New media in standalone drive not being labelled.

[-matt-]

Netbackup 7.0 running on AIX 5.3

I have an IBM3592 HCART2 standalone tape drive with AVR enabled, but when inserting a brand new tape, no automatic labelling occurs.

"Enable standalone drive extension" is enabled on the Master server which is also the media server, but AVR just doen't seem to be kicking in.

I'm not an AIX admin but I've had the server restarted and tried a number of new tapes, but I can't make any progress with this.

Any ideas would be appreciated.

Thanks
Matt

[Marianne]

NetBackup will only assign mediaid and label automatically when a backup job starts.

It is a lot easier if you assign media id and label tapes manually.

Please see this discussion:

https://www-secure.symantec.com/connect/forums/unknown-or-corrupt-media-detected-drive

Also have a look at this TN: http://www.symantec.com/docs/TECH56903

[-matt-]

Hi Marianne.  Thanks for your post.

When running a backup I end up with a Pending action, as netbackup wants other media from the pool, which isn't loaded into the drive.

I want the new tape to be assigned the default "A" prefix by avr so I can move into a seperate volume pool and configure the backup policy to use that pool.

At the moment, the new tape does not appear in the media list so I can't do anything manually.

Regards

Matt

[Marianne]

From your description it sounds as if standalone drive extensions is disabled or NBU does not 'see' the tape in the drive. In both these cases NBU will select a previous tape that's not Full to append to. Does Device Monitor see the mounted tape in the drive before the backup is started?

Media id's can manually be added in the Media section of the GUI (Actions-> New -> New Volumes) or from cmd using vmadd.

Extract from Admin Guide I:

About adding stand-alone volumes

Because NetBackup does not label volumes until it uses them, you can add volumes
even though they do not reside in a drive. The additional volumes are available
for use if the volume in a drive becomes full or unusable. For example, if a volume
in a stand-alone drive is full or unusable because of errors, NetBackup ejects
(logically) the volume. If you add other stand-alone volumes, NetBackup requests
that volume; NetBackup does not generate an out of media error.

The easiest way to manage media with Standalone drives is to pre-add media id's, then insert them one-by-one in the drive and use bplabel from cmd to label them.

Also, to be 100% that NBU will not request a previously used tape, Suspend the tape after backup is completed or when you eject the tape from the drive. When the backup images expire, the tape will be automatically unsuspended.

You can use the Media section of the GUI to suspend or from cmd with

bpmedia -suspend -m <media-id>

[-matt-]

Hi Marianne,

Standalone drive extension is enabled, but I'm thinking it's more about the drive not seeing the tape, although from the actual drive led panel is does show a tape is loaded.

I created a new volume pool and manually created a media id, then changed my test backup policy to use the new volume pool.

When running the backup I receive a pending request;

"1/10/2011 10:03:49 AM - awaiting resource uk01wrs6008-hcart2 A pending request has been generated for this resource request.
     Operator action may be required. Pending Action: No action.,
     Media ID: A00002, Barcode: --------, Density: hcart2, Access Mode: Write,
     Action Drive Name: N/A, Action Media Server: N/A, Robot Number: N/A, Robot Type: NONE,
     Volume Group: ---, Action Acs: N/A, Action Lsm: N/A"

From device monitor if I drag the request to the tape drive, I receive a "The drive is not ready or inoperable" error.

[Marianne]

If the 'Ready' light is on, NBU should be able to pick up the mounted state.

The tape drive sends the 'Ready' state to the O/S, where NBU will pick up status to update Device Monitor.

Please insert the tape you want to use, wait for 'Ready' light, then post a screendump of Device Monitor or output from cmd:

/usr/openv/volmgr/bin/vmoprcmd -d

also from O/S perspective:

tctl -f <tape-device> status

About your feedback 'I created a new volume pool and manually created a media id...'

Did you also label the tape after you have added the media id?

It takes a bit of time to add and pre-label media, but avoids lots of headaches later on...

[-matt-]

output from the commands attached.  I can see that Ready = No, but does it go into a Yes status when the drive is actually in use?

[Marianne]

PLEASE try to pre-label the tape - while tape A00002 is in the drive, do

bplabel -m A00002 -d hcart2 -o -p <pool-name> -u 0

The 'Pending' request will go away and media will immediately be assigned to backup job.

[-matt-]

Hi Marianne.  That failed with the following;

# /usr/openv/netbackup/bin/admincmd/bplabel -m A00002 -d hcart2 -o -p Test-Backups -u 0
Error reported by media operation: resource request failed

Log states;

12:33:59.480 [585846] <16> manage_media_request: Error reported by media operation: resource request failed (800)
12:33:59.480 [585846] <16> bplabel: Mount of media A00002 failed (252)

[Marianne · Accepted solution]

OK - So NBU does not 'see' the tape mount....

I'm wondering about the TWO entries in vmoprcmd output:

Drv Type   Control  User      Label  RecMID  ExtMID  Ready   Wr.Enbl.  ReqId
  0 hcart2   AVR                -                     No       -         0
  0 hcart2   AVR                -                     No       -         0
 
                             ADDITIONAL DRIVE STATUS
 
Drv DriveName            Shared    Assigned        Comment
  0 IBM.03592E05.010      No       -
  0 IBM.03592E05.010      No       -
 
 

Dual path to drives?

Device names for the dual paths?

Please check output of the following:

/usr/openv/volmgr/bin/scan
/usr/openv/volmgr/bin/tpconfig -l
/usr/openv/volmgr/bin/tpautoconf -t

This is what my lab media server reports when I mount an unlabelled tape - note the Ready state.

# vmoprcmd -d

                                PENDING REQUESTS

                                     <NONE>

                                  DRIVE STATUS

Drv Type   Control  User      Label  RecMID  ExtMID  Ready   Wr.Enbl.  ReqId
  0 hcart2   AVR               No                     Yes     Yes        0

                             ADDITIONAL DRIVE STATUS

Drv DriveName            Shared    Assigned        Comment
  0 Drive000              No       -
#

I am not sure how to relate your tctl status output to Solaris mt command... This is what I get when I run 'mt -f /dev/rmt/0cbn status' when tape is mounted:

HP Ultrium LTO 2 tape drive:
   sense key(0x0)= No Additional Sense   residual= 0   retries= 0
   file no= 0   block no= 0

When unmounted:

#  mt -f /dev/rmt/0cbn status
/dev/rmt/0cbn: no tape loaded or drive offline

[-matt-]

Dear Marianne,

Its' been some time but I did eventually get to the bottom of the problem.

A simple, and wait for it, reboot of the tape drive resolved the issue!

Thanks for help and time on this.

Regards

Matt