backup and restore issues

[John_Constantin]

Dear sysmantec team:

What is the difference from the process of database backup & restore? as i know, for database backup operation on media server,the bptm child process receives the image and stores it block by block into a shared memory segment on the media server,then the parent bptm process retrieves the image from shared memory and directs it block by block to the allocated storage media.for database restore operation on media server,the parent bptm process reads data from the media or disk and delivers it to shared memory block by block,then the child bptm process delivers the data to the tar process on the client system, which writes the data to the client disk.

So i make a script to test the above that mentioned.

The backup script is as follow:

---

OPERATION BACKUP
DATABASE "SAPDATA"
SQLHOST "SIH-SQLDB"
NBSERVER "SIH-NETBACKUP"
STRIPES 3
MAXTRANSFERSIZE 6
BLOCKSIZE 7
POLICY SAPDATA_FULL
BROWSECLIENT "SIH-SQLDB"
NUMBUFS 2
ENDOPER TRUE

---

Result:Successful.

Then i change the script for restore as follow:

---

OPERATION RESTORE
OBJECTTYPE DATABASE
DATABASE "SAPDATA"
# The following image is type: Full
NBIMAGE "SIH-SQLSRV1.MSSQL7.SIH-SQLDB.db.SAPDATA.~.7.001of003.20140329221029..C"
SQLHOST "SIH-SQLDB"
NBSERVER "SIH-NETBACKUP"
STRIPES 003
BROWSECLIENT "SIH-SQLDB"
MAXTRANSFERSIZE 6
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE RECOVERED
NUMBUFS 2
ENDOPER TRUE

---

Result:Failed

Then i change the restore script to try again

---

OPERATION RESTORE
OBJECTTYPE DATABASE
DATABASE "SAPDATA"
# The following image is type: Full
NBIMAGE "SIH-SQLSRV1.MSSQL7.SIH-SQLDB.db.SAPDATA.~.7.001of003.20140329221029..C"
SQLHOST "SIH-SQLDB"
NBSERVER "SIH-NETBACKUP"
BROWSECLIENT "SIH-SQLDB"
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE RECOVERED
NUMBUFS 2
ENDOPER TRUE

---

Result:Successful

Now,why backup operation is successful and restore operation is failed while using stripes 003 and maxtransfersize 6?

[Deb_Wilmot]

If you set STRIPES to 3 versus 003 does it work?

Just wondering since the STRIPES setting is an integer with valid values between 1 and 32.

Deb

[Marianne]

We cannot tell you why the different parameters resulted in the failed restore. You will need logs to troubleshoot restore failure.

On media server: bptm and bpbrm

On SQL client: dbclient

It is also possible that some parameters are meant for backups only - not restores.

[John_Constantin]

Dear Marianne:

please find the attachments,the restore operation start time is 9:05AM,this timestamp can help you to locate the error info of restore operation which is failed.

[John_Constantin]

here is dbclient log on client system

[Marianne]

According to NBU for SQL manual, MAXTRANSFERSIZE is supported for backup and restore.

I do see this error in the dbclient log:

09:24:36.895 [1688.6816] <16> CDBbackrec::ProcessVxBSAerror: ERR - Error in DBthreads::dbclient: 6.

Have a look at this TN and see if this is a similar issue: http://www.symantec.com/docs/TECH212281

[John_Constantin]

i don't think the TN is useful for me, because i am sure what the difference between my two scripts are just only to set stripes to 003 and to set maxtransfersize to 6.what i can find in the TN is that you must add Move and TO to ndf data file,and my script includes them as follow,you can see detail in my attachment

MOVE  "WSS_Content"
TO  "S:\bak_temp\WSS_Content.mdf"
#
#
#  Replace the file path <M:\SQL DB\WSS_Content2.ndf>
#  with a new file path. Also remove the hash mark <#> which precedes the keyword <TO>.
#  The target of the MOVE keyword must be "WSS_Content2".
MOVE  "WSS_Content2"
TO  "S:\bak_temp\WSS_Content2.ndf"
#
#
#  Replace the file path <S:\SQL LOG\WSS_Content_log.LDF>
#  with a new file path. Also remove the hash mark <#> which precedes the keyword <TO>.
#  The target of the MOVE keyword must be "WSS_Content_log".
MOVE  "WSS_Content_log"
TO  "S:\bak_temp\WSS_Content_log.LDF"
#
# The following image is type: Full