Video Screencast Help

snapshot errors on information store and systesm?state

Created: 17 Sep 2013 | 17 comments

I am new to the company I am working for and have been going through my new clients list of backups.

This is the second one I have found with similar but of course not the same VSS snapshot errors.  

marcusolini helped me last time so hope will see this one also.

I used some of the research I got on the last one with diskpart etc. but all looks good there.

Adding a pic of the error I am getting.   Along with I have a lot more writers failing.

This like last time is the server that has the BE software on it.

It has a decent history till recently.

I got the company to buy new external disks so we have more room for the backups.



Operating Systems:

Comments 17 CommentsJump to latest comment

Donald Eady's picture


      Backup exec is alerting on writer failures, You did well to run the vssadmin list writers command and as you see its results detailed a number of failed writers. Have you restarted the server yet. Restarting the server will reset the writers and restore them to a state 1 stable state. Then reattempt the backup job and report the results. 

Also take a look at the following article


I hope this posting was helpful


mli7777's picture

Yes actually this happened last week also and the server was rebooted over the weekend.

While the backup was running I looked at the writers and they were ok.

Kind of like I have seen one in the past look all good until it goes to the verify pass.


Donald Eady's picture

if you would please run vssadmin lis providers and confirm as to whether or not MS is the only provider... 

See .2 under the solution of the documenation provided above

I hope this posting was helpful


mli7777's picture

Yes MS is the only provider.

I read that article and not sure what they mean by the changes in the exchange properties ? Pic included

But I can change that setting in the AOFO part from Auto to system and use microsoft shadow.

Both these pics are how those to places are setup right now.



AOFOSettings.PNG exchangesettings.PNG
Rahul Kumar1's picture

In the Snapshot Technology where you have selected Use Microsoft VSS, click on the drop down and select System: Use Microsoft VSS writers.

If Exchange is in cluster then in the other window where it says let backup exec choose the best copy to backup from, click on the drop down ans select backup from the active copy.


marcusolini's picture

Hi Mike.  If you are still in need of assistance, I can certainly help with this issue as well.  When able, please collect the diagnostic information requested in the previous post for this system ‘SBS2K8.sss.local’.  I have provided a command line script file that can be used to automatically collect all of the diagnostic information into a file to make the collection more convenient.  Note, this command line script includes addition diagnostic information from the native Windows command line tool FSINFO.  Thank you.

(1) Copy the compressed file “” into a local directory.
Note, right-click the compressed file and choose “Save target as…” (or equivalent).

(2) Extract the compressed file “” into the same local directory.

(3) Execute the command file “vss-diag.cmd”.
Note, the command file utilizes the text file “diskpart-script.txt” for scripting of the DISKPART commands.

(4) Collect the resulting text file “vss-diag-output.txt” and attach the file to this post.

AttachmentSize 863 bytes
mli7777's picture

Here is the info you requested.

I plan to do the changes Rahul suggests and get the server rebooted tonight.

If you come up with anything else before I get them to reboot look foward to hearing back from you.

Thanks for the help you guys.


vss-diag-output.txt 17.27 KB
marcusolini's picture

Hi Mike.  Thank you for collecting the requested diagnostic data so quickly!  The root cause of your issues is that the J: volume is on a disk that has native 4 Kilobyte sectors (4Kn).  This is reported by command “FSUTIL FSINFO NTFSINFO J:” with the output “Bytes Per Sector : 4096”.  4Kn disks are not supported on Windows 2008.  4Kn disks are only supported on Windows 2012 and newer operating systems.  For Windows 2008, 512 native and 512 emulated disks are supported.  Please review the following article for further detail:

With that said, there are a couple of choices on how to resolve your issues:  (A). Remove the unsupported 4Kn disk (and replace with a 512n or 512e disk).  (B). Configure the Shadow Storage associations (and do not include the 4Kn disk specifically within a backup job).  Below are the steps to configure the Shadow Storage association.  When able,  please perform either option (A) or (B) and rerun the failing backup jobs.  Let me know if you have any questions or concerns.  Hope this proves helpful.  Thank you.


Note, From the diagnostic output this setting is already present.

Note, NO reboot is required.
Note, Consider deleting the existing snapshot of the J: volume.

mli7777's picture

Well that would make sense I guess ?   That drive is the new backup external drive so it is from a newer technology ?

So if I do A or B that takes that drive out of the loops ?

Not sure I can do that?   Or does plan B use it by over riding something?

Like I said that is a brand new 3TB external drive the got to replace ones that were to small.



marcusolini's picture

Hi Mike.  I understand that you are in a difficult situation and that you have a tough decision to made since 4Kn disks are not supported in your current configuration.  With the assumption that the disk is exclusively used as Backup Exec backup disk storage…

For Option A, this WOULD require the 4Kn disk to be removed.  Would it be possible to replace the 4kn disk with multiple 512n or 512e disks (as a pool or in rotation)?  Overall, this would be most beneficial.  Keep in mind that 512n and 512e disks are typically a maximum of ~2TB.

For Option B, this would NOT require the 4Kn disk to be removed but it’s consistent success would be unpredictable.  This option additionally assumes that the 4Kn disk is not backed up by a Backup Exec Media Server (itself of another).

Hope this helps clarify.  Let me know what you decide and the outcome.  Good luck.

mli7777's picture

Well I think we are getting closer ?   :)

This system backs up two servers.   So far one has succeeded everytime since I added these 3TB drives.  It is the server that does not have backup exec. on it.

The one that keeps failing is the server with backup exec. on it and is running exchange.

So if I understand your explanation right niether option can be used as the server that is failing is the backup exec. server ??  Maybe I misread that as I read it again ?

I can start off and send a email right at 4 before I leave and see if they can take those drives back.   :(

They have not had them for over 30 days.

I have tried once to put SP2 on backup exec. and it did not fly the first time.   So was going to try it again later will that help at all ?

Thanks for the help.


marcusolini's picture

I guess I need to understand your configuration better.   There are 3 servers.  2 servers are BE Media Servers; one of which has Exchange installed.  Which servers have the 4Kn disks and how many (please use the FSUTIL command to confirm)?  Also, are there any other applications installed?  Thank you.

mli7777's picture

If we have to switch these out is there a 3TB one that is on a compatible list ??



mli7777's picture

Well this is a small company with two servers.  So like many  there is a few things running on each.

Both are small business server 2008 with SP2.

One of the servers has backup exec. 2012 SP1.   I am new to this client but under all programs that server shows Exchange server 2007 and SQL 2005.   I know Exchange is working not sure about SQL.   It also runs peach tree and a retail program called MOM.

This server has the new drivers we backing up to and they are a external USB drive.   That is that J: drive you say we are having issues with.

But since we swapped a week or so ago the server without backup exec. on it has succeeded everytime.   The server with backup exec. on it has not.   Now just before I got these drives we had smaller 700 gig ones and I did get a backup of the server that is failing on those so you maybe on to something ?

Is that enough new information ?

I am still going to start to warm them up to the idea we may need to take them back.



mli7777's picture

And what would cause what looks like it combinging full backups ?

I have about 350 gig to backup and this full is at 1TB and still climbing.


marcusolini's picture

Hi.  Unfortunately, I do not know the compatibility details for specific disk sizes.  In the interest of acquiring successful backups, I would recommend configuring the Shadow Storage associations on the servers, as described previously, and similarly for the other Media Server.  If you capture the same VSS diagnostics for the other Media Server, then I can give you the exact configuration settings.

As for the unexpected backup size, the possible reason is that the Media Server’s disk storage backup device is included in the backup.  When able, another question… are the Media Servers backing up themselves: MS1 is backing up MS1 -AND- MS2 is backing up MS2?  Or are the Media Servers backing up each other: MS1 is backing up MS2 -AND- MS2 is backing up MS1?   The latter would be recommended.  Thank you.

mli7777's picture

Well some things learned this weekend.   They were understandably getting concerned as the one server had not been backed up for a week or so.   So we went back to the smaller external USB drives and it backed up successfully.   So the problem with the newer technology of the bigger drive appears to be the issue.   I think that is even why it was backing up what looked to be 3 times the size of the servers storage and still climbing as I canceled that backup.   As all we did was put the old drive back on and point to it and it succeeded.

So moving foward for right now I am going to monitor the backups to see how full those drives get.   We do have a cloud backup offering so the client is interested in that.   But it is a small company so I am going to guess it will be to expensive but we sill see.

As for the media server backup relationship.   We have two servers.   Both are getting backed up by one server with BE 2012 on it.   For what ever reason I don't understand is that with the newer drive technology of the bigger USB drives the server that does not have BE on it and is being pulled to the USB drive on the backup server was saying successful for its backups.   It was when backing up the media server itself on to the newer USB drive connected to it that kept failing.   And appeared to think it needed to get the last backup along with the new one over agian ??  :)    If that made sense?

So for now if they do not like the price of the cloud offering I will try to get them to purchse say a 1.5TB drive and hope it is the 512 type storage technology.   I talked with the guy at Best Buy where they are going to buy it and he says the manufaturers do not put anything on the box saying which one it is setup for.

Sound like a plan ??

Thanks again for the help.