Veeam Quick Migration Bug

Veeam Quick Migration: A Lesson Learned

As a seasoned IT professional, I’ve had my fair share of experiences with various virtualization tools and techniques. Recently, I was tasked with moving some VMs for a client using Veeam Quick Migration. While the migration process seemed to complete successfully, the VMs failed to boot upon arrival at the destination. This led me on a troubleshooting journey that uncovered an interesting behavior in Veeam Quick Migration.

The source VMs were part of an Oracle RAC cluster, using shared VMDKs. One of the requirements for setting up shared VMDKs is that they must be thick eager zeroed disks. During the migration, I noticed that the destination side had thick provisioned lazy zeroed disks. This disparity in disk types caused the VMs to fail to boot.

At first, I thought something might have gone wrong during the migration process, so I tried it again. However, after the second run completed, the VMs still wouldn’t boot. That’s when I started digging around for more information. The error message I received was vague and simply stated “Incompatible device backing specified for device ‘0’.”

After verifying the configuration of both nodes, I decided to investigate the disk type on the destination side. That’s when I noticed that the disk type was thick provisioned lazy zeroed. Ah ha! This discovery led me to suspect that this might be a bug in Veeam Quick Migration.

To test my theory, I built some more test VMs and paid close attention to the disk type selection during the migration process. Sure enough, when I explicitly selected thick provisioned eager zeroed or used the same format as on the source side, the result was a VM with lazy zeroed disks. Time to submit a ticket!

Veeam support was very helpful and investigated the issue. A couple of days later, they confirmed that this was indeed a bug that would be fixed in an upcoming version. This bug may seem like a minor inconvenience, but it’s important to note that there is an easy workaround.

If you encounter this issue, you can simply login to an ESXi server using SSH and convert the VMDK using the command vmkfstools –eagerzero /vmfs/volumes/DatastoreName/VMName/VMName.vmdk. More information on how to convert a VMDK can be found in this VMware KB article.

In conclusion, while Veeam Quick Migration is an excellent tool for migrating VMs, it’s important to be aware of potential issues like this one. By understanding the behavior of thick provisioned lazy zeroed disks and the consequences of selecting the wrong disk type during migration, you can avoid potential headaches and ensure a smooth migration process.

As always, I appreciate your time and attention, and I hope that this lesson learned will help you in your own IT endeavors.