Backup on Scratch media pool instead of CatalogBackup pool

[pisquik]

Hi all,

I have a problem with netbackup v 7.0.1. My catalog policy is backed up on wrong media pool : Netbackup take tape into scratch media pool instead of CatalogBackup pool..

I have tried to move the catalog tape into scratch media with bpexpdate command and move it again into CatalogBackup volume pool and netbackup still take tape into scratch volume pool..

On my Policy options i have fill "policy volume pool" to "CatalogBackup" so why it takes tape into scratch media ?

My media is full ?

Thanx for your answer and Sorry for my english :)

[Andy_Welburn]

Are you sure it has actually *written* to media from the scratch pool & that said media is *still* in the scratch pool?

Is it possible that, as your current catalog media was full, it hasn't just taken media *from* the scratch pool & it is now in the CatalogBackup pool & it's just your GUI display that needs 'refreshing'?

For a catalog backup policy (Policy type: NBU-Catalog) it is impossible for you to select any other volume pool than the default CatalogBackup (new this was wrong even as I was typing it - see EDIT!!)

Please show us the output of:

vmquery -m media_id             (for this suspect media)

and

bppllist policy_name -U           (for your catalog backup policy)

***EDIT***

Sorry, just had a deeper look - has someone changed your Scratch_Pool to be a Catalog Backup pool

e.g. by right-click/change on volume pool (this would also mean that your scratch pool is no longer a scratch pool)

Can check via CLI by:

vmpool -list_catalog_backup

vmpool -list_scratch

[pisquik]

It takes a media into the scratch pool and it move it on the catalogBackup pool to write the on it.

If the tape is full, how can i erase it ? Because i have tape only reserved for this type of backup.

vmquery output :

========================================================

media ID:              HZM500
media type:            1/2" cartridge tape (6)
barcode:               HZM500L4
media description:     -------------
volume pool:           CatalogBackup (3)
robot type:            TLD - Tape Library DLT (8)
robot number:          0
robot slot:            41
robot control host:    cbt-backup
volume group:          000_00000_TLD
vault name:            ---
vault sent date:       ---
vault return date:     ---
vault slot:            ---
vault session id:      ---
vault container id:    -
created:               4/29/2011 10:33:54 AM
assigned:              1/25/2013 3:02:51 PM
last mounted:          2/14/2013 9:59:33 AM
first mount:           4/29/2011 11:01:47 AM
expiration date:       ---
number of mounts:      378
max mounts allowed:    ---
status:                0x0
========================================================

bppllist output :

Policy Name:       CatalogPolicy1

  Policy Type:         NBU-Catalog
  Active:              yes
  Effective date:      02/14/2013 09:57:06
  Mult. Data Streams:  no
  Client Encrypt:      no
  Checkpoint:          no
  Policy Priority:     0
  Max Jobs/Policy:     1
  Disaster Recovery:   0
  Collect BMR info:    no
  Residence:           cbt-backup-hcart-robot-tld-0
  Volume Pool:         CatalogBackup
  Server Group:        *ANY*
  Keyword:             (none specified)
  Data Classification:       -
  Residence is Storage Lifecycle Policy:    no

Granular Restore Info:  no
Ignore Client Direct:  no
  HW/OS/Client:  PC            Windows2003   cbt-backup

  Include:  CATALOG_DRIVEN_BACKUP

  Schedule:          Full
    Type:            Full Backup
    Frequency:       every 7 days
    Maximum MPX:     1
    Synthetic:       0
    PFI Recovery:    0
    Retention Level: 1 (2 weeks)
    Number Copies:   1
    Fail on Error:   0
    Residence:       (specific storage unit not required)
    Volume Pool:     (same as policy volume pool)
    Server Group:    (same as specified for policy)
    Residence is Storage Lifecycle Policy:     0
    Daily Windows:
          Sunday     00:00:00  -->  Sunday     24:00:00
          Monday     00:00:00  -->  Monday     24:00:00
          Tuesday    00:00:00  -->  Tuesday    24:00:00
          Wednesday  00:00:00  -->  Wednesday  24:00:00
          Thursday   00:00:00  -->  Thursday   24:00:00
          Friday     00:00:00  -->  Friday     24:00:00
          Saturday   00:00:00  -->  Saturday   24:00:00

  Schedule:          Differential-Inc
    Type:            Differential Incremental Backup
    Frequency:       every 1 day
    Maximum MPX:     1
    Synthetic:       0
    PFI Recovery:    0
    Retention Level: 1 (2 weeks)
    Number Copies:   1
    Fail on Error:   0
    Residence:       (specific storage unit not required)
    Volume Pool:     (same as policy volume pool)
    Server Group:    (same as specified for policy)
    Residence is Storage Lifecycle Policy:     0
    Daily Windows:
          Sunday     00:00:00  -->  Sunday     24:00:00
          Monday     00:00:00  -->  Monday     24:00:00
          Tuesday    00:00:00  -->  Tuesday    24:00:00
          Wednesday  00:00:00  -->  Wednesday  24:00:00
          Thursday   00:00:00  -->  Thursday   24:00:00
          Friday     00:00:00  -->  Friday     24:00:00
          Saturday   00:00:00  -->  Saturday   24:00:00

Catalog Disaster Recovery Configuration:
  Email Address:   (none specified)
  Disk Path:       E:\CatalogDisasterRecoveryFiles
  User Name:       (none specified)
  Pass Word:       (none specified)
  Critical policy: (none specified)
 

Thanx for your Help !

[Andy_Welburn]

It takes a media into the scratch pool and it move it on the catalogBackup pool to write the on it.

If the tape is full, how can i erase it ? Because i have tape only reserved for this type of backup.

That's how NetBackup works - the media will only be available again for use once *all* the images on it have expired which is based on the retention period & when the last image was written, in your case the media will be available again 2 weeks after the last image was written.

You'll need to be aware & make provision for the fact that you're going to need more than one tape to fulfil your catalog backup requirements for the amount of data you're saving & your retentions.

***EDIT***

If you really, really, really do need to re-utilise this tape then you could use bpexpdate to expire it - but if all your catalogs are on it you obviously won't have one & sod's law says that when this is done you're going to need it!

***EDIT #2***

Because i have tape only reserved for this type of backup.

I read this as having only *one* tape for catalog backups, but I suppose it could mean several. If you only want to use specific media from specific volume pools for specific policies & not draw from scratch you'd essentially have to stop using a scratch pool altogether, but this would mean that you would have to ensure that there are media available in *all* volume pools when backups are run otherwise you leave yourself open to job failures - not good. Or just make sure that there are enough available catalog media in the catalog volume pool to cover your requirements at all times in which case the catalog backup should *never* draw from scratch.

Apologies for the numerous edits, second edits and re-edits, but there are so many possibilities of what you can do (& shouldn't for that matter) & personally I would always try & reduce the amount of micro-management needed to successfully maintain a NetBackup environment - if you let it, NetBackup will do it all for you!

[Marianne]

In addition to Andy's excellent explanation, you can run this report from cmd on a regular basis:

available_media

This commands is in 'goodies' folder.
If you have lots of tapes, you may want to redirect output to a text file.

[pisquik]

Actually i have 2 tapes for the catalog backup and I swapp these tapes every 6 month. (1To of data for each tape is enough for 6 month of catalog backup).

How can i know the available space on my tape ?

With my configuration, do you know if i can save 6 months of catalog backup on my tape or it's to long ?

[Andy_Welburn · Accepted solution]

How can i know the available space on my tape ?

Unfortunately not possible - well not accurately.

With my configuration, do you know if i can save 6 months of catalog backup on my tape or it's to long ?

That would all depend upon the size of your catalog & how much it changes daily (as you do differentials).

Actually i have 2 tapes for the catalog backup

- presumably one of these remains in the library & the other is offsite? So your library blows up taking your on-site copy & all your 'normal' backup tapes with it. In your anxiety to get a recovery underway you're all fingers'n'thumbs & so drop your offsite catalog tape ..... oh dear!

I know nothing of your environment nor your business requirements, but we used to take media off site on a weekly/monthly basis and an up-to-date catalog backup/tape too. Our catalog tapes were very probably *never* full, they weren't in the library long enough. These tapes stayed together until recalled/expired a month or year later. Essentially, we could have had around 16 catalog tapes on the go at the same time. Obviously the data on the older media becomes less and less relevant but I still didn't think it too many.

What was important to us was the ability to restore in the event of a disaster, & to at least something relatively recent - the number of tapes we needed was unimportant it was the data itself that mattered.

Imagine the 'hypothetical' scenario outlined above, you're coming up to 6 months, nearly time to change your catalog tape when the unthinkable happens - your most recent catalog is 6 months old!

***EDIT***

Sorry I'm editting again!!

What was it Marianne said the other day in another discussion? Something along the lines of:

"Don't base your backup requirements on the number of tapes you have, base the number of tapes you need on your backup requirements"

.... goes without saying really.

[pisquik]

it's very weird : I put all my tape of the scratch pool into Netbackup pool (pool used for data backup), I launch a catalog backup and NetBackup tooks the good tape into the catalogBackup pool -> So the tape is not frozen and not full !

So why when i have tapes into scratch pool NetBackup take these tapes ???

Thank's again for your help !

[Andy_Welburn]

Not sure I'm 100% with you here but ....

- if there are media available in the volume pool NetBackup will use that/those in preference to scratch media (by available we mean: not full, not frozen, not suspended, not physically expired, are of the same 'density', not physically expired, are in library etc & has images of same retention (by default)).

- if there are no media within the volume pool then NB will select an appropriate one from scratch.

If this *hasn't* actually occurred I couldn't explain it, but fulfilling the above 'requirements'/criteria is what *should* happen.

Some of the formatting leaves a lot to be desired, but:

About selecting media in robots on UNIX/Linux
http://www.symantec.com/business/support/index?page=content&id=HOWTO32854

[Dan4]

Pisquik,

Not sure if it helps but in your catalog backup policy, did you check Schedules. under Schedules (either full or incremental) you need to make sure "Override Policy Volume Pool" nothing is mentioned. If it is mentioned, teh backups will be directed to that volume pool instead of correct pool.

Please check and let us know.

Thanks.

+ Dan