Error after vSphere update

[pizang]

One of my backups stopped working after upgrading whole infrastructure to vSphere. I received and error:

Retrieving file list
VCB error: Error: Other error encountered: Snapshot creation failed: File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore> An error occurred, cleaning up...

This is a one machine backup. This machine has 2 disk
system disk of size 10 GB - 175 GB of free space
data disk of 500GB - 200MB of free space

Data disk is on VMFS volume that has maximum data size of 512GB (block size of 2MB).I guess this is a problem. But why? Snapshot is made on system disk's VMFS. Is it somehow "added" to base disk and than it exceeds 512 GB limit?

[Vitaliy S.]

Hello Miroslaw,

Are you able to do the snapshot manually using vSphere client? Is it a thin provisioned disk (10 GB and free 175 GB of free space)?

[pizang]

Hello Vitaliy,

No I am not able to do a manual snapshot. vSphere Client shows exactly the same error.
This is thick drive, all my drives are thick.

I thought that free space on VMFS volume where VM resides is enough to make a snapshot. Should I create a new VMFS with larger block?

[Vitaliy S.]

Miroslaw,

Yes, everything points that 2MB block size is not enough to make a snapshot, you can format new VMFS volume with larger block size and then move VMs causing issue there, I believe it'll resolve your issue.

Thanks.

[drbarker]

vSphere changed the limits in a few places... Max datastore (& file?) sizes all shrunk by 512 bytes.

See (under 'storage maximums'):
http://www.vmware.com/pdf/vsphere4/r40/ ... ig_max.pdf
http://www.vmware.com/pdf/vi3_35/esx_3/ ... ig_max.pdf

Apparently, ESX 3.x had the same limitation with the last 512 bytes - it just never bothered inforcing it.

As vsafarov says, to get around it, create a new VMFS store with a 2Mb block size. You won't be able to do an online storage motion, so you'll have to shut down the guest, then move from the old datastore to the new one.

[pizang]

I have created a new LUN the same size as the old one - 512GB. I have created a new VMFS volume with 4MB block size (so it can hold files as big as 1TB). I have powered down my VM and moved disk 2 (data disk) to new VMFS.

To my surprise I received exactly the same error:

File <unspecified filename> is larger than the maximum size
supported by datastore '<unspecified datastore>

Is, by any means, block size of configuration file's partition involved? What I mean is that VM configuration file and VM system disk is stored on partition with block size of 1MB. I have checked and only two VM has problem with snapshot and both of them (and only them) have virtual disks larger than 256GB. As snapshot file is stored on disk where configuration file is located maybe it can cause a problem? Maybe system is checking block size of snapshot partition?

Second idea is lack of space on data disk. There are only 170MB left as I maximized virtual disk size. But it should not be a problem as snapshot file is on different disk. I have another VM with data disk of 200GB, no significant space left and snapshot is working.

[pizang]

I have resolved this. I have moved VM config file to VMFS with 8MB block. Snapshot was created without any error. So ESXi 4 changed the rules. It is checking whether snapshot destinantion VMFS has block size equal or grater than disk that you are snapshotting.

[DocMac100]

I'm trying to backup a guest that is on a datastore with a block size of 1mb, which is the block size of all of my datastores, and I'm getting "Snapshot failed: File is larger than the maximum size supported by the datastore".

I understand that a block size of 1mb limits my datastore size to 256gb, but why would the snapshot process be creating a file larger than 256gb? Could it be that the datastore itself is too small?

[DocMac100]

The failure depends on the size of the virtual disk. All virtual machines having disks with a maximum supported size by VMFS may experience this error.

For example, a virtual machine with disk sized 256 GB stored on VMFS with 1MB block size fails to snapshot because of the additional overhead required for the delta files. Though the maximum file size for a 1MB block size is 256GB - 512, creating a vmfsSparse (delta) disk requires additional overhead space about 2GB. The resulting delta has potential to grow beyond the file system limit.

To resolve this issue, move the virtual machine to a VMFS volume with 2MB block size.

Overhead for the snapshot is roughly about 2GB for a disk size of 256GB. If snapshots are to be used, consider the overhead while deciding the size of the disks.

[Vitaliy S.]

Hello Doc,

Thank you for answering your own question I'm sure it will be helpful for other users also!

[Gostev]

Also here's the existing thread with additional details about this issue:
http://www.veeam.com/forums/viewtopic.p ... +datastore

[lewislampkin]

Hi, I'm new here. But, I am getting the same error message as the original poster.

His problem was something with block size, so I'll share the details about the particular VM and datastores giving an error (others are backing up just fine so far).

All thin provisioned (if that matters)
VM disk1 (System): 80 GB datastore: 183 GB, 1 MB block size
VM disk 2 (Data): 512 GB datastore: 1.82 TB, 8 MB block size
VM disk 3 (Data): 512 GB datastore: same as VM disk 2
VM disk 4 (Data): 512 GB datastore: same as VM disk 2

Target datastore: 1.82 TB, 8 MB block size

OK, I resolved my own issue, before I could complete this post, but including it for completeness ...

The issue is that there is an option to "check space" when you are setting up the job properties. Had I done this in the first place, I would have prevented the error. Irregardless, I went into the job properties, checked the space, and saw this error. The error said I didn't have enough space to do this job, within the retention policy. In this case, I modified the job, and took it from the default of 14, back to 13. (Not a big deal, as it is only a test server, in the first place.)

OK, problem solved, and the job now runs just fine.

Can now use this as a "planning tool" to calculate space required, based on the number of backups I want to retain.

Thanks for making such an easy to use product!

[lewislampkin]

OK, problem not sorted. I'm going to break out this server, and set up a job for each drive, and see if a particular drive fails the job.

EDIT:
Individually, each drive still fails.

Not sure what the problem is ... anything about my setup suspicious?

If it would tell me where it is placing "unspecified filename" and what was the "unspecified datastore" I might know better, but I posted my setup, and not sure what I'm missing.

[Vitaliy S.]

Lewis,

It won't make much difference if you brake the job to backup particular virtual disks separately. The reason why you see the error is that a snapshot is created on the same datastore where main VM configuration files are located (in your case it 1 MB datastore, which doesn't meet the requirements). As you probably know when snapshot is taken it affects all disks attached to the VM and the only way to exclude the disk from snapshotting is to make it independent, thus meaning you won't be able to backup it either. To resolve the issue you need to reformat the datastore (with VM disk1 on) with a larger VMFS block size.

Thank you!

[lewislampkin]

Lewis,

It won't make much difference if you brake the job to backup particular virtual disks separately. The reason why you see the error is that a snapshot is created on the same datastore where main VM configuration files are located (in your case it 1 MB datastore, which doesn't meet the requirements). As you probably know when snapshot is taken it affects all disks attached to the VM and the only way to exclude the disk from snapshotting is to make it independent, thus meaning you won't be able to backup it either. To resolve the issue you need to reformat the datastore (with VM disk1 on) with a larger VMFS block size.

Thank you!

Vitaly,

I did not realize this about the snapshots, and that provides an important warning to me, with regards to planning my storage layout. I am probably at novice level, with regards to this stuff, just another hat to put on at the office, you know?

Thanks for your help, it makes perfect sense now.

[Rohail2004]

[merged into existing discussion]

i am keep getting the following error while trying to run the back up..

Freezing guest operating system
CreateSnapshot failed, vmRef "432", timeout "1800000", snName "VEEAM BACKUP TEMPORARY SNAPSHOT", snDescription "Please do not delete this snapshot. It is being used by Veeam Backup.", memory "False", quiesce "False"
File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore>

[milfner]

[merged into existing discussion]

Before explaining my problem, I will give you a breif overview. I have an SBS 2008 VM on a datastore that has a block size of something small (i think it was 1MB), this was created inadvertently. I now realize that the total size of the OS hard disk will not exceed something like 250GB due to this 1MB cache size limitation on the datastore (called "Second Mirror). I have a feeling that the following error I am getting on Veeam Backup & Replication is due to this???

I tried backing up this VM and I receive the following error.....

0 of 0 files processed

Total VM size: 256.00 GB
Processed size: 0.00 KB
Processing rate: 0 KB/s
Backup mode: HOTADD/NBD without changed block tracking
Start time: 16/02/2011 7:15:31 PM
End time: 16/02/2011 7:15:55 PM
Duration: 0:00:24


Freezing guest operating system
CreateSnapshot failed, vmRef "208", timeout "1800000", snName "VEEAM BACKUP TEMPORARY SNAPSHOT", snDescription "Please do not delete this snapshot. It is being used by Veeam Backup.", memory "False", quiesce "False"
File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore>

[pwynne2009]

[merged into existing discussion]

Hi,

My backup job for a file server is failing for me. below is the mesage I get for the failure. Has anyone seen this before?

"CreateSnapshot failed, vmRef "vm-164", timeout "1800000", snName "VEEAM BACKUP TEMPORARY SNAPSHOT", snDescription "Please do not delete this snapshot. It is being used by Veeam Backup.", memory "False", quiesce "False"
File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore>"

[dilpz213]

[merged into existing discussion]

Hi All,

We have been using Veeam Backup and Replication 5.0.0.179 to backup smaller VMs from our ESXi 4 environments to other Windows servers in our LAN. This weekend we did a P2V of a Windows 2003 fileserver to one of our new ESXi installations. The ESXi host has raided internal disks (no shared storage) divided into 2 VMFS datastores. Datastore 1 has a default block size and is 924GB in size. Datastore 2 has a 4mb block size and and is also 924GB in size as it hosts a large 800GB VMDK for this fileserver.

The fileserver consists of 2 VMDKs. The first one is 220GB in size and is the system volume/app store hosted on datastore 1. Datastore 1 has around 556GB free. The second, as mentioned above is 800GB in size and is on Datastore 2 which has 124GB free on it.

When attempting to do a Veeam backup job to a windows share on another server in network mode, the job fails with the following error:

Freezing guest operating system
CreateSnapshot failed, vmRef "vm-422", timeout "1800000", snName "VEEAM BACKUP TEMPORARY SNAPSHOT", snDescription "Please do not delete this snapshot. It is being used by Veeam Backup.", memory "False", quiesce "False"
File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore>

Is this error because there isnt enough space to make the snapshot or because the snapshot veeam is trying to make is too big for the block size of the datastore? If it is the latter, is there any way of getting round the problem without having to trash the datastore and reformat? Or am I going to struggle anyway as I only have 124GB free left on datastore 2? I have already tried doing the backup job excluding the 220GB VMDK and doing the 800GB VMDK on its own. I still get the same error. Any advice would be much appreciated.

Thanks

[Vitaliy S.]

Dilpz, you need either move VM configuration files to a datastore with larger block size or reformat the existing datastore (storing VMX file) with larger block size.

[dilpz213]

Hi Vitaliy,

Thanks for your input. Ok, we will try to reformat the datastore. As this VM is spread over datastore 1 and 2, will both datastores required formatting to the 8mb block size or just the one with the larger VMDK?

[Alexey D.]

Just make sure that VM's configuration files are either on the datastore with "larger VMDK" or on the datastore with 8 MBs block size.

[dilpz213]

Hi Alexey, I had a hunch it might be something like that. I moved the config files - and it seems to have worked! Thanks. I guess it makes sense as ESX will try to make the snapshot in the same place as where the config files are and in my case, the snapshot had the potential to exceed 1tb and therefore it failed as where the config files were previously, the block size was only 2mb. Am I right?

[Alexey D.]

You're 100% correct!

[ritterbruder]

[merged into existing discussion]

First time poster, 10 days into trial. I know there have been a few threads about this but my question stems more of how can I specifically fix this in my scenario. I am new to virtualization and created a file server vm and threw some files on it. It is thin provisioned and has about 29gb of space used so far (it is all one large vmdk file, not cut into small chunks). When I attempt my first backup this is the error I receive.

Total VM size: 1.00 TB
Processed size: 0.00 KB
Processing rate: 0 KB/s
Backup mode: HOTADD/NBD without changed block tracking
Start time: 4/8/2011 1:36:18 AM
End time: 4/8/2011 1:37:09 AM
Duration: 0:00:51


Freezing guest operating system
CreateSnapshot failed, vmRef "vm-88", timeout "1800000", snName "VEEAM BACKUP TEMPORARY SNAPSHOT", snDescription "Please do not delete this snapshot. It is being used by Veeam Backup.", memory "False", quiesce "False"
File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore>

My target is a 2tb iscsi target mounted onto a server 2008r2 as a partition. From what I've read I think it has something to do with block size. I would like to know how to deal with this the best way possible without totally redoing my vm. Thanks everyone.

[ivan_cecoel]

[merged into existing discussion]

Hi everyone, I have a problem when I try to do a backup from a VM, the error is

File <unspecified filename> is larger than the maximum size supported by datastore '<unspecified datastore>

I found the solution in this forum, format the partition with another block size, but in my case that's a problem.

I have the VM with the OS disk in a Fiber Disks, and I have the other disks into SATA disks (6 disks 600Gb every disk) , I only want to backup the OS disk, the other disks I'll do a tape backup. I select only de OS disk in the veeam backup but the error stills...
I can't format the datastore, there are more servers and this server it's in production and I can't stop them.
There are any possibilities to backup only the OS disk?
I saw another possibilitie are copy de config file of the VM to another datastore with a block size bigger, it's possible? Could be some problem?

Thank you so much.

[Vitaliy S.]

If re-formatting the datastore is not an option for you, then moving VM configuration files to the datastore with a larger block size should make a trick.

I select only de OS disk in the veeam backup but the error stills...

That's expected. Reading back through my responses above, you should notice that when snapshot is created it affects all disks attached to the VM and the only way to exclude the disk from snapshotting is to make it independent, thus meaning you won't be able to backup it either.

[Daveyd]

I have a VM that is configured like this:

C: (O/S) - 30GB vmdk on 400GB Datastore formatted with 1MB block size with 40GB free
D: (Data) - 1.48GB vmdk on 1.5TB Datastore formatted with 8MB block size with 20GB free

I am obviously receiving the same Veeam error when trying to do a backup. It looks as if I need to move the .vmx file from the OS datastore to the Data datastore. My question is, in doing so, how much free space do I need for the snapshot? As you can see I do not have tons of free space on either datastore.

[Gostev]

Depends on how active the VM in question is. You need enough space to accommodate incoming disk changes while backup is running (which can be anywhere between few minutes to few hours depending on amount of incremental changes).

Our product starts throwing warnings at 10GB free space. 20GB or 40GB is better, and should be enough in most cases (unless VM in question is highly active Exchange or SQL server).

Thanks.

[Daveyd]

Thanks for the info. Is the recommended procedure to move the vmx file to power down the VM, unregister it from vCenter, copy the vmx file to the 8MB block datastore the re-register it in vCenter?