Recently I had a problem with an HyperV/Failover cluster I just put up. We have a back end SAN with 2.61 TB on the LUN. I have 2 nodes in my cluster and a 10 GB Quorum disk.
So I get this error on 5 of my VMs.
Log Name: Microsoft-Windows-Hyper-V-VMMS-Admin Source: Microsoft-Windows-Hyper-V-VMMS Date: 9/16/2012 12:02:47 PM Event ID: 16060 Task Category: None Level: Error Keywords: User: SYSTEM Computer: <removed by me> Description:<hostname> has been paused because it has run out of disk space on 'C:\ClusterStorage\Volume1\vm\<hostname>\'.
Now, our CSV storage space is 2.61 TB. See the allocation below:
VM---------Disk Utlzd (GB)--Disk Allocated(GB)--Allocation Type (Thin/Thick/PassT)
vs1-OS-----14---------------127-----------------Thin
vs2-OS-----12---------------127-----------------Thin
vs2-data---535--------------1035----------------Thin
vs3-OS-----22---------------127-----------------Thin
vs4-OS-----9----------------50------------------Thin
vs5-OS-----15---------------127-----------------Thin
vs5-Data---1024-------------1024----------------Thick
Totals-----1631-------------2617
Cluster Shared Volume available = 2.61 TB
Ratio = 1:1
Now I did have a 500 GB backup of VS5 from when the cluster previously crashed. Still, I had 111 GB of free space left on the CSV according to the Failover Cluster Manager.
I guess my question is how much overhead is really required here? I don’t want my VM’s to “pause” because they don’t think they have enough space, even though I do. I don’t know what to do here because I’ve used VMware and Xen and the allocated space is what you have. So I’m kind of confused.
I’ve been reading the forums and one suggestion was that there should be double the capacity that is allocated which seems unnecessary. Another thing that keeps coming up said to delete the snapshots, but I’m not using any snapshots currently. ???
I just saw this when I was about to post my question:
Is this telling me that if I’m using dynamic (I call them thin) disks that I cannot overallocate?