Hi,

My CSVs (iSCSI SAN volumes) are filling up, but I can't see what is taking the space.  I have a server 2012 cluster with 4x 3TB CSVs on an iSCSI SAN.  One of the CSVs has only 250GB free according to the properties of the disk in Cluster manager, however the total amount of disk used by my VMs is 1.8TB, so I have almost a TB of space being used but I can't see what by.

I backup the VMs using DPM 2012 SP1 so I think VSS may be to blame (no VSS hardware providers install, using the Microsoft VSS provider), but running a vssadmin list shadowstorage doesn't return any info about whats in use by VSS

vssadmin list shadowstorage
vssadmin 1.1 - Volume Shadow Copy Service administrative command-line tool
(C) Copyright 2001-2012 Microsoft Corp.

No items found that satisfy the query.

Are there any other things I can do to see what is eating the space and to recover it?

Cheers,

James

Hi,

Thank you for your question.

I am currently looking into this issue and will give you an update as soon as possible.

Hi Lawrence, thanks for looking in to this.

Still no idea what's using the space, but I have found if I change the owner of the CSV the space becomes available again.

Hi, James

> but I have found if I change the owner of the CSV the space becomes available again.

Its wired question, so when you change owner back, what happened? Still only have 250 GB free space?

I think maybe this space was used by CSV I/O synchronization metadata.

A CSV enables multiple nodes to have simultaneous read-write access to the same shared storage. When a node performs disk input/output (I/O) on a CSV, the node communicates directly with the storage, for example, through a storage area network (SAN). However, at any time, a single node (called the coordinator node) owns the physical disk resource that is associated with the LUN. The coordinator node for a CSV is displayed in Failover Cluster Manager as Owner Node under Disks. It also appears in the output of the Get-ClusterSharedVolume Windows PowerShell cmdlet.

However, when certain small changes occur in the file system on the CSV, this metadata needs to be synchronized on each of the physical nodes that access the LUN, not only on the single coordinator node. For example, when a virtual machine on a CSV is started, created, or deleted, or when a virtual machine is migrated, this information needs to be synchronized on each of the physical nodes that access the virtual machine.

Use Cluster Shared Volumes in a Windows Server 2012 Failover Cluster
http://technet.microsoft.com/en-us/library/jj612868.aspx

Hope this helps!

TechNet Subscriber Support

If you are TechNet Subscription user and have any feedback on our support quality, please send your feedback here.

Hi Lawrence, after moving the volume back to the original node it still has 1.1TB free.

Could this have been an issue of that IO Sync metadata space not being released after being written to?  Is there a way I can view how much disk space is in use for this cache?

The article you linked had a powershell command to reserve some space for cache?  Will this also limit the amount of space used as cache?  I'm happy to allocate a decent chunk of space, but I can't have a TB disappearing as happened.

Thanks,

James

Hi,

Wish below Blog can help you:

http://blogs.technet.com/b/askcore/archive/2011/04/13/where-s-my-disk-space.aspx

http://blogs.technet.com/b/tommos/archive/2011/12/03/enumerating-free-disk-space-on-remote-computers-with-powershell.aspx

http://blogs.msdn.com/b/ntdebugging/archive/2008/07/03/ntfs-misreports-free-space.aspx

http://blogs.msdn.com/b/clustering/archive/2010/06/19/10027366.aspx

Regards,

Hi,

How are things going on your end? Please keep me posted on this issue. I appreciate your time and efforts.

Regards,

Hi!

I'm having almost the same problem - CSV free space is disappearing strangely quickly.
I had problems with Netbackup, but I resolved them and just made new full backup of VMs. But after completing that, it didn't free any space on CSV.
I also tried to changing Owner Node of the CSV, but that didn't help either.

Any suggestions?

This was resolved by running DISKSHADOW -> DELETE SHADOWS ALL in an elevated command prompt on the Hyper-V host.

Found it: https://social.technet.microsoft.com/Forums/en-US/44cde22c-e404-485d-8e71-246525204208/dpm-decreases-csv-free-disk-space?forum=dpmhypervbackup