On 11/10/10 09:32, Igor Zinovik wrote:
>   Hello.
> 
> I'm deploying bacula in our network.  Before it will go into production I need
> to solve one problem for myself how should I manage pools in my setup.
> 
> I have an ordinary linux box running centos and bacula 5.0.3  I'm going
> to store all my copies on NFS share that is mounted from netapp NAS.
> 
> I read bacula documentation and I always see that it says that Pools
> are very good for managing tapes, but what about disks?  Should I ever
> bother about defining several pools for disks, e.g. should I create
> pool for each client
> so that bacula would write all data thats belong to special client into 
> separate
> pool and though into separate volume.  And will have something like this:
> client1 -> pool1 -> client1-vol
> client2 -> pool2 -> client2-vol

This would be a very bad idea, because it will effectively mean that you
can only ever have one job running at a time per storage device.  Any
storage device can only have one volume at a time mounted, and if each
client has its own individual pool and can use volumes only from that
pool, then every job has to wait for its turn to own the storage device
to mount a volume it's allowed to write to.

> Or maybe I should not bother about Pools in my disk setup?  I have rather
> big NFS share which capacity is about 2 terabytes.  netapp NAS
> protects my copies
> with raid-dp (modified raid6 that protects against double disk
> faults).  Maybe i should
> just use one `Default' pool and should not care about pool management.

This would mean that all volumes have the same retention.  Whether this
is a problem for you depends on how you choose to use and manage your
volumes.

> Or maybe it is better to create separate pool for full, incremental
> and differential backups?

This is the way I do it.  I use separate Full, Differential and
Incremental pools, each with its own volume retention time.  Volumes are
automatically created and autolabelled by date as needed, with a volume
use duration window to make sure each volume is used for only one set of
backups.  Purged volumes are recycled into the scratch pool, and an
admin job goes through the scratch pool once a week, finds al;l of the
purged volumes, and deletes them both from the catalog and from disk.
Full backups additionally get copied to tape after completion, by a
separate backup job that runs after all full backups have completed.


-- 
  Phil Stracchino, CDK#2     DoD#299792458     ICBM: 43.5607, -71.355
  alaric AT caerllewys DOT net   alaric AT metrocast DOT net   phil AT 
co.ordinate DOT org
         Renaissance Man, Unix ronin, Perl hacker, Free Stater
                 It's not the years, it's the mileage.

------------------------------------------------------------------------------
The Next 800 Companies to Lead America's Growth: New Video Whitepaper
David G. Thomson, author of the best-selling book "Blueprint to a 
Billion" shares his insights and actions to help propel your 
business during the next growth cycle. Listen Now!
http://p.sf.net/sfu/SAP-dev2dev
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users