cancel
Showing results for 
Search instead for 
Did you mean: 

Newbie quesion: New set of hcart3 on a stand-alone drive

kilroy1103
Level 3

Hi all.

We have Veritas v6.0MP4 on a Windows Storage Server 2003 R2 using a hp ultrium stand-alone drive. This backs-up data from our Oracle RAC Server on schedules. We manually change the tape before every schedule runs: volume A00001 - every Sat, Sun, and Mon; volume A00002 - every Tue and Wed; and volume A00003 - every Thu and Fri. These three tapes are now full so we acquired a NEW SET of tapes and label them A00001 (for Sat, Sun, and Mon), A00002 (for Tue and Wed), and A00003 (for Thus and Fri) using blabel. Our problem is, no back-ups were ever performed when we started using the new set of tapes. The report status message said...

"Unable to allocate new media for backup; storage unit has none available." Status code 96

We informed our supplier about this and we were told to just use blank tapes without labels as Veritas will do the labeling and update itself of the volumes. We don't want to buy new tapes yet so we erased the labels using bplabel -erase but the above message still appears. We need to get the back-up schedules running asap. Please help. Thanks in advanced.


kilroy

1 ACCEPTED SOLUTION

Accepted Solutions

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
Trying the long erase is a good idea. If that doesn't work, contact your supplier - probably faulty tape.
Also enable VERBOSE logging to ensure that all device/media related errors are logged to Event Viewer.
Add VERBOSE on a new line in ...\Veritas\volmgr\vm.conf
Stop/Start Device Management service. Errors will now be logged in Event Viewer System and Application log.
Check that you have bptm log directory under ...\veritas\netbackup\logs.

If there's still a failure after the long erase, you should be able to find enough evidence of a faulty cartridge in these logs.

View solution in original post

16 REPLIES 16

Abesama
Level 6
Partner
Inside NetBackup catalog (...\NetBackup\db\images directory, that is) there are records for all the backup data on the original A00001/A00002/A00003 and NetBackup knows they are full.

That's why when NetBackup saw label A00001/A00002/A00003 (from the new 3 tapes), it decided not to write anything on them, although physically your new 3 tapes are blank and empty.

Unless you want to tell NetBackup to delete (bpexpdate command, that is) all the past backup history on the original 3 tapes (I guess you do not, because you bought new 3 tapes rather than recycling original 3 tapes) you should not label the new tapes as A00001/A00002/A00003.

Label them as A00004/A00005/A00006 using bplabel command syntax, and stick paper labels on them accordingly - then NetBackup will see new media called A00004/A00005/A00006 which it haven't used at all in the past, so it will happily start writing on them and use them until they become full.

(Seems you do not need assistance with bplabel command syntax, so I will leave it with you)

Abe

Abesama
Level 6
Partner
What your supplier said is correct - NetBackup decides what to use, if it sees a brand new tape.

But in your case, because the new 3 tapes were once labeled as A00001/A00002/A00003, the label "stays" on the tape although you ran bplabel command again to erase.

The "erase" option in the bplabel command syntax does not mean erasing the label itself - it means "keep the label" and "erase the rest(data)" on the tape.

That's why the backup still failed even after you ran bplabel again.

If you really want to delete/remove a label from a NetBacukp tape, use some other tool (NT backup or GNU tar binary) to write some data on to the tape - then the media ID (label) written on the tapes by NetBackup will be overwritten by that application.

And then next time you let NetBackup to see the tape, NetBackup will see this is not a NetBackup tape, and it will try label it with next available A0000? number before it writes any data on them.

Abe

kilroy1103
Level 3
Thanks also for the very clear explanation. But since this server was installed with no 3rd party app other than NetBackup, we'd prefer it to stay that way as the process of installing/running the new app might mess up something in the background. Is there any other way to delete/remove the labels using native NetBackup tools/commands? (I'm hoping there is one that we could use.)

Again, thank you so much.

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
Some documentation regarding Standalone drives:
How to perform Media Management with standalone Drives in NetBackup 5.x and 6.x

Also have a look at NetBackup Admin Guide II under 'Reference Topics' -> How NetBackup Selects Media -> Selecting media in stand-alone drives.
Extract:
With NetBackup stand-alone drive extensions, NetBackup tries to use any labeled or any unlabeled media that is in a stand-alone drive. This capability is enabled by default during installation. The media selection process is as follows:
....
If an appropriate drive contains a volume, NetBackup tries to select and use that volume.
....
NetBackup uses media that was not used previously. If the media is unlabeled, the following actions occur:
■ NetBackup labels the media.
■ NetBackup adds a media ID to the volume configuration, if necessary. If a media ID is added, the NetBackup Media ID prefix (non-robotic) is used as the first characters of the media ID.
■ If a media ID prefix is not specified, the default prefix is the letter A. For example, A00000.
■ NetBackup adds the requested volume pool to the volume configuration (if the backup policy specifies a volume pool).

.....

Disabling stand-alone drive extensions

You can disable the stand-alone drive extensions by clearing the NetBackup media server host property, Enable standalone drive extensions. If you clear this property, NetBackup uses the same method to select media for stand-alone drives as it uses for robotic drives.


Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
You will notice that the media that have been used for backups have an entry in 'Time Assigned'. Only these 'Assigned' media will show up in the media reports.
The status 96 is because NBU could not find suitable available media, i.e new tape in required pool or assigned tape with space available in pool with the required retention level.
Run 'available_media' report from a cmd window (in ...netbackup\bin\goodies) and compare output with policy/schedule requirements.

kilroy1103
Level 3
@Abesama

Apologies but I was so sure the renaming method will work but unfortunately, it didn't. We renamed the new set as A00005 for Tue-Wed sched, A00006 for Thu-Fri sched, and A00007 for Sat-Sun-Mon sched. We can see the the new labels in the Media twig under the Media and Device Management branch (see link below).
img163.imageshack.us/img163/7910/netbackup01.jpg

But this new set doesn't exist in th Media List twig in the Media ID drop-down list. Why is that? (see link below)
img4.imageshack.us/img4/822/netbackup02.jpg

And the same Status of Backup report message appears. (see link below)
img535.imageshack.us/img535/5051/netbackup03.jpg

@Marianne

Thanks for the link and info. I'm sure I've read that before but I'll be reading that again afterwards.

@All

My aplogies for the image hotlinking. Don't know how to do thumbnailing thingy.


-Kilroy

kilroy1103
Level 3
Yes, you read it right. My boss decided to shutdown our Oracle RAC servers and storage server because the aircon is malfunctioning. When we are in front of the servers, it's like we're in front of a furnace. So darn hot! And since they're already shutdown, we cannot try any of your suggestions yet. Luckily, our old Novell file servers and  previous program versions are still operational so we can use them temporarily for our data transactions. What a predicament we're in!

@Marianne

Thank you so much.  We'll try the available_media command once the servers are up and running. I guess we need to change also the RL of the new set and pattern them from the old ones.

-Kilroy

kilroy1103
Level 3
@Marianne

We tried the available_media command and this is what was shown:


media        media        robot        robot        robot        side/        ret    size        status
 ID        type        type          #        slot        face        level  KBytes
----------------------------------------------------------------------------
CatalogBackup pool


DataStore pool


NetBackup pool


None pool


RMAN_SSM pool

A00007        HCART3   NONE          -          -         -         -     -        AVAILABLE
A00002        HCART3   NONE          -          -         -         9     1042165754        FULL

RMAN_TF pool

A00006        HCART3   NONE          -          -         -         -     -        AVAILABLE
A00004        HCART3   NONE          -          -         -         9     1041105984        FULL

RMAN_TW pool

A00005        HCART3   NONE          -          -         -         9     24883488        ACTIVE
A00003        HCART3   NONE          -          -         -         9     999209248        FULL

As you can see, both A00006 and A00007 are "AVAILABLE" but not "ACTIVE" unlike A00005. I think this is the reason for our status code 96. Also I followed this link but up to step #2 only. Kinda doubtful of what might happen if I follow step #3 onwards. Here is what it generated using A00006 as the media id:


NBEMMCMD, Version:6.0MP4(20061102)
====================================================================
Media GUID:                     561be4c3-dff1-4f4e-a4f4-237ce1d435c1
Media ID:                       A00006                              
Partner:                        -                                   
Media Type:                     HCART3                              
Volume Group:                   ---                                 
Application:                    Netbackup                           
Media Flags:                    1                                   
Description:                    ------                              
Barcode:                        --------                            
Partner Barcode:                --------                            
Media Server:                   NONE                                
Created:                        02/02/2010 16:33                    
Time Assigned:                  -                                   
First Mount:                    02/02/2010 16:35                    
Last Mount:                     02/02/2010 16:35                    
Volume Expiration:              -                                   
Data Expiration:                -                                   
Last Written:                   -                                   
Last Read:                      -                                   
Robot Type:                     NONE                                
Robot Control Host:             -                                   
Robot Number:                   -                                   
Slot:                           -                                   
Side/Face:                      -                                   
Cleanings Remaining:            -                                   
Mounts:                         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:                     00000000-0000-0000-0000-000000000000
Rsm Guid:                       00000000-0000-0000-0000-000000000000
Origin Host:                    NONE                                
Upgrade Conflicts Flag:                                             
Pool Number:                    6                                   
Volume Pool:                    RMAN_TF                             
Previous Pool Name:             -                                   
Vault Flags:                    -                                   
Vault Container:                -                                   
Vault Name:                     -                                   
Vault Slot:                     -                                   
Session ID:                     -                                   
Date Vaulted:                   -                                   
Return Date:                    -                                   
====================================================================
Command completed successfully.

Still the same error messages as in the previous posts. Suggestions are much appreciated. Thanks again in advanced.


-Kilroy
Philippines

kilroy1103
Level 3
Hi. Well, there were another messages that appeared during backup. Here is the log:


Date        Time        Media Server        Client        Job ID        Severity        Description
02/09/2010        12:00:31 AM        backup        thp01ora        2825        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        12:00:32 AM        backup        thp01ora        2825        Error        cannot position media id A00005
for write
02/09/2010        12:11:31 AM        backup        thp01ora        2825        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        12:11:32 AM        backup        thp01ora        2825        Error        cannot position media id A00005
for write
02/09/2010        02:01:59 AM        backup        thp02ora        2827        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        02:02:00 AM        backup        thp02ora        2827        Error        cannot position media id A00005
for write
02/09/2010        04:00:34 AM        backup        thp02ora        2828        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        04:00:35 AM        backup        thp02ora        2828        Error        cannot position media id A00005
for write


No back-up was performed and it exited with a status code of 86. I've read that it has something to do with the tape drive or its driver. It was running perfectly before the arrival of the new set of tapes so I doubt it's the problem. What else could it be? Thanks again in advance.


-Kilroy
Philippines

kilroy1103
Level 3
Good day to all.

Thu-Fri schedule runs with no problem (using tape A00006) and it seems Sat-Sun-Mon schedule will be running good also. Only Tue-Wed schedule (using tape A00005) is having problems. The log says:


Date        Time        Media Server        Client        Job ID        Severity        Description
02/09/2010        12:00:31 AM        backup        thp01ora        2825        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        12:00:32 AM        backup        thp01ora        2825        Error        cannot position media id A00005
for write
02/09/2010        12:11:31 AM        backup        thp01ora        2825        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        12:11:32 AM        backup        thp01ora        2825        Error        cannot position media id A00005
for write
02/09/2010        02:01:59 AM        backup        thp02ora        2827        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        02:02:00 AM        backup        thp02ora        2827        Error        cannot position media id A00005
for write
02/09/2010        04:00:34 AM        backup        thp02ora        2828        Warning        cannot locate on drive index 0,
No more data is on the tape.
02/09/2010        04:00:35 AM        backup        thp02ora        2828        Error        cannot position media id A00005
for write

I've unfreeze and unsuspend tape A00005 before it's scheduled back-up runs. Still having the same error messages. Maybe long erasing the tape will reset it back?

Andy_Welburn
Level 6
You say:

"I've unfreeze and unsuspend tape A00005 before it's scheduled back-up runs"

& repeated messages from the logs:

"Error cannot position media id A00005 for write"

Does it keep freezing? If so there must be an issue with it. Give it a good check over externally. Maybe the tape leader has dislodged (can happen if tape is dropped)? Could you load the tape & access outside of NetBackup, maybe even just try & fast-forward, rewind?

kilroy1103
Level 3
After visual inspection, there seems to be nothing wrong with the tape. Well, there's nothing to inspect much except the sliding latch on the right-hand side (it's a HP LTO3 Ultrium C7973A btw). And comparing it with the other tapes, they look pretty much the same. Also, I don't remember mishandling these tapes as I'm the only one assigned to handle them right now but will soon endorse to a colleague. And yes, it keeps on freezing after exiting with status code 86.

"Could you load the tape & access outside of NetBackup, maybe even just try & fast-forward, rewind?"

Sorry I can't. There's no fast-forward or rewind button on the drive (HP StorageWorks Ultrium 960) or maybe am just too dumb to locate it. (Sorry... getting desperate now...)


-Kilroy
Philippines

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
Trying the long erase is a good idea. If that doesn't work, contact your supplier - probably faulty tape.
Also enable VERBOSE logging to ensure that all device/media related errors are logged to Event Viewer.
Add VERBOSE on a new line in ...\Veritas\volmgr\vm.conf
Stop/Start Device Management service. Errors will now be logged in Event Viewer System and Application log.
Check that you have bptm log directory under ...\veritas\netbackup\logs.

If there's still a failure after the long erase, you should be able to find enough evidence of a faulty cartridge in these logs.

Andy_Welburn
Level 6
Maybe it is just a plain old bad tape?

Even new ones could be 'faulty' (Friday afternoon jobs! ;) ).

Maybe worth getting your suppliers to replace it if it's a new one.

kilroy1103
Level 3
Finally, the tape with problem is now ok! Here's what I did:
  • Long-erased tape A00005 and re-label that same tape to A00008 but under the same volume pool.
  • Removed A00005 from the media list by following the instructions here.
  • Then added A00008 to the media list.

Commands used:

bplabel
bpmedia
bpimmedia
nbemmcmd
bptm
bpexpdate
available_media

I don't know what 's with media label "A00005". But by removing it from the back-up database and creating a new entry solved the status code 86 problem. Still, I have doubts because status code 86 may also caused by hardware-related issues so I still need to monitor this particular media. But as of this writing, back-up is now doing what it's supposed to do.

Thanks to all the people who took time in responding to my (noob) query (Abesama, Marianne and Andy) and pointing me to the direction where to look at. Your assistance is very much appreciated. I hope the next time I'm under the same predicament, you guys will be again available to assist me. Thank you so much. Maybe you could visit the Philippines some time, I'd like to you a drink or two. ;)

Andy_Welburn
Level 6
One way to see if your set-up doesn't like A00005 would be to label another tape with that tag! It shouldn't have any impact, but I know I've had many instances of "shouldn't have any impact" that have, even with 'trivial' items like names or passwords!