Enormement de points de sauvegarde

[lp33]

Bonjour à tous,
 
Je travaille avec NetBackup et j'ai un problème avec la sauvegarde d'un serveur Windows 2008 R2.
En effet, je me retrouve avec des dizaines voire des centaines de "points de restauration" chaque jour de sauvegarde différentielle.
Mon planning de sauvegarde est le suivant :
- sauvegarde différentielle journalière
- complète hebdomadaire
 
Je n'arrive pas à comprendre le nombre conséquent de sauvegarde quotidienne (en moyenne, un point de restauration toutes les 3min !).
Le réglage de la Policy est exactement le même que pour d'autres serveurs où il n'y a aucun problème.
 
Auriez-vous une idée ?
 
Au passage, en pièce jointe ma console de restauration. Voyez la taille de l'ascenseur à droite...
 
Merci d'avance pour vos idées.

[Marianne]

Google Translate says:

Hi all,

I work with NetBackup and I have a problem with the backup of a Windows Server 2008 R2.

In fact, I find myself with dozens or even hundreds of "restore points" daily differential backup.

My backup schedule is as follows:

- Daily differential backup

- Weekly full

I do not understand the significant number of daily backup (on average, a restore point every 3min!).

Setting the Policy is exactly the same as for other servers where there is no problem.

Do you have any idea?

Incidentally, my attachment Recovery Console. See the size of the elevator on the right ...

Thank you in advance for your ideas.

[Marianne]

Do you see multiple backups in Activity Monitor for this client?

Same policy with multipe streams or same policy kicking of multiple times during backup window?

Please show us your policy config for this client. From cmd, run this command:

bppllist <policy-name> -U
(command is in ...\netbackup\bin\admincmd)

[Yasuhisa_Ishika]

Just multistreaming backups with auto-discovery, I bet.

[lp33]

I post you this in a minute...

Update : OK. The settings are uploaded.

And from the console :

Policy Name:       SERVER01

  Policy Type:         MS-Windows
  Active:              yes
  Effective date:      18/09/2012 17:29:49
  Backup network drvs: no
  Collect TIR info:    no
  Mult. Data Streams:  no
  Client Encrypt:      no
  Checkpoint:          no
  Policy Priority:     0
  Max Jobs/Policy:     Unlimited
  Disaster Recovery:   0
  Collect BMR info:    no
  Residence:           STU_dedup
  Volume Pool:         NetBackup
  Server Group:        *ANY*
  Keyword:             (none specified)
  Data Classification:       -
  Residence is Storage Lifecycle Policy:    no
  Application Discovery:      no
  Discovery Lifetime:      0 seconds
ASC Application and attributes: (none defined)

  Granular Restore Info:  no
  Ignore Client Direct:  no
Enable Metadata Indexing:  no
Index server name:  NULL
  Use Accelerator:  no
  HW/OS/Client:  Windows-x64   Windows2008   server01.domain.local

  Include:  D:\

  Schedule:              Full
    Type:                Full Backup
    Frequency:           every 7 days
    Maximum MPX:         1
    Synthetic:           0
    Checksum Change Detection: 0
    PFI Recovery:        0
    Retention Level:     3 (1 month)
    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
    Schedule indexing:     0
    Daily Windows:
          Saturday   21:00:00  -->  Sunday     12:00:00

  Schedule:              Differential-Inc
    Type:                Differential Incremental Backup
    Frequency:           every 1 day
    Maximum MPX:         1
    Synthetic:           0
    Checksum Change Detection: 0
    PFI Recovery:        0
    Retention Level:     3 (1 month)
    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
    Schedule indexing:     0
    Daily Windows:
          Monday     22:00:00  -->  Tuesday    06:00:00
          Tuesday    22:00:00  -->  Wednesday  06:00:00
          Wednesday  22:00:00  -->  Thursday   06:00:00
          Thursday   22:00:00  -->  Friday     06:00:00
          Friday     22:00:00  -->  Saturday   06:00:00

[Yasuhisa_Ishika]

no multistreaming, I lose.

How the activity monitor shows?

There must be numerous jobs for this client. Check if policy name is displayed as SERVER01.

if so, someone initiate manual backups through GUI or command.

[lp33]

Yes, there are many many jobs for this client in the activity monitor... But I just don't understand why it doesn't just backup once, like all the other servers...

No one initiate manual backups, especially when they're launched (post 10pm, there's just no one in the office ;)). And I think that manual backups are showing with a single red dot, not the blue one as programmed differentials, no ?

[Andy_Welburn]

The single red-dot in the BAR GUI is for a user initiated backup from the client which is 'allowable' by the presence of a 'User Backup' type schedule - but that's strange as from the bppllist output you've provided there isn't such a schedule in place in that policy.

The blue-dots are incrementals whether initiated manually or via the schedule.

Is the bppllist output definately for the same client as the output from the BAR GUI?
Is the client in any other policies?

When you click on the blue-dots what sort of content is apparent? The same or backing up different 'things'?

From the output you've supplied I cannot see *any* reason why you should have so many backups for this client (unless it is being produced via another policy). I was expecting something similar to Yasuhisa, multiple data streams with a backup selection of D:\path_to_somewhere\* which could create '00's of jobs.

If you've supplied all the info for this client & this is definately the only policy that it is a part of, then I would be tempted to create a new policy (do not copy the original) to see if that remedies the situation.

[Marianne]

We have often seen this over the years across NBU versions but we have not been able to pinpoint the problem. There seems to be some sort of corruption between the policy and the scheduler.

The only solution so far has been to create a new policy or else delete the STREAMS file for this policy on the master server under ...\netbackup\db\images\<client-name>.

Take care that you only do this before a FULL is due to prevent incremental from running as Full.

[lp33]

Yep, this server is only in this policy. I had a manual backup schedule for testing at the beginning, so it's normal I had red dots ;)

In BAR, I almost have nothing to restore (only empty folders).

I'll try to create a new policy to see if it's better then.

[lp33]

OK. I'll create a new policy and delete everything from netbackup/db/images/<client-name>. I'll let you know if it corrects the issue. Thanks.

[Andy_Welburn]

I'll try to create a new policy to see if it's better then.

Further to Mariannes post below, if you go this route you'll have to run a Full backup (manually or schedule it to run first initially maybe) for this new policy otherwise all your Incrementals until the next scheduled Full will run as Full backups.

[Andy_Welburn]

I'll create a new policy and delete everything from netbackup/db/images/<client-name>

NO! NO! NO!

& again NO!

[Marianne]

ONLY delete STREAMS file - NOT everything from <client-name> folder....

[lp33]

What are you calling "streams" ? In this folder, I just have three folders named "1347000000", "138000000" and "139000000". Then in every each of them, folders "catstore" and "tmp", and a ".lck" file + many ".f" and ".lck" files. Which one have I to delete ?

[Andy_Welburn]

I would just go thru' setting up a new policy for this client then.

Could be STREAMSpolicy only created if you have multiple data streams allowed in your policy, which you haven't. (That was certainly the case with a test I've just done, but on NB6.5.6)

[Marianne]

Seems STREAMS file is only created when multistreaming is done. Apologies for causing confusing. Just create a new policy.

[lp33]

OK guys... New policy created, old one deleted, manual full backup made yesterday afternoon (OK) then wait for differential backup of last night and... again, hundreds of restoration points ! About 3min/5min between every one of them. I really don't understand what is the trouble here. And every differential backup referenced in the activity monitor is marked as "Done", no error, no warning. I don't know what to do.

[Andy_Welburn]

I really don't understand what is the trouble here

Well that makes two of us then.

Can you confirm NetBackup versions for client & master/media? Worth re-installing on client?

Do you have bpbkar log enabled on client? May tell you exactly *what* it's backing up ...... but not *why* & that's the issue!

Was thinking along the lines of exclude/include lists, but *again* that wouldn't explain the vast number of jobs this policy/client is spitting out.

[lp33]

Well, it's good to know I'm not alone not to understand ^^

The Netbackup version for client and master is 7.5 on each. I can try to re-install the client.

I just enabled bpbkar log, but like you said, it'll not tell me why it's backing up so regularly.

I do not use any exclude files list, and I tried to change what is backed up (the full partition, then only the folders, then only one folder...) and it's always backing up hundreds times...

I'll try the client reinstall and let you know.

Thanks.