rasmith1
Contributor
Contributor

Corrupted Files after snapshot deletion - recovery

I was attempting to increase disk size on the primary disk of a VM. I followed instructions to delete all snapshots. After I noticed that the disk type went from Thin provisioned to Thick provisioned. After increasing the disk size in the VM properties I powered on the machine and received multiple errors:

An error was received from the ESX host while powering on VM <NAME>.

Failed to start the virtual machine.

Cannot open the disk '/vmfs/volumes/NUMBER/NAME.vmdk' or one of the snapshot disks it depends on.

The system cannot find the file specified

VMware ESX cannot find the virtual disk "/vmfs/volumes/NUMBER/NAME-000003.vmdk". Verify the path is valid and try again.

Module DiskEarly power on failed.

Cannot open the disk '/vmfs/volumes/NUMBER/NAME_4-000004.vmdk' or one of the snapshot disks it depends on.

The system cannot find the file specified

VMware ESX cannot find the virtual disk "/vmfs/volumes/NUMBER/NAME.vmdk". Verify the path is valid and try again.

I have attempted to follow the instructions to recreate the vmdk file, but have had no success. Is there any other option to get this machine back?

0 Kudos
8 Replies
RAJ_RAJ
Expert
Expert

Hi,

1. please follow the steps create vmdk

vSphere Advanced Technics: Recreating a missing virtual machine disk descriptor file

2. If you have backup then you can restore

3. create a clone of vm  , open vmx file  and make modification (vmdk file names as per vms configuration)  and try to power on

RAJESH RADHAKRISHNAN VCA -DCV/WM/Cloud,VCP 5 - DCV/DT/CLOUD, ,VCP6-DCV, EMCISA,EMCSA,MCTS,MCPS,BCFA https://ae.linkedin.com/in/rajesh-radhakrishnan-76269335 Mark my post as "helpful" or "correct" if I've helped resolve or answered your query!
0 Kudos
a_p_
Leadership
Leadership

First of all, it's important to get an overview of the current state.

For this, please provide a complete list of files in the VM's folder on the datastore, i.e. run

ls -lisa > filelist.txt

from the command line in the VM's folder, then download the newly created filelist.txt, and attach it to a reply post. Also attach the VM's configuration (.vmx file).

Other questions:

Which ESXi version do you use?

Which vSphere Client do you use? (in case of the Embedded Host Client, please provide its version and build - "Help" -> "About")

André

0 Kudos
msripada
Virtuoso
Virtuoso

AFAIK you cannot increase space on the vmdk when you have snapshots. So am not clear on how you got the error which indicates delta files which are part of snapshots post increasing the size of the vmdk... Did you take another snapshot post increasing the disk space. How many snapshots are there currently..

Did you find all the linking disk descriptor files when you take a ssh session on the esxi host?

Thanks,

MS

0 Kudos
continuum
Immortal
Immortal

> AFAIK you cannot increase space on the vmdk when you have snapshots.
Wrong - of course you can do that. Many folks have done that before because they do not know how to find out if their VM has snapshots.
The mistake is to trust Snapshotmanager.

Do you need support with a recovery problem ? - send a message via skype "sanbarrow"
0 Kudos
msripada
Virtuoso
Virtuoso

Though the snapshot manager identifies or not identifies, the snapshot database is having the information and will try to grab info from that.. Incase of failure of consolidation still. I don't believe u will be allowed to change the vmdk space.. It might happen when there is corruption with snapshots database which I have not seen in regular but it may happen..

0 Kudos
continuum
Immortal
Immortal

FYI the snapshot database is a very fragile textfile with the extension *.vmsd.
With a little bit of practice you will realize that the vmsd gets corrupted quite often - especially if tools like Veeam are used.

Do you need support with a recovery problem ? - send a message via skype "sanbarrow"
0 Kudos
Jitu211003
Hot Shot
Hot Shot

Hi,

From the errors you could do the below things.

1). Look for snapshot manager if there any snapshot present. If not, create a test sanpshot and delete it again (This is just a trick).

2). If above steps fails, power down the VM and run offline consollication if online consollidation fails.

3) If still does not solve your issue, then you will have to manually map the original disk to the VM.

Remove the delta files from VM edit settings (vmname-00001, 0002 and so on) and look for original vmdk (i.e vmname.vmdk, vmname1.vmdk and so on).

4). Power on the vm.

Note: A small data loss would be there since the snapshot was corruped. If it is not so financial or trading server, no one gonna know what happened and you VM will start working perfectly.

Thanks

If it solves your query, please mark it resolved.

My own blog is www.vmwarediary.com

Visit and get more answeres,article, tips, trics, ebook and many more.

0 Kudos
rasmith1
Contributor
Contributor

Support case opened. I think this machine is not recoverable. A vmkfstools --fix check NAME.vmdk command run on the 120 GB disk file now results in a "The file specified is not a virtual disk. (15)."

Multiple attempts were made to recreate the vmdk file using the referenced article.

Snapshots were consolidated using the steps here: KB 1004047 (except for the backup ;-( )

No other backups besides snapshots. Live and learn.

I will no longer trust snapshots for any purpose. If the work, good. If they don't I will have some other more robust alternative.

0 Kudos