lead to this.
 
was the backup which is running with control file instead of oracle catalog?

was oracle catalog enabled/disabled recently?
 
is there any way which can delete the data from the TSM without the intervention of RMAN?
Also the backup name is unique in the backup script of RMAN

Short of deleting the file space I have found no way to address this issue without RMAN.
However, assuming you have access to the client system take a look at TDPOSYNC.
It is a TSM utility that should be located in the TDPO bin/bin64 directory.
Usage documentation can be found here: http://publib.boulder.ibm.com/infoc...topic=/com.ibm.itsmm.doc_5.4/anrcms587792.htm

It will query TSM and the Oracle catalog/repository and provide a list of inconsistencies, at which point you can choose to delete them (or not).
Once deleted run the crosscheck, delete obsolete, and delete expired processes from RMAN. The objects will then take a state of "inactive" on the TSM server and if your backup copy group is configured correctly will be deleted during the next expire inventory process.

~Rick
 
Back
Top