Oracle redirected restore failing

[Ramesh111]

Hi,

   We are doing Orcale redirected restore but it's getting failed with below error.

Master Server = Windows 2k8

Client1 = Solaris 10   

Client2 = Solaris 10

Backup is taken via Oracle inteligent policy for the client1 and trying to restore in Client2.

Error :

23:35:37.567 [12817] <4> logCommFile: <22:46:04 (461365.001) Skipping to next file header...>

23:35:37.567 [12817] <4> logCommFile: <22:46:04 (461365.001) /u007/oraprod12/proddata/system/system11.dbf>

23:35:37.567 [12817] <4> logCommFile: <22:46:04 (461365.001) Read access to /u007/oraprod12/proddata/system/system11.dbf was denied>

23:35:37.567 [12817] <4> logCommFile: <22:46:34 (461365.001) Skipping to next file header...>

23:35:37.567 [12817] <4> logCommFile: <22:46:34 (461365.001) /u007/oraprod12/proddata/data1/a_summ01.dbf>

23:35:37.567 [12817] <4> logCommFile: <22:46:34 (461365.001) Read access to /u007/oraprod12/proddata/data1/a_summ01.dbf was denied>

23:35:37.567 [12817] <4> logCommFile: <22:47:01 (461365.001) Skipping to next file header...>

23:35:37.567 [12817] <4> logCommFile: <22:47:01 (461365.001) /u007/oraprod12/proddata/system/system06.dbf>

23:35:37.567 [12817] <4> logCommFile: <22:47:01 (461365.001) Read access to /u007/oraprod12/proddata/system/system06.dbf was denied>

23:35:37.567 [12817] <4> logCommFile: <22:47:23 (461365.001) Skipping to next file header...>

23:35:37.567 [12817] <4> logCommFile: <22:47:23 (461365.001) /u007/oraprod12/proddata/system/system07.dbf>

23:35:37.567 [12817] <4> logCommFile: <22:47:23 (461365.001) Read access to /u007/oraprod12/proddata/system/system07.dbf was denied>

23:35:37.567 [12817] <4> logCommFile: <22:47:46 (461365.001) Skipping to next file header...>

From Netbackup console:

17/11/2015 10:44:20 PM - Info bptm(pid=10180) EXITING with status 0 <----------        
17/11/2015 10:44:20 PM - Info master server (pid=10180) StorageServer=PureDisk:media server ; Report=PDDO Stats for (mediaserver): read: 163475745 KB, CR received: 48944985 KB, CR received over FC: 0 KB, dedup: 0.0%
17/11/2015 10:44:20 PM - Info tar32(pid=12837) done. status: 183: tar received an invalid archive     
17/11/2015 10:44:20 PM - Error bpbrm(pid=5064) client restore EXIT STATUS 183: tar received an invalid archiv

We did the client name change before restore, but restore is still failing.

[Marianne]

What is username and userid of the Oracle user that did the backup on Client1? What is username and userid doing the restore on Client2? Especially userid must be the same.

[Ramesh111]

Dear marianne,

                      Thanks for your update. We used same userid for Client1 and Client2 for backup and also for restore. When we are doing restore of Client2 backup to Client2 it's working fine. 

[Marianne]

You have a permission problem. This is OS-related. Please compare oracle user entries in /etc/password file. Can you list Oracle backups as Oracle user on Client2 for Client1 using bplist?

[Michal_Mikulik1]

Hello,

verify that you redirected restore is allowed via %altnames% on Master Server.

http://www.veritas.com/docs/000073334     

Michal

[Nicolai]

Agree with Marianne - Absolutely look like file system permissions.

The GID - group permission is likely wrong for the DBA  group.

You need to restore the database as root using rman on the destination system

Alternative:

• fix group dba access rights on source system
• Stop/start oracle on source system in order to re-read the new file permissions
• Run new backup
• New re-directed restore using new backup.

[Ramesh111]

Sorry for late reply.. I was busy with symantec engineer. 

Dear Michal,

                  we have allowed from master server. Restore is getting triggered from server. Only problem is not writing the data.

Dear Marianne,

                     You are correct.It's permission issue. Now I have updated new user details and taking the fresh backup..

I will update the restore result as soon as restore done. 

[Ramesh111 · Accepted solution]

I really want to thank you all for your support, 

I worked with Symantec engineer and resolved the issue. After adding below entry in bp.conf  taken new backup and restored to another server. Finally Restore got success. If anyne is taking Oracle DB backup and the DB directory is on another mount point, please test your restore it's really working or not. 

Symantec engineer update:

Some times when proxy backup is initiated and the DB directory is on another mount point than netbackup is unable to backup that directory files. however we can see the backup  is completed.

this is a known issue and it is under working of engineering.

A workaround for this is to add parameter in the bp.conf file.
So we added the parameter "ORACLE_OVERRIDE_DATA_MOVEMENT = 1" in the bp.conf.

[Marianne]

Thanks for sharing the solution.