Video Screencast Help
Symantec to Separate Into Two Focused, Industry-Leading Technology Companies. Learn more.

Restoring a SqlServer Database

Created: 14 Nov 2013 • Updated: 14 Nov 2013 | 8 comments
This issue has been solved. See solution.

Hello Everybody

I am performing a SqlSever restore and it´s failing with the messages from the Activity Monitor:

Error bpbrm socket read failed: errno=62 - Timer expired
the restore failed to recover the request files (5)
 
From the Windows clint side the error showed is:
 
TIMESTAMP 20131114085544
MASTERSERVER SVBACKUP
CLIENT bancoerp2
INSTANCE bancoerp2
DATABASE RST0811
OPERATION RESTORE
OBJECTTYPE DATABASE
STATUS -1
INFO Server Status:  Communication with the server has not been initiated or the server status has not been retrieved from the server.
INFO Error in VxBSAGetData: 3.
INFO System detected error, operation aborted.
ENDOPER TRUE
 
My SqlServer is 2005, running in Windows Server 2003.
My Netbackup Server is 6.5.6 running in Solaris 10.
 
Can anyone help me?
 
Thanks
 
Alexandre Andrich
Operating Systems:

Comments 8 CommentsJump to latest comment

Mark_Solutions's picture

Where are you running this from?

It needs to be run from the client via the NetBackup SQL GUI

You also need to ensure that no one is accessing the database in question

Give us more detail of exactly what you have done so that we can assist further

Any logs would also assist

Authorised Symantec Consultant

Don't forget to "Mark as Solution" if someones advice has solved your issue - and please bring back the Thumbs Up!!.

andrich's picture

Mark

I am restoring from the Netbackup Database Extension GUI from the client.

In this GUI I have chosen

1)Restore SQL Server Objects (F3)

2)Backup History Options

3)Then I find my backup list and I have select the date I need

4) Manage Scripts Files (F4)

5)Then I save the script and I have modified it, seting a new PAth for the restore;

6) Then I select the script and fire START

All this actions are performed in clint side.

So a Restore job starts at Netbackup server. This job take the tape, start reading the tape and so it fails with error 5.

What log do you need? I heve attached the dbclient. Take a look please.

AttachmentSize
dbclient.txt 54.51 KB
Mark_Solutions's picture

Could we see your edited script file please

The log seems to indicate that it restired the database but then crashed out - possibly in trying to mount it / add it to SQL

What do you see in the directiry it was being restored to and in SQL Studio?

Had the original database been removed first?

Authorised Symantec Consultant

Don't forget to "Mark as Solution" if someones advice has solved your issue - and please bring back the Thumbs Up!!.

andrich's picture

Mark

I am restoring to a different place from the Sql folder, then my DBA will take this files and aplly in a Test Database.

When I start the script, it creates in the folder the two specified files, but the restore fails.

The scripts follow attached.

AttachmentSize
bancoerp2-restore0811.txt 3.98 KB
Mark_Solutions's picture

There are some long time gaps while this is running between the operations

It could well be that during the restore running it hits a timeout

It looks like the restore does actually work OK but it shows a failure as the client connect / readout timer has expired and so the job never gets its final status

I am guessing that the restored files are actually OK

If so then increase the client read and client connect timeouts on the Media Server and Client handling the restore and hopefully that will resolve the final status messages - try 3600 for both and try it again (but it looks like the restored files should actually be OK)

Hoep this helps

Authorised Symantec Consultant

Don't forget to "Mark as Solution" if someones advice has solved your issue - and please bring back the Thumbs Up!!.

andrich's picture

Mark

The weird is that my restore takes 10 (ten) minuts to give me the error and cretae the files, but the biggest file is a 100GB files. And my Netbackup generally takes 1 hour to restore 100GB.

Mark_Solutions's picture

A single 100GB file will always restore way faster than 100GB of document files - so maybe it is just that?

The log indicates that the process is taking much longer than that but maybe i have read the logs wrong - could we see the full text of the job detailed status

Authorised Symantec Consultant

Don't forget to "Mark as Solution" if someones advice has solved your issue - and please bring back the Thumbs Up!!.

andrich's picture

Mark

I was talking to my SQLServer DBA and he told me that the restore should be done to a differente server that contains a SQL instance installed.

This solved the case.

Thanks

SOLUTION