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

user initiated SQL backups are failing with 230 though the policy and schedule are in place.

Created: 04 Dec 2012 • Updated: 11 Feb 2013 | 15 comments
anishC5's picture
This issue has been solved. See solution.

User initiated SQL backups are failing with 230 though the policy and schedule are in place. the client name is getting recognised as A=a, policy= NULL.

Please could someone let us know where do we look and which logs do we refer. There is nothing on the detailed job screen to refer.

:(

Our env details are as follows:

NB master svr: LINUX 5.1

Client: win svr 2008

Thanks,

Anish

Comments 15 CommentsJump to latest comment

Marianne's picture

How is user backup started? By kicking of script from client or just selecting db's and running the backup?

We need to determine if SQLHOST is defined when sending request to master server.

dbclient log on the client as well as bprd log on the master server will contain this info.

If you are not using a script that defines SQLHOST and POLICY, Master needs to resolve client's IP address to hostname, then looks for the resolved hostname in appropriate policy.

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

anishC5's picture

Hi Marriane,

The backup is fired by a script.

We are trying to gain accessto the client for the dbclient log. We would also check the bprd log in master.

To check the name resolution shoudl we use the command bpclntcmd please?

Marianne's picture

Start by checking the script.

Look for :
SQLHOST
POLICY

This must be the same as in the policy on the master server.

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

anishC5's picture

Hi Marianne,

Any ideas please about the standard name of the script.

 

Thanks,

Anish

Marianne's picture

There is no 'standard' name. The extention should be .bch. Default location is ..\netbackup\dbext\mssql but it is also possible that script was saved elsewhere.
Maybe SqL dba / server owner will know?

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

RamNagalla's picture

 

I am more intrested to see the bch file which triggers the job..

FYI..

Policy Names are case sensitive, 

anishC5's picture

Hi Nagalla, Marianne,

I will be giving an extensive search for the bch file. One information I would like to share here is that the failure is occurring while the backup is trying to copy the flat files to tape, the databse copy to E:/SQLBK2FS is ok.

Please let me know if that changes our observation please?

Thanks for all your help,

Anish

Marianne's picture

Backing up this way is not a SQL agent backup.
Probably SQL backup to disk followed by bpbackup command to send flat file to NBU?

So, the script will not be a NBU .bch file.

You really need to work with your SQL dba to find out what exactly is happening on the client.

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

anishC5's picture

Hi Marianne,

yes please you are right. I just asked them to provide me the command / script they use to initiate the backup to tape and this is the response:

The below command is used to fire the backup to tape process:

[c:\progra~1\veritas\netbackup\bin\bpbackup -s DailySQL -L "E:\SQLBK2FSLOGS\TapeBackup_Server01_20121204_145810.txt" -w "E:\SQLBK2FS\Server01"]

I went to the help list of bpbackup and found the following:

bpbackup [-p policy] [-s schedule] [-k "keyword phrase"]
           [-L progress_log [-en]] [-S master_server...]
           [-t policy_type] [-w [hh:mm:ss]]

Does it need the "-p" POLICY switch adding please? is that mandatory?

Thanks,

Anish

Marianne's picture

-p <Policy> will certainly help, otherwise the first policy with 'DailySQL' schedule will be used.

Confirm that DailySQL is a User-Backup type schedule with open window when script will call bpbackup.

Next, confirm NBU 'Configured Client' name. 
Use Client BAR GUI: File -> Properties -> Client Name.

Ensure the name in this field corresponds with the Client name in the policy.

 

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

anishC5's picture

Hi Marianne,

 

We just checked the host file of one of the client and found that the backup IP entered there and the one being used for backup are different.

We have rectified the same and asked DBA to reinitiate the job once more to check.

Rest settings as advised by you looks ok.

I will let you know the results.

 

Thanks,

Anish

Marianne's picture

Please ensure that bprd log folder exists on the master server. If not, create folder and restart NBU Request Service.

This will help to troubleshoot connection from client.
bprd log will show Connection received from client's IP address and how master is resolving IP to hostname.

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

anishC5's picture

Hi,

Unfortunately the host file setting change did not help. :(

Anything else we can check please?

Thanks,

Anish

Marianne's picture

Check bprd log on the master as per my previous post.

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

anishC5's picture

hi All,
Sorry for not posting for this thread in a long time.
The issue is indeed resolved.
We observed that in registry settings of NBU, a client entry was made as A=a instead of the actual client name.
When we changed that, the issue was resolved.

SOLUTION