ESXi Cloned Snapshot through Plugin, Same UUID?

[1parkplace]

I cloned a snapshot using the plugin and everything went fine, but when I went to hot-add the VMDK to the VM, I was presented with a prompt telling me that the UUID of the VMDK is the same and I should not proceed as it will cause boot problems, etc. I restored some files from a cloned VMDK in the past and do not recall seeing this message, but maybe I did and just added it because I was under pressure to restore. This time I am just performing a disaster recovery test and more hesitant to mash buttons

Any insights here?

[mamatadesaiNim]

Are you trying to add a VMDK to a VM that already has a disk with the same UUID?  Can you let me know your exact workflow?  e.g. you are using vcplugin, you snapshot a VMFS datastore?  or a vVol datastore?

https://kb.vmware.com/s/article/1021189

[1parkplace]

Hi there,

My workflow is as follows:
Clone Nimble Snapshot from 1 day ago (new vol name: SQL-BACKUP)
Attempt to Hot-add SCSI device to VM targeting VMDK in SQL-BACKUP

This is where I get the message about the VMDK having the same UUID.

Ultimately I am trying to access the snapshot volume with my SQL server to mount the DB contained within so I can access records from the snapshot period. I feel that I have done this before and not sure if I just accept the warning and proceed as I will only be removing it shortly after adding, or if this will cause system instability.

What is the proper process to access VMDK data from a snapshotted VMFS?

[mamatadesaiNim]

"Clone Nimble Snapshot from 1 day ago" --> so you are attempting this operation from the array GUI, correct?  And it looks like you have a VMFS datastore?

Are you familiar with using our Nimble vCenter Plugin?  If you clone the snapshot through the plugin, the workflow executes the necessary code to resignature the new datastore.

Since you only created the snapshot, you would have to perform the resignature operation as well.  Could you try that please?

--m

[1parkplace]

Sorry I did not make that more clear, I did it through the vCenter plugin and it was re-signatured.

It's not the VMFS that is the problem, it's when I try to attach a VMDK from that re-signatured VMFS to a current VM that I get the notification. I am just trying to figure out if I select YES and proceed to add it, if it will cause any ill-harm to my running production VM.

[mamatadesaiNim]

Thank you for the additional informaiton.  Could you please post a screenshot of the notification?

[1parkplace]

Thanks for sticking with me on this, I can't get a screenshot at this time, but it is the exact message from the VMWare KB you shared earlier:

https://kb.vmware.com/s/article/1021189

"Virtual disks "/vmfs/volumes/<DatastoreName>/<VMName>/<vDiskName.vmdk>" and "/vmfs/volumes/<DatastoreName>/<VMName>/<vDiskName.vmdk>" have the same UUID 60 00 C2 97 55 c1 8d 49-cd 59 4f 55 6e d9 0f 31. Virtual disks with same UUID should not be assigned to a VM since this can lead to problems when the guest boots. Please make sure virtual disks have unique UUIDs. Do you want to continue?"

[1parkplace]

Hi there, any chance for an update on this?

Thanks!

[mamatadesaiNim]

Since you are adding the cloned disk to the original VM, this message is expected, and if this is what you want to do, you can follow the steps in the VMware KB.

There is another option though: since you've cloned the VMFS datastore via the vCenter Plugin, you can browse this datastore, and find the VM and register this with a new name, to access files from the cloned datastore/disk.

Does that help your workflow?