Video Screencast Help
Search Video Help Close Back
to help

Cross restore of SQL database doesn't work

Created: 07 Aug 2012 | 10 comments
ziro's picture
0 0 Votes
Login to vote

Hello,

 

We are trying to restore whole SQL database from one server to another (cross restore)

Backup master server is running on NetBackup 7.1.0.3 , server where SQL is running and backup is made on run on 6.5.5, server where should be restored run on 7.1.0.1. I think different versions of Netbackup is not reason of issue.

Clients run on Windows 2008, backup server is Solaris 10.

Files named as servers in /usr/openv/netbackup/db/altnames are created.

Connection should be ok, filesystem backups run properly, backups of sql on first client as well.

Here is script which we are using:

OPERATION RESTORE

OBJECTTYPE DATABASE

##DATABASE $ALL

DATABASE "TTPB"

SQLHOST "TTPBP-A0001"

SQLINSTANCE "TTPB_D"

NBSERVER "BKNBU"

STRIPES 2

MAXTRANSFERSIZE 6

BLOCKSIZE 7

NBSCHED "back_3m"

BROWSECLIENT "ttpbr-d0007bk"

RESTOREOPTION REPLACE

RECOVEREDSTATE NOTRECOVERED

NUMBUFS 2

ENDOPER TRUE

 

Error event log:

Event log:

Faulting application name: dbbackex.exe, version: 6.5.2009.1105, time stamp: 0x4af3859e

Faulting module name: NBCTRLU.dll, version: 6.5.2009.1105, time stamp: 0x4af37a65

Exception code: 0xc0000005

Fault offset: 0x0000000000049d59

Faulting process id: 0x41d0

Faulting application start time: 0x01cd73d208ec2aba

Faulting application path: C:\Program Files\Veritas\NetBackup\bin\dbbackex.exe

Faulting module path: C:\Program Files\Veritas\NetBackup\bin\NBCTRLU.dll

Report Id: 46bc6a8a-dfc5-11e1-94e0-cd6f5b9d5faf

 

Do you have some idea what can cause it?

Thanks.

 

Comments 10 CommentsJump to latest comment

captain jack sparrow's picture

pls post dbclient log from SQL client. If log directory is not created , create it and perform the restore again. Make sure you increase verbosity of logging from backup archive and restore (general 2, verbose 5)

 Cheers !!!

 

EV Outlook Client Add-ins

0
Login to vote
  • Actions
Yogesh9881's picture

below URL will help you....

https://www-secure.symantec.com/connect/articles/how-backup-and-restore-microsoft-sql-server-backup-netbackup

http://www.symantec.com/business/support/index?page=content&id=TECH35344

 

If this post has helped you, please vote or mark as solution.

Before break-up, make sure you have a good backup.....  ;-)

0
Login to vote
  • Actions
rookie11's picture

ur trying to do redirect SQL restore.

1 ON master server

- Create this directory:  <install_path>\NetBackup\db\altnames 

- Inside the altnames directory, create a zero byte "touch file" called: No.Restrictions 
2 u have to login to destination system then open NBU SQL GUI -- set SQL connection proerties, n set destination server name and instance name. 
3 file -restore -SQL host n source client anem should be same.
 
create the script , edit it n put new DB location in it. save it.
 
 
0
Login to vote
  • Actions
Marianne van den Berg's picture

First of all, test SQL backup on destination client.

Comms is different when database backup is done - SQL client initiates connection to master, master then needs to resolve client's incoming IP address to valid hostname. 
This does not happen with filesystem backup.

During restore, client initiates connection to master. Master needs to resolve client's IP address to valid hostname and check/verify permissions for this client name to browse/restore source client's data.

Please follow steps in NBU for Microsoft SQL Server Admin Guide  .

 

Look for topic called: Redirecting a restore to a different host.

Step 1 is done on master server. Creating No.Restrictions file is the easiest.

Step 2 onwards is done on Destination Client.

Also read this section in the same manual: Minimizing timeout failures on large database restores.

Be sure to create log folders before you start. This is the only way to troubleshoot failures:

On master: bprd (NBU needs to be restarted to enable bprd logging)

On media server: bpbrm and bptm (no restart needed)

On destination client: (bpcd and dbclient)

Please rename logs to reflect process name (e.g. bprd.txt) and post as File attachments.

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

0
Login to vote
  • Actions
ziro's picture

Hello,

Thanks you for hints.

 

Finally I see restore job on master server but it failed with RC 5. Here is status:

 

10.8.2012 10:49:23 - begin Restore
10.8.2012 10:49:24 - Info dbclient (pid=9184) INF - RESTORE STARTED USING
10.8.2012 10:49:24 - Info dbclient (pid=9184) Microsoft SQL Server 2008 R2 (RTM) - 10.50.1600.1 (X64)
10.8.2012 10:49:24 - Info dbclient (pid=9184) Apr  2 2010 15:48:46
10.8.2012 10:49:24 - Info dbclient (pid=9184) Copyright (c) Microsoft Corporation
10.8.2012 10:49:24 - Info dbclient (pid=9184) Enterprise Edition (64-bit) on Windows NT 6.1 <X64> (Build 7600: )
10.8.2012 10:49:24 - Info dbclient (pid=9184) Batch = C:\Program Files\Veritas\NetBackup\DbExt\MsSql\redirect_restore_ttpb_full.bch, Op# = 1.
10.8.2012 10:49:24 - Info dbclient (pid=9184) INF - Restoring from <2> images. The first is <TTPBP-A0001.MSSQL7.TTPBP-A0001\TTPB_D.db.TTPB.~.7.001of002.20120808075135..C>.
10.8.2012 10:49:24 - Info dbclient (pid=9184) INF - Number of stripes: 2, Number of buffers per stripe 2.
10.8.2012 10:49:24 - Info dbclient (pid=9184) INF - Created VDI object for SQL Server instance <TTPB_D>. Connection timeout is <300> seconds.
10.8.2012 10:54:26 - Info dbclient (pid=9184) ERR - Error in GetConfiguration: 0x80770003.
10.8.2012 10:54:26 - Info dbclient (pid=9184)     CONTINUATION: - The api was waiting and the timeout interval had elapsed.
10.8.2012 10:54:27 - Info dbclient (pid=9184) DBMS MSG - ODBC return code <-1>, SQL State <37000>, SQL Message <3201><[Microsoft][ODBC SQL Server Driver][SQL Server]Cannot open backup device 'VNBU0-9184-12172-1344588564'. Operating system error 0x80070002(The system cannot find the file specified.).>.
10.8.2012 10:54:27 - Info dbclient (pid=9184) DBMS MSG - SQL Message <3013><[Microsoft][ODBC SQL Server Driver][SQL Server]RESTORE DATABASE is terminating abnormally.>
10.8.2012 10:54:34 - Info dbclient (pid=9184) ERR - Error in VDS->Close: 0x80770004.
10.8.2012 10:54:34 - Info dbclient (pid=9184)     CONTINUATION: - An abort request is preventing anything except termination actions.
10.8.2012 10:54:34 - Info dbclient (pid=9184) INF - OPERATION #1 of batch C:\Program Files\Veritas\NetBackup\DbExt\MsSql\redirect_restore_ttpb_full.bch FAILED with STATUS 1 (0 is normal). Elapsed time = 311(311) seconds.
10.8.2012 10:54:36 - Info dbclient (pid=9184) INF - Results of executing <C:\Program Files\Veritas\NetBackup\DbExt\MsSql\redirect_restore_ttpb_full.bch>:
10.8.2012 10:54:36 - Info dbclient (pid=9184) <0> operations succeeded. <1> operations failed.
 

 

Do you have some idea what is wrong now? Thanks in advance.

0
Login to vote
  • Actions
Marianne van den Berg's picture

Have you read through all the documentation in my previous post?

You need to increase timeouts on media server dramatically for restores!

The default of 5 minutes (300 seconds) is not nearly enough.

Please also see the following troubleshooting logs (also in my previous post):

 

Be sure to create log folders before you start. This is the only way to troubleshoot failures:

On master: bprd (NBU needs to be restarted to enable bprd logging)

On media server: bpbrm and bptm (no restart needed)

On destination client: (bpcd and dbclient)

Please rename logs to reflect process name (e.g. bprd.txt) and post as File attachments.

 

**** EDIT *****

Collection of TN's for SQL backup and restore issues: http://www.symantec.com/docs/TECH74475

(Ignore the reference to NBU 6.0 - All info still valid)

 

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

0
Login to vote
  • Actions
ziro's picture

Hi,

I tried to find something in dbclient, bpcd logs and in mssql_backup_failures log.(see attached files) 

In mssql_backup_failures is:

STATUS -1
INFO Error in GetConfiguration: 0x80770003.
INFO The api was waiting and the timeout interval had elapsed.

I found some topic regarding this:

http://www.symantec.com/connect/forums/failed-sql-...

Connection checked through commands

bpclntcmd -hn <hostname>
bpclntcmd -ip <ip-address>

seems to be ok, right names and IP addresses, reachable...

 

Is something in logs what I overlooked?

AttachmentSize
081012bpcdlog.txt 117.59 KB
081012rpt.txt 27.95 KB
081012dbclientlog.txt 638.73 KB
0
Login to vote
  • Actions
Marianne van den Berg's picture

I don't see any of the following logs:

 

master's bprd log

media server's bpbrm and bptm logs.

We need to first of all see how master interpreted client's restore request. This will tell us why we see status 37's in dbclient log.

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

0
Login to vote
  • Actions
ziro's picture

Really sorry, but these logs are huge 50 - 100 MB. I cant upload it here.

0
Login to vote
  • Actions