none
DPM continually gives 'out of disk space' errors RRS feed

  • Question

  • I've just finished building a new DPM 2010 server. This has available ~4TB of storage for replica's, and I'm not using tape. The server is patched to the latest level (build 7707.0)

    I've created the first protection group, and am primarily adding Windows 2008 R2 servers, intending to protect SS and BMR on each server. The problem is that, even through the protection group settings are set to 'Auto Grow', I'm continually getting 'DPM is out of disk space for the Replica' errors, even on first backup! If I manually add extra replica space (which I believe I shouldn't have to if Auto Grow is set), it get's so far, then fails with the same error. The server I'm trying to backup (2K8R2) only has a 25GB system disk. I can repleat this on other servers. Even for servers that might complete the initial replica's will eventually fail with the same disk space error!

    This is a really frustrating problem that I'd like to get resolved. What's the point of allowing the replica's in the protection group to autogrow, if it doesn't make any difference. This really should be a 'set and forget' solution. I also don't want to start adding silly amounts of disk space to the replica's, as this will quickly exhaust the space I have available.


    Monday, August 8, 2011 2:28 PM

All replies

  • Hi,

    We log autogrow activity in the application event log, filter on 33143 and 33144.  Autogrow grows - 10GB or 25% of volume whichever is larger.

    Log Name:      Application
    Source:        MSDPM
    Date:          MM/DD/YY HH:MM:SS AM
    Event ID:      33143
    Task Category: None
    Level:         Information
    Keywords:      Classic
    User:          N/A
    Computer:      DPMSERVER.DOMAIN.com
    Description:   The volume for load5 has grown from 1.62 GB to 11.62 GB.

     

    Log Name:      Application
    Source:        MSDPM
    Date:          MM/DD/YY HH:MM:SS AM
    Event ID:      33144
    Task Category: None
    Level:         Warning
    Keywords:      Classic
    User:          N/A
    Computer:      DPMSERVER.DOMAIN.com
    Description:   DPM was unable to expand the volume for <DataSource_name>.

     

    There are several situations in which autogrow may not happen and this activity is also logged in C:\Program Files\Microsoft DPM\DPM\Temp\DPMAccessManagerCurr.errlog

    1. If another alert of the same type for the same datasource was raised and resolved within 15 minutes of the current alert, then it will fail.  Retry another backup after 15 minutes.  In this case, the following would have been logged:
    AutoHeal: A grow alert was recently resolved; AlertId=[95b6c1f0-5c81-4a3a-84e5-7d75f3ca52ee]

    2. The ProtectionGroup that contains the data source does not have AutoGrow enabled.  If so, the following will be logged:
    AutoHeal: AutoGrow not enabled for datasource. AlertId=[95b6c1f0-5c81-4a3a-84e5-7d75f3ca52ee]

    3. The Disk threshold exceeded or Recovery point volume threshold exceeded alert was not raised.  This can be checked from the DPM alerts.

    4. If the system is close to hitting the LDM limit, then the autogrow will fail and log the following:
    AutoHeal: LDM database within error threshold, skipping grow; AlertId=[95b6c1f0-5c81-4a3a-84e5-7d75f3ca52ee]


    Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.
    Monday, August 8, 2011 3:04 PM
    Moderator
  • Thanks. I've managed to get one (problematic) server to backup now. The DPM estimated amount of space for the BMR/SS backup was calculated as 30GB, which I generally leave alone, but this always caused the replica to fail. I removed the protection group, and created it again with the same server. This time I manually set the initial volume size to 150GB, and ran the backup. This did complete ok, and the final reported size for the backup is 103GB! I'm going to try some other servers, but wondered if there's a manual way of calculating what the actual amount of data will be for a BMR/SS backup on a server, so if I need to override the recommended settings, I can.

    One other question. If you are doing a BMR backup of a server, and the server just has a single system disk (say 50GB), I understand the BMR backup should never exceed 50GB. If you now have a physical disk with a 50GB system partition and 150GB data partition, would BMR still expect to backup 50GB, or would it try to backup 200GB (i.e. the entire physical disk)? I assume if you're doing a BMR restore, then DPM will want to restore the ENTIRE 200GB physical disk, and not just the (50GB) system partition.

    Tuesday, August 9, 2011 1:42 PM
  • HI,

    Yes, that explains why you had trouble.  Since each protected servers will most likely require varying amounts space, you ca run this command manually on each server to see what volumes are included in the backup.

     

    1) Set up a network share on a remote machine \\server\bmrshare

    2) From an administrative command prompt on the Protected Server, type:

                  wbadmin.exe start backup -allcritical -backuptarget:\\server\bmrshare


    This should show you the list of volumes included in the BMR backup and ask "Do you want to start the backup operation?. - Type Y to continue if you want to see how big the resulting backup is, or just say N and you can calculate space required based on the volumes it included.


    Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.
    Tuesday, August 9, 2011 2:22 PM
    Moderator
  • Thanks. I've seen reference to this command a few times now, and while it does show the volumes that will be backed up, it doesn't give an actual summary of the amount of data it will backup. I suspect you'd have to select [Y]es and let it run to see for sure, but this is not ideal.

    Wednesday, August 10, 2011 11:32 AM