cancel
Showing results for 
Search instead for 
Did you mean: 

SQL Backup Partial Successful

pankaj_narang1
Level 2

Hi,

 

Need help on the issue.

SQL Backup is getting partially successful, this is skipping a single database.

Master Server : Solaris

Client : Windows 2003

 

Script :

 

 

OPERATION BACKUP
DATABASE $ALL
SQLHOST "gbbrf-sqtopcat"
NBSERVER "SLNX-NBUMST-LONSRD-PROD.ESC.LON.DLH.DE"
MAXTRANSFERSIZE 6
BLOCKSIZE 7
NUMBUFS 2
VDITIMEOUTSECONDS 1800
ENDOPER TRUE

 

5 REPLIES 5

inn_kam
Level 6
Partner Accredited

well enable db logs on client 

also check the permissions

 

NetBackup for Microsoft SQL Server database agent: What permissions are required for the backup and restore of Microsoft SQL Server databases?

Article:TECH43512  |  Created: 2009-01-16  |  Updated: 2013-10-28  |  Article URL http://www.symantec.com/docs/TECH43512
 

pankaj_narang1
Level 2

What other logs are required ?

 

I think its not an issues with the permissions, as this is an intermittant issue.

Pangal
Level 5

There is a bug in the netbackup where dbbackex remains active even after backup completion. So kindly check the dbbackex process in the client and kill those process

Marianne
Level 6
Partner    VIP    Accredited Certified

I am late to this discussion but it seems that nobody even tried to have a look at the logs provided.

It seems that backup for db cognos_content is failing after 100% of db was read and sent to server.

20:01:40.500 [5932.4816] <4> CODBCaccess::LogODBCerr: INF - SQL Message <3211><[Microsoft][SQL Native Client][SQL Server]100 percent processed.>

There is not enough info in the level 0 log to say what went wrong.
All we can see is a network-related error (but we can see that all other db's are backing up fine):

20:01:41.719 [5932.1524] <16> writeToServer: ERR - send() to server on socket failed:

Please increase logging level for this client to a 3 and then run a backup for this client only.
Before you start, please rename exist dbclient log so that we have a 'clean' log for this db only.

Other logs that need to be checked on the SQL client:
SQL errorlog
SQL VDI log.

Other NBU logs that may be needed:
On master: bprd (restart NBU after folder is created)
On media server: bpbrm and bptm.

One more thing (as we can see that master is on a different network to the client):
Is there a time difference between the client and the master?
The difference should not be more than a couple of minutes.

watsons
Level 6

You didn't tell us your Netbackup version... but in your dbclient logs, I can also see the following error:

20:01:42.031 [5932.1524] <32> CDBbackcat::GetSysBackupTimes: ERR - No rows found in SQL request <SELECT replace(convert(varchar, s.backup_start_date,111),'/','') + replace(convert(varchar, s.backup_start_date,108),':',''), replace(convert(varchar, s.backup_finish_date,111),'/','') + replace(convert(varchar, s.backup_finish_date,108),':','') FROM msdb.dbo.backupset s inner join msdb.dbo.backupmediafamily m ON s.media_set_id = m.media_set_id WHERE s.database_name = 'cognos_content' AND m.physical_device_name = 'VNBU0-5932-4816-1418932820' ORDER BY s.backup_finish_date>.
20:01:42.031 [5932.1524] <16> ReadFromVirtualDevice: ERR - Internal error. See the dbclient log for more information.

This is consistent with a known bug as described in: http://www.symantec.com/docs/TECH204104.