Forum Discussion

Peter_Sheridan's avatar
Peter_Sheridan
Level 6
13 years ago
Solved

BE 2012 Backup Stalled on Insufficient Disk Space

Hi Guys,

 

Any ideas what might be going on here…

 

So I come in this morning and the backup from last night is still running. It is saying not enough disk space on the USB Drive.

 

 

 

I go to the USB Drive and check out the backup sets on the drive. There is a backup set there from the 12/04 which has a expiration date of the 19/04 (It is the 20/04 today in Australia, for those who are reading this elsewhere). So what it looks like has happened is the Backup Set is still protected even though the date said it shouldn't. The only thing I can think of is the system date of the machine is incorrect, but I checked that and it is set correctly.

 

 

 

Is there anything that I can try that you guys can think of?

 

Thanks in Advance

Peter

2012
Backing Up
Backup and Recovery
Backup Exec
  • Colin_Weaver's avatar
    Colin_Weaver
    13 years ago

    Just for info of others and bear in mind this still needs confirmation as this may not be the cause of Peter's problem

     

    DLM will not groom the last complete recovery point chain (i.e. full + Incrementals) for a given resource as such if the set that appeared to have expired was part of the last recovery point chain then it will not be deleted with the result that you may run out of space if this has not been planned into your storage strategy.

    Please see this blog

    https://www-secure.symantec.com/connect/blogs/data-lifecycle-management-be-2012

     

     

     

  • pkh's avatar
    pkh
    Moderator
    13 years ago

    It is not just the date that is important, the exact time is also important.  If your job starts on 19/4 at 10 p.m. and the media is protected until 19/4 11 p.m., then it is still protected when the job runs and cannot be overwritten.  The OPP starts from the END of the job that last write to the media.  Thus if you want to protect your media for a week, you got to set the OPP to 6 days, instead of 1 week.

  • Peter_Sheridan's avatar
    Peter_Sheridan
    Level 6
    13 years ago

    Hi Phk,

    Thanks for that. The expiration date/time on the media set it the 19/04/2012 2:09:17AM (This is for the G: Backup Set as seen in the image above). The current time here is 20/04/2012 3:44:00 PM so it has definatly past that time.

    The only other thing i can think of is that the Backup Set G: is no longer included in the backup job that is currently stalled. Could that be why it is not overwriting...?

  • pkh's avatar
    pkh
    Moderator
    13 years ago

    The question is when your job runs.  Is it before the expiry of the protection period?

    Also, did you set the global default to "overwrite recycleable media before scratch".  This setting is found at BE button ---> Configuration and Settings ----> Storage.

  • Peter_Sheridan's avatar
    Peter_Sheridan
    Level 6
    13 years ago

    The start time of the job was 19/04/2012 @ 10:00 pm, the media was protected until 19/04/2012 @ 2:09:17am.

    So the job started after the media expiration time.

    And no, it is set on the default of Overwrite Scratch Media before overwriting recyclable media..

  • pkh's avatar
    pkh
    Moderator
    13 years ago

    ... then your job is going to use a new media.

  • pkh's avatar
    pkh
    Moderator
    13 years ago

    The setting explains it all.  You are asking BE to use a scratch media before overwriting your expired media.

  • Colin_Weaver's avatar
    Colin_Weaver
    Moderator
    13 years ago

    AFAIK Data Lifecycle Management will not groom/delete a full disk based backup if an incremental backup exists that depends on the full.

     

    As such if your incremental backup has a protection date in the future then the full will not be groomed even though the expirt date has passed - not sure  if this is your issue as you may not be running incremnetal or differential backups.

     

    Also as grooming occurs independent of the actual jobs - has this USB drive been detached from the server until just before your new backup attempted to start??