Backup Exec not using disk with free space in disk pool

 First of all, are your disks physically located within the same machine? For backups that's a definite no-no.

What operating system is this running on?

If you are adamant about having them run inside the machine you are backing-up you can create dynamic disks, creating a kind of software RAID, therefore backup exec will see just one disk.
However this is not ideal, really the only solution for backups is tape, or remote-site backups (sorry to harp on about this but I had customers lose there entire business due to inadequate backups).

Jean

 Backup Exec is running on Windows 2003 Server.  And why would having the disks physically located on the Backup Exec server be a definite no-no?

The backups are done to disk first for performance reasons.  After the backups are complete, a job is run that duplicates the backups to tape, so there are copies that are stored off-site.  Also, having the backups available on disk provides for faster restores as well when dealing with restoring files or directories.

We are not going to RAID the disks together, because as the amount of data that needs to be backed up grows, so will the space need for backups, and all we would have to do is add another disk and make it a member of the pool.

What I want to know is if my assumption on how Backup Exec treats the disks in the pool is correct.

Hi,

No. When a disk target is created, you create a B2D folder on that specific disk. The job is then targetted at that, and will cease to write data to the drive once it has hit its limit that you would have specified.
It doesn't automacially move on to another folder like a tape to my knowledge, unless set up correctly.
You would need to look at your overwrite protection period, and make sure that is correct. If those settings aren't, and it isn't returning *.bkf files to scratch, it will fill up to the point of no further backups until you do something manually. Check that, and your append periods, and make sure they are correct.

You don't say how much space you are backing up, or what your process is (Full, Incremental or Differential).

In this case, as an easy fix, I would target the 1TB drive first. Maybe look at splitting some days off onto the smaller drive as well (like a weekly/monthly job).
Having internal drives doesn't mean a definite no-no, but RAID does help your protection of disks though. If you COULD do it, I would suggest that.

However, the guy above you has it all wrong too...tape is DEFINITELY not the way of the future. When you move on to BEWS 2010, and want to use data deduplication, you can only do this on disk. Restores are easier from disk; staging Information Store restores are easier when done to disk first; disk is considered faster than tape when backing up and restoring (for now...LTO5 is out!), so there are a lot of reasons for using disks.
Added protection of your data would see you stream the files off to tape once you have finished your B2D...

 I guess my question now is what is the advantage of using a disk pool?  If you still have to manage each disk individually anyway, then why put it in a pool?

From the Backup Exec Help, a device pool is defined as follows: 

Device pool
A grouping of devices for Backup Exec operations. Jobs assigned to the device pool are run on the first available device.

I dont understand why people are going off on one.

The question was simple why isnt the server using the rest of the disk which has free space.

It has nothing to do with what type of backup job. Where the disks are located or even the operating system.

If you dont know the answer then dont go off on one with all the junk that has no effect on disk pooling.


I will speak to tech support on this matter as I have the same problem and will get back to you.

The posting is old, and we never heard anything else from John so I suspect this is dead.
The idea is not to slag off people, and certainly not in public.

I did not post anything else as my questions were not answered, and no one seems to be able to provide me with one.  We are still experiencing the same problem.  Since version 10d is no longer supported, we have been looking at upgrading to the latest version of Backup Exec, but still have not been assured that we will not have this problem with the new version.  I have been avoiding the job failures by reviewing the disk space each night and making any changes necessary for the jobs to use whichever disk has the space needed.  If we do not get a definitive answer, we will end up changing from Backup Exec to something else, which is not what we want to do, but if given no other choice, we certainly will do.

BackupExec does not do any capacity screening before selecting the disk to use.  If the disk with the highest priority has ANY space on it, BackupExec will start writing to it.  When it fills the space up, with a standard B2D file, the job will fail

If you have created Removeable B2B devices on each disk, have placed both into a pool, and then pointed the job at the pool, when the first device fills up, the job should span to the second device