Netbackup 7.0 with Standalone drive

[pruvn]

Hi everybody,

We are implementing NBU at one of my sites and We just have an IBM System Storage TS2340 model L4 drive to work as a Standalone drive. IT's really difficult to manage & I have some questions that need help from you:

1. NBU 7 admin console do not give us a change to type in the label, neither do the bplabel command. Some tape this Standalone scan & label as A00000, a00001 ... some as BE0000 . How can we specify label for media in NBU 7 ?

Technical support guy give me a link : http://seer.entsupport.symantec.com/docs/296268.htm [tape handling instruction with Standalone drive, it just help me in scan inserted tape, but do not help me in labeling tape ...]

2. The jobs backup to tape keep hanging and said that the current media is completed and request next any available resource. Though the current media is not frozen & not suspended. This issue happens again and again whatever we do or change new tape ... Could you pls help me to solve this?

The detailed log is:

7/21/2010 11:41:01 AM - requesting resource ttnnetbk.NBU_CLIENT.MAXJOBS.ttnnetbk
7/21/2010 11:41:01 AM - requesting resource ttnnetbk.NBU_POLICY.MAXJOBS.thang
7/21/2010 11:41:01 AM - awaiting resource ttnnetbk-hcart A pending request has been generated for this resource request.
  Operator action may be required. Pending Action: No action.,
  Media ID: BE0000, Barcode: --------, Density: hcart, 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

Thanks so much.

Vic

[Marianne · Accepted solution]

I don't understand what you mean with the following statement:
"admin console do not give us a change to type in the label, neither do the bplabel command."

What happens when you try to to run bplabel or label from the GUI?

Say you have 6 physical tapes and you decide to put labels/stickers on them:
A00001
A00002
A00003
A00004
A00005
A00006

You next go to Media Gui and add new volumes: A00001 - 6
Ensure density matches that of the tape drive and select a pool.

Once added, you insert tape A00001 into the tape drive, wait for the 'Ready' light on the drive, then use bplabel as per the TechNote:
bplabel -m A00001 -d hcart -o -p NetBackup -u 0


Repeat for rest of tapes.

[pruvn]

Thanks, I mean:

In GUI: Administration console --> media --> right-click a media --> Label... --> OK to confirm media server --> OK to confirm lost data --> OK to see in Activity Monitor =>> Then we have no chance to specify the label cause we do not have an option to input our wanted label.

NBU run this process & keep the same label as it is recognized when we insert & run the command "vmphyinv -n IBM.ULT3580-TD4.000 -h <media server>" to scan new inserted tape.

In CMD: with the bplabel command, when we run the command (ex: bplabel -m A00001 -d hcart -o -p NetBackup -u 0), it require the media A00001 name is right [this is the exact name NBU recognizes when we run the command to scan new inserted tape]. If we change it to MON001 then NBU throw an error:

Error reported by media operation: resource request failed.
or
Error reported by media operation: requested media id is not assigned to this host in the EMM database.

That's what I want to say.

In my procedure, I insert a new tape. Then I run command "vmphyinv -n IBM.ULT3580-TD4.000 -h <media server>" to scan for new tape, at this time NBU automatically assign a label to my tape. Then I do not know how to change the label as I want.

Thanks for your quick comment,

Vic

[Marianne]

The "vmphyinv" is not a requirement.
If there is Backup Exec data on the tape, it has probably assisgned it as BE tape to be imported. I'd like to see what vmphyinv is actually adding to the EMM database.

Please post output of one of the media-id's added by vmphyinv:
nbemmcmd -listmedia -mediaid <ID>

[pruvn]

Here you are:

NBEMMCMD, Version:7.0
====================================================================
Media GUID:                     e0d252bc-d244-4396-bf85-60e9df2ad6d9
Media ID:                       BE0000
Partner:                        -
Media Type:                     HCART
Volume Group:                   ---
Application:                    Netbackup
Media Flags:                    1
Description:                    HI000002
Barcode:                        --------
Partner Barcode:                --------
Last Write Host:                NONE
Created:                        07/21/2010 11:39
Time Assigned:                  -
First Mount:                    07/21/2010 11:46
Last Mount:                     07/21/2010 13:29
Volume Expiration:              -
Data Expiration:                -
Last Written:                   -
Last Read:                      -
Robot Type:                     NONE
Robot Control Host:             -
Robot Number:                   -
Slot:                           -
Side/Face:                      -
Cleanings Remaining:            -
Number of Mounts:               5
Maximum Mounts Allowed:         0
Media Status:                   ACTIVE
Kilobytes:                      0
Images:                         0
Valid Images:                   0
Retention Period:               -
Number of Restores:             0
Optical Header Size Bytes:      0
Optical Sector Size Bytes:      0
Optical Partition Size Bytes:   0
Last Header Offset:             0
Adamm Guid:                     2124ea5d-04ba-40bd-bf61-743476717019
Rsm Guid:                       00000000-0000-0000-0000-000000000000
Origin Host:                    NONE
Master Host:                    ttnnetbk
Server Group:
Upgrade Conflicts Flag:
Pool Number:                    4
Volume Pool:                    Scratch
Previous Pool Name:             -
Vault Flags:                    -
Vault Container:                -
Vault Name:                     -
Vault Slot:                     -
Session ID:                     -
Date Vaulted:                   -
Return Date:                    -
====================================================================
Command completed successfully.

[Marianne]

There should be no problem writing to BE000, but this is not what you want, right?

"If we change it to MON001..."

So you want your labels to be MON001, MON002, TUE001, etc... ?

Please delete whatever is in the Media GUI (all volumes/media id's) that are still unassigned.

Manually add the volumes that you need (MON001, MON002....) Choose correct media type/density that corresponds with the tape drive density and select a pool (e.g. NetBackup).

Next, insert MON001 into the tape drive.

Label from CMD:
bplabel -m MON001 -d hcart -o -p NetBackup -u 0

Please let us know what the result is.

[pruvn]

There are 2 cases:

1. If we just insert the tape & do nothing to scan for the tape.

in CMD, run command: bplabel.exe -m MON001 -d hcart -o -p Scratch -u 0

=> Error reported by media operation: requested media id is not assigned to this host in the EMM database.

2. If we insert & scan for the tape. [Symantec support guide me to do this to scan inserted tape, and I assume that I have to do this any time I take a tape out & insert another one]

scan tape in CMD: vmphyinv -n IBM.ULT3580-TD4.000 -h ttnnetbk
=> NBU will automatically assign a label for the tape [tape status is assigned in Backup Exec pool or Netbackup pool] & we can not change the label ... [mentioned before]

So troublesome :(

I leave the assigned label with the tape intact, I just bpexpire the tape & move it to scratch pool. I create the policy to backup & backup job hang and and said that the current media is completed and request next any available resource. Though the current media is not frozen & not suspended. This issue happens again and again whatever we do or change new tape ... My second issue !!

[Marianne]

PLEASE do the following:

1. Delete whatever is in the Media GUI (all volumes/media id's) that are still unassigned.

2. Manually add the volumes that you need (MON001, MON002....) Choose correct media type/density that corresponds with the tape drive density and select a pool (e.g. NetBackup).

3. Next, insert MON001 into the tape drive.

4. Label from CMD:
bplabel -m MON001 -d hcart -o -p NetBackup -u 0

[Andy_Welburn]

labelling is just that, labelling a volume. It is not creating the volume in NB, you need to do that first before trying to associate a media label with it.

***EDIT***
Beat me to it with your reply - this was just a follow-up to Mariannes previous post!

[pruvn]

Hi Marianne,

You solved my first issue :D , thank you very much. Can't believe that Symantec Support guide us to follow a difficult way than your way !!

The backup job keep hanging & request next media :-s

Detailed Log:
7/21/2010 4:19:00 PM - estimated 0 Kbytes needed
7/21/2010 4:19:00 PM - started process bpbrm (4388)
7/21/2010 4:19:00 PM - connecting
7/21/2010 4:19:00 PM - started
7/21/2010 4:19:01 PM - connected; connect time: 00:00:01
7/21/2010 4:19:05 PM - mounting MON001
7/21/2010 4:19:06 PM - Error bptm(pid=5616) The tape device at index -1 has a maximum block size of 32768 bytes, a buffer size of 65536 cannot be used
7/21/2010 4:19:06 PM - Warning bptm(pid=5616) media id MON001 load operation reported an error    
7/21/2010 4:19:06 PM - current media MON001 complete, requesting next resource Any
7/21/2010 4:19:11 PM - current media -- complete, awaiting next media Any A pending request has been generated for this resource request.
  Operator action may be required. Pending Action: No action.,
  Media ID: MON001, Barcode: --------, Density: hcart, 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
 
7/21/2010 4:19:59 PM - granted resource MON001
7/21/2010 4:19:59 PM - granted resource IBM.ULT3580-TD4.000
7/21/2010 4:19:59 PM - granted resource ttnnetbk-hcart
7/21/2010 4:20:00 PM - mounting MON001
7/21/2010 4:20:02 PM - Error bptm(pid=5616) The tape device at index -1 has a maximum block size of 32768 bytes, a buffer size of 65536 cannot be used
7/21/2010 4:20:02 PM - Warning bptm(pid=5616) media id MON001 load operation reported an error    
7/21/2010 4:20:02 PM - current media MON001 complete, requesting next resource Any
7/21/2010 4:20:06 PM - current media -- complete, awaiting next media Any A pending request has been generated for this resource request.
  Operator action may be required. Pending Action: No action.,
  Media ID: MON001, Barcode: --------, Density: hcart, 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

[Marianne · Accepted solution]

This is your current problem:
Error bptm(pid=5616) The tape device at index -1 has a maximum block size of 32768 bytes, a buffer size of 65536 cannot be used


Something at SCSI and/or O/S level has set block size at 32768 - NBU uses default of 64K (65536).

This TN is VERY old (http://seer.support.veritas.com/docs/251807.htm),
 but use the method described in the TN to load the tape into the drive and use nt_ttu to determine block size.

Also have a look at experience shared by one of my colleagues:

Check the Windows MaximumSGList setting, it may be set so low that the default 64k buffer is not being allowed.

This snippet is from a Tivoli/QLogic setup, but the principle is the same.
 You need to identify the driver name for the SCSI or FC interface you are using to find the parameter.

<<
Important: If your environment includes an IBM LTO Ultrium tape drive and your storage agent will be located on a computer using Windows 200x, you must verify and possibly change the host-bus-adapter (HBA) setting. For
most HBA's this value is located in the MAXIMUMSGLIST registry setting.
The MAXIMUMSGLIST value on the storage agent must be equal to, or greater than HEX 41 (decimal 65) or a data integrity error may occur. To verify or change the current registry setting perform the following:

Edit the Registry
Click HKEY_LOCAL_MACHINE | SYSTEM | CurrentControlSet | Services |(vendor HBA name*) | Parameters | Device
Note:
For (vendor HBA name*) in the previous description, substitute the vendor's device name for your environment (for example, for the Qlogic 2200 and 2300, the name appears as Ql2200 and Ql2300 respectively). If your vendor's HBA name does not appear as a registry entry, contact the HBA vendor to determine the location of the setting and then verify that the setting is correct.
If the setting's hex value is 41 or higher, verification is complete.
Close the registry editor, and continue with the storage agent installation and configuration procedure. If this value is less than 41, change the value to 41 or higher, then save the new registry entry, close the registry editor, and continue with the storage agent installation and configuration procedure.
 
You can set any maximum block size between 64 Kbytes and 1 Mbyte. The formula to calculate the proper value for MaximumSGList is:

     MaximumSGList = ((Maximum Block Size)/4 Kbytes) +1

For 256 Kbytes: 256 Kbytes/4 Kbytes = 64. Add 1 for 65 (decimal) or 0x41 (hexadecimal). The maximum value allowed for MaximumSGLIst is 255 or 0xFF. For the particular value of 0xFF, the internal value passed to Windows is increased to 0x101, allowing support for a full 1 Mbyte transfer.

The system must be rebooted for the new registry setting to be effective.

NOTE: For 64-bit systems, the OS page size is 8 Kbytes instead of 4 Kbytes.
      Therefore, the maximum transfer size is 2 Mbytes, the formula becomes:
 
          MaximumSGList = ((Maximum Block Size)/8 Kbytes) +1
>>

[pruvn]

Thanks Marianne, that's right. I followed the kb & changed data buffer size to 32KB then the backup jobs ran successfully. Sr for my late response cuase I was off yesterday :)

After I changed buffer, I can backup successfully but the second time I ran a job to backup to tape, NBU can not find available media & failed with error code 96.

 7/23/2010 1:54:01 PM - mounting MON001
7/23/2010 1:54:03 PM - Error bptm(pid=5884) tapemark or blank tape encountered reading media header, drive index 0 
7/23/2010 1:54:03 PM - Error bptm(pid=5884) non NetBackup media found in drive index 0, FREEZING MON001  
7/23/2010 1:54:03 PM - mounted; mount time: 00:00:02
7/23/2010 1:54:03 PM - current media MON001 complete, requesting next resource Any
7/23/2010 1:54:12 PM - end writing
unable to allocate new media for backup, storage unit has none available(96)

[Marianne]

non NetBackup media found in drive index 0, FREEZING MON001 

Look at All Log Entries and bptm log to see if you can find more info.
Since you labelled the tape with NBU, you shouldn't encounter this error.

You will have to enable logging to troubleshoot tape drive problem:
bptm log
as well as VERBOSE entry in volmgr\vm.conf (restart NBU after adding this entry)
This will increase logging to O/S Event Viewer (if Windows).

The 32 Kb block size seems odd - I have never had to reduce NBU buffer size from the default of the 64Kb - we actually try to increase it wherever possible. Even 10-year old SCSI cards can do 64 Kb......

This TechNote mentions a similar problem with VTL, but since the actual problem is caused by HBA firmware and driver, it could very well be related:
http://seer.entsupport.symantec.com/docs/307482.htm

[Andy_Welburn]

Old, but the workaround may assist?
GENERAL ERROR: DLT media is frozen the second time it is used. A "media mismatch" error is seen.
http://seer.entsupport.symantec.com/docs/288474.htm

Driver issue?
error bptm pid=5008 tape device at index -1 has maximum block size of 32768 bytes buffer size of 65536 cannot be used
http://seer.entsupport.symantec.com/docs/303961.htm

[Denis_Koo]

Hi,

I have a similar question, but your information already solved a lot.

What if I need to replace a new tape for a backup (say A00001, assume that I use your procedure to label A0000 to A00006), Should I need to perform a inventory first to release A0000 from the LTO4, before the tape is recognise a new tape with lable A00001 to carry the backup?

What is the procedure to use A00001 after A00000

Denis

[Denis_Koo]

Hi,

I have a similar question, but your information already solved a lot.

What if I need to replace a new tape for a backup (say A00001, assume that I use your procedure to label A0000 to A00006), Should I need to perform a inventory first to release A0000 from the LTO4, before the tape is recognise a new tape with lable A00001 to carry the backup?

What is the procedure to use A00001 after A00000

Denis

[Marianne]

Inventory is associated with a robot, not Standalone drive.

When you eject A00000 from the tape drive, it will automatically be 'released' from the tape drive - this is handled by the avrd process (automatic volume recognition daemon).

As per previous post:

Go to Media Gui and add new volumes: A00001 - 6
Ensure density matches that of the tape drive and select a pool.

Once added, you insert tape A00001 into the tape drive, wait for the 'Ready' light on the drive, then use bplabel as per the TechNote:
bplabel -m A00001 -d hcart -o -p NetBackup -u 0

[Denis_Koo]

Thanks for your reply. But I never make it works.

I even use default nameing, say A0000 - 6

For the first tape, it will automatic assigned A0000, then I create A00001 - 6 in Netbackup pool with correct drive density. Then I usse command line to label.

Once i exec the command with your syntax, I got IPC error.

[Marianne]

Denis, this post has actually been solved.
Please start a new discussion, post the exact command that you are using as well as the exact error that is displayed.