Windows SAN Client backup failed for Oracle

You might find some addinational clues in the bphdb log and it might be worth to try run with the NOSHM touch file which should make the SAN backup act like the LAN backup.

The oracle alert log and the server event log might also give some indication of what the problem is

Can you run a file backup with FT transport ?

Regards

Michael

Wilson,

plz test filesystem backup via FT and let us know result ?? is that OK or failed ??

Yes, 7.6.0.2 began support for 12c and would be required here. That will also get you into supported setup with Symantec in case it doesnt resolve the issue.

The master server, media server (5230 appliance) and the client all upgraded to 7.6.0.2. But same result for RMAN backup via SAN Client.

In the dbclient log file, there is some error below:

14:52:47.184 [4684.5948] <16> AttachShMForBackup: Unable to open shared memory (Global\NetBackup Media Manager SHM Path 18633156 2280): OS error 5
14:52:47.184 [4684.5948] <16> SetupShMforBackup: Returned with error 12 from AttachShMForBackup
14:52:47.184 [4684.5948] <16> handshake: ERR - failed setting up for shared mem
14:52:47.184 [4684.5948] <16> CreateNewImage: ERR - handshake() failed
14:52:47.309 [4684.5948] <16> VxBSACreateObject: ERR - Could not create new image with file /bk_u1gp8p0tk_s48_p1_t848069556.
14:52:47.309 [4684.5948] <16> xbsa_CreateObject: ERR - VxBSACreateObject: Failed with error:
   Server Status:  Communication with the server has not been initiated or the server status has not been retrieved from the serve
14:52:48.200 [4684.3688] <16> AttachShMForBackup: Unable to open shared memory (Global\NetBackup Media Manager SHM Path 18634156 2280): OS error 5
14:52:48.200 [4684.3688] <16> SetupShMforBackup: Returned with error 12 from AttachShMForBackup
14:52:48.200 [4684.3688] <16> handshake: ERR - failed setting up for shared mem
14:52:48.200 [4684.3688] <16> CreateNewImage: ERR - handshake() failed
14:52:48.325 [4684.3688] <16> VxBSACreateObject: ERR - Could not create new image with file /bk_u1fp8p0tk_s47_p1_t848069556.
14:52:48.325 [4684.3688] <16> xbsa_CreateObject: ERR - VxBSACreateObject: Failed with error:
   Server Status:  Communication with the server has not been initiated or the server status has not been retrieved from the serve

The NetBackup Client service and the NetBackup SAN Client service are both using the local system account for the logon account to start the services.

After we set the logon account same as the Oracle services, now the backup can go thru FT transport. But it only can run successfully for one job only. The subsequent job will still failed with error 6. 

Once the RMAN backup job is failed, I straight trigger normal file backup via FT transport, it is still working fine.

Now the error in the dbclient is different:

17:46:51.532 [6336.10000] <4> handshake: INF - valid newdataSock: 5020
17:46:51.532 [6336.10000] <4> handshake: INF - nameSock = 3020, dataSock = 5020, commSock = 2516
17:46:51.532 [6336.9420] <16> CopyToShM: connection broken. Error #40
17:46:51.532 [6336.9420] <16> writeToServer: ERR - CopyToShM() failed, no files backed up
17:46:51.532 [6336.9420] <16> dbc_put: ERR - failed sending data to server
17:46:51.532 [6336.9420] <4> closeApi: entering closeApi.
17:46:51.532 [6336.9420] <4> closeApi: INF - EXIT STATUS 6: the backup failed to back up the requested files

17:46:51.532 [6336.9420] <4> closeApi: INF - closing commSock 4664
17:46:51.532 [6336.9420] <4> closeApi: INF - closing dataSock 4808
17:46:51.532 [6336.9420] <4> closeApi: INF - setting linger on nameSock 776
17:46:51.532 [6336.9420] <4> closeApi: INF - closing nameSock 776
17:46:51.532 [6336.9420] <16> VxBSASendData: ERR - Could not do a bsa_put().
17:46:51.532 [6336.9420] <2> xbsa_ProcessError: INF - entering
17:46:51.532 [6336.9420] <2> xbsa_ProcessError: INF - leaving
17:46:51.532 [6336.9420] <16> xbsa_SendData: ERR - VxBSASendData: Failed with error:
   Server Status:  Communication with the server has not been initiated or the server status has not been retrieved from the serve
17:46:51.532 [6336.9420] <2> sbterror: INF - entering
17:46:51.532 [6336.9420] <2> sbterror: INF - Error=7501: VxBSASendData: Failed with error:
   Server Status:  Communication with the server has not been initiated or the server status has not been retrieved from the serve.

It seems like there is some connection drop, but still can't manage to find where is it.

Problem:

NetBackup SAN Client backup of Microsoft SQL Server fails with status code 6

15:21:38.819 [8108.7324] <16> AttachShMForBackup: Unable to open shared memory (Global\NetBackup Media Manager SHM Path 84222078 2900): OS error 5

15:21:38.819 [8108.7324] <16> SetupShMforBackup: Returned with error 12 from AttachShMForBackup

15:21:38.819 [8108.7324] <16> handshake: ERR - failed setting up for shared mem

Error 5 is ACCESS_DENIED

 

Since the shared memory is created by the NetBackup San Client Service and written to by the database agent, the database agent must have permission to access the shared memory.

 

Resolution:

The database agent runs under the NetBackup Client service. This service must be started with an account that will have privileges to access shared memory created by the account that the NetBackup SAN Client service runs under. This can be controlled via the services section of the Computer management interface.

Simply right click on "My Computer", then select manage. In the Computer Management interface. select "Services and Applications" then Services. In the Right panel find the "NetBackup Client Service". right click and select properties. In properties select the "Log On" tab. In this tab the account used by this service can be designated. Be sure that the designated account has the privileges needed in order to access shared memory created by the account the NetBackup San Client service logs on as.  

Mentions SQL instead of Oracle though it looks like the same trouble! 

http://www.symantec.com/business/support/index?page=content&id=TECH74570

After we set the same logon account with the Oracle service, the backup via FT transport is working, but only for the first stream job. Subsequent job still failed.

The normal file backup using FT transport is working fine. The RMAN backup using lan transport also working fine. Only RMAN with FT transport is not working.

Master: hqmsbk04

Media: hqststr10

Client: hqmsdb42

Attached the log from respective server:

Master: bprd, bpdbm

Media: bptm, bpbrm

Client: dbclient

Some how look likes is a communication problem, but bpclntcmd and bptestbpcd are all working correctly, so still no idea where is the problem.

Hope someone can give some info on this.

Thanks.