ntfs disk errors being reported every time Avamar backup is running

Question

Hello Everyone,

I have a strange error to which I haven't been able to find an exact answer and hope someone here might point me in the right direction.

We have a 3 node Fileserver cluster (Windows 2003 sp2). Backups for this cluster was being done through Networker before (Networker 7.6 SP1) we never had any issues with that.

I recently moved these servers to Avamar (6.1.0-402). Ever since Avamar backup was initiated on this cluster I am seeing alot of "ntfs disk errors" (please see attachment).

Avamar activity log shows multiple entries of "Unable to open directory (code 1392: The file or directory is corrupted and unreadable)" and doesn't backup the file.

"vss:*=off" parameter was set with Networker which is not the case with Avamar. Does that mean that this is a VSS related issue?

Has anyone seen this issue? Is there a way to not use VSS when doing backups through Avamar?

Please let me know if anymore details are needed.

Thanks,

Sri

Bebo2k · Accepted Answer

Hi Sri,

Most probably this error appears in the event viewer due to corruption in the file system on that volume, and most of the fixes are related to Microsoft. For that I recommend you to follow firstly the steps mentioned in the following Microsoft Technet Article:

http://www.microsoft.com/technet/support/ee/transform.aspx?ProdName=Windows+Operating+System&ProdVer=5.2.3790.1830&EvtID=55&EvtSrc=ntfs&LCID=1033

If there is no issues with the chkdsk and chkntfs commands, so you will have to patch the machine with the hotfix i provided before. In my case this works fine after patching the hotfix and reboot the machine. But it is better to follow the steos for chkdsk and chkntfs firstly then update the machine.

Hope this helps you,

Ahmed Bahaa

Sandeep_Sinha · Answer

Hi Sri, Please refer to this article : http://solutions.emc.com/EMCSolutionView.asp?id=esg108530&usertype=C

Sridhar246 · Answer

Hi Sandeep,I have tried chkdsk /f and it doesn't do anything. It reports a lot of 'file is unreadable' kind of messages but that's about it.In the meantime, I tried --freezemethod=none and there were no ntfs errors reported last night. At this point it looks to me like a VSS issue as similar issue has been reported by other backup products.I would like to keep this discussion open to see if anyone else had this issue and if they were able to resolve it (obviously not using VSS is not a resolution ).

ionthegeek · Answer

This sounds very similar to an issue with corrupted shadow copies that is documented on technet. I would recommend working with the OS vendor to troubleshoot this issue. The backup is exposing the issue but it is unlikely to be the root cause.

Bebo2k · Answer

Hi Sri,

I saw this issue before, It was Microsoft issue and has been solved by applying this hotfix:

http://support.microsoft.com/kb/932578

Reboot is required after applying that hotfix.

Hope this helps you,

Ahmed Bahaa

Sridhar246 · Answer

I will certainly bring this issue with the server team, but looking at the KB article it doesn't look like it addresses specifically to the issue we are having.

J_H_ · Answer

Just want to clarify on one point. you said you converted it to Avamar, as this is a cluster, you did do the special cluster setup for the backups correct?

Sridhar246 · Answer

Yes sir, I did setup cluster configuration

ionthegeek · Answer

I'm glad you were able to resolve it. Thanks for following up!

Sridhar246 · Answer

Just a quick note, I was able to convince server team to install that hotfix and have been running VSS based backups for 3 days now and not a single disk error is reported.

Thank you everyone for your input

Avamar

ntfs disk errors being reported every time Avamar backup is running

Was this post helpful?