Video Screencast Help
Symantec Appoints Michael A. Brown CEO. Learn more.

NTFS event Error during backup BE2012.

Created: 23 Oct 2013 | 12 comments
Dennis Ib's picture

Hi there.

I'm running Backup Exec 2012 fully patched, primary backup of VM Ware virtual machines.

Last week we suddenly see this error ind the EventViewer on the backup server and backup started to fail and some jobs startet to take much longer time than normal. I have triede to find a solution, but no luck.

The error from the eventviewer

The file system structure on the disk is corrupt and unusable. Please run the chkdsk utility on the volume .

We have run chkdsk [driveletter] /f on the backup server, but no error reported.

As you can see there is no indication in the error, that shows us which drive its regarding.

Hope someone can ghelp me.

Best regards

Dennis

 

 

Operating Systems:

Comments 12 CommentsJump to latest comment

pkh's picture

You should run chkdsk on the volume that you are backing up, not the media server.

Avkash K's picture

Hi,

Can you please confirm the event id getting logged..

If it is 55 then refer below article.

http://www.symantec.com/docs/TECH136896

Cause:
Event ID 55 may be logged in the System log when you create many files on an NTFS partition on a Windows Server 2003.
Event Type: Error
Event Source: Ntfs
Event Category: (2)
Event ID: 55
Description:

The file system structure on the disk is corrupt and unusable. Please run the chkdsk utility on the volume B:.

 

Resolution:
A hotfix is available from Microsoft to resolve the issue.

 http://support.microsoft.com/kb/932578
Event ID 55 may be logged in the System log when you create many files on an NTFS partition on a Windows Server 2003-based or Windows XP-based computer

Regards,

Avkash K

Dennis Ib's picture

Yes it is event id 55 but i have found out that it is when backup of some windows 2012 virtual servers.

Run chkdks on one of them and no problems there.

Dennis

Avkash K's picture

plz confirm, if above mentioned hotfix from microsoft is installed or not??

If not, then try installing the same.

Regards,

Avkash K

Dennis Ib's picture

it is not because it´s not for Windows 2012 server.

Dennis

Avkash K's picture

are you running BE 2012 on Windows server 2012??

Regards,

Avkash K

Dennis Ib's picture

BE server Windows 2008R2 and BE 2012 fully patched.

VJware's picture

Does the event 55 generate each time when the backup fails ? (Though its a bit uncommon that the event does not specify which volume). Are there any corresponding errors under the Application events when this system event occurs ?

What is the Backup Exec error which you receive from the job log ?

Lastly, are you backing up to disk or tape ? If disk, has chkdsk been run on this storage as well ?

Dennis Ib's picture

Yes each time, but there is no error in the job. No corresponding errors under application events.

The backup is to a deduplication store and chkdks have been run on this drive.

I have installed this hotfix http://support.microsoft.com/kb/2754784 now i'm trying to run backuop again.

I will report back with the result.

Dennis

Dennis Ib's picture

Hi

The hotfix didn't help, the Event 55 is still in the event viewer. But backup run fine and done some restore test.

Dennis

Peter Faber's picture

any updates on this one?

i have the same problem when backing up VMs over SAN.

running 2012 SP3 on all media servers and caso.

Colin Weaver's picture

If this only affactes Windows 2012 virtual servers, as we mount the vmdk files in order to catalog the file system GRT, it's possible that this error is being generated because the operating system of the media server briefly sees the disk system that is held inside Windows 2012 as if it is a local disk.

 

If it only does the error for some but not all Windows 2012 VM's then try looking at differences in the configurations of the vmdk files and the windows volumes held within the vmdk files between VMs that appear to generate the error and those that don't, it's possible you may find somethingn that is not supported yet.

 

BTW I'd ecnorarge formal support cases to be logged for this issue just for us to confirm what is happening as my comments above are a theory / educated guess.