Can't Label Media Attached to NetApp NDMP Host

[Dan_DeFord]

We are currently running Netbackup 6.5.4 on a Solaris 9 master/media server, which is connected to a fiberchannel hub along with our tape library and NetApp filer (configured for NDMP).  The changer has two robots and eight DLT drives.  The NetApp controls the TLD(1)robot device and four of the drives while TLD(0) and the remaining four drives are used for everything else (Windows/Solaris clients) and controlled by the NetBackup master/media server.

Everything works fine.

However, I've just been tasked with configuring a Windows 2008 server running NetBackup 7.0, which is configured exactly the same way (e.g. connected to the same fiberchannel hub, NDMP host set up the same, policy set up identically, our NetApp NDMP config is set to allow ALL hosts, etc.)

I can back up the normal clients (tested both Windows and Solaris) using the non-NDMP robot (TLD(0)), but whenever I attempt to back up the NetApp filer, I get an error that it can't find any available media to perform the job. 

 unable to allocate new media for backup, storage unit has none available(96) 

So I tried to label/erase some scratch media in a separate Volume Pool that is accessible by TLD(1).  But when I try, I get the following error in NB:

 error requesting media (tpreq)(98) 

The Windows event viewer shows the following errors as well:

 TLD(1) key = 0x5, asc = 0x3b, ascq = 0xd, MEDIUM DESTINATION ELEMENT FULL 

 TLD(1) Move_medium error 

 Invalid operation in move_medium(), op = 31 

I realize that having multiple master/media servers attempting to connect to a single NDMP server and tape library probably isn't the best situation, but I don't see any way around it (although I AM open to suggestions).

Any help would be greatly appreciated!

[mrinal_sarkar62]

Hi,

The error can occure due to various factors.

You need to pin point by testing one by one.............no other short cut........sorry

Check the following:-

1. Check with a new tape volume, with unasigned tape, to check there is no issue with tape volume/tapes..tape expire sort of problem .........and like wise..issue...

2. Check for the Netbackup device manager service is configured correctly..

3. Check for the network configuration with the librery(if there is multiply network u need to take car of.. ....that...).

4. Check for bptm debag log. If there is a problem look for the log in the path :-

nstall_path\Volmgr\debug\daemon\xxxxxx.log  for what has happened.

5. Is there no DNS related issue. Check the host file if required.

Thanks.

[Karthikeyan_Sun]

I See this messgae looks some problem with Library Side..

How many tapes in Slot allocated for TDL(1)?

How many tapes in drives allocated TLD(1)?

Do a fresh inventory of TLD(1) and keep Drives of TLD(1) and tapes settled.

After inventory do you see media under TLD(1) ? Change the volume pool appropriately and re fire backup.

As per the error message you showed, NB is trying to eject a tape to slot thats already full.

[mrinal_sarkar62]

Hi,

Yes Sundaram that's what I gussed...

There may be some tape related issue...

But as DAN identified that it is a new server configured, might happen any other like service related. That's what I have given step by step as provided by symantec.

To seperate, it is a tape issue, please follow my previous post.

But first of need to check that tape related issue.

If the jammed issue try this:-

 robtest

http://seer.entsupport.symantec.com/docs/274063.htm

You will probably need to use the unload & m options.

If you do remove a tape from your library in this way you will need to inventory the robot via NetBackup to record that the tape has been moved/removed otherwise this could cause you issues further down the line.

Thanks.