While DRS is on, the machine changes places many times during the day. Sometimes I get the error xxx.vmdk cannot be synced because it is inaccessible. The problem is solving when I manually move the virtual machine to the other host. There is no IO error on the side of VMware.
This is Kalsang from Zerto.
VMware DRS enables balancing computing workloads with available resources in a cluster.
DRS is automatically disabled by Zerto while updating recovered virtual machines in the recovery site from the journal for these recovered virtual machines. After the promotion of the data from the journal to the recovered virtual machine completes, DRS is automatically re-enabled.
If DRS is disabled for the site, VMware removes all resource pools in the site. If the recovery was defined to a resource pool, recovery will be to any one of the hosts in the recovery site with a VRA installed on it.
I may have told you wrong.DRS is only open on the production side. This is completely random. For example, the virtual machine replicates 20 times on the other host in a day. But this problem happens 1 time (I get the error xxx.vmdk cannot be synced because it is inaccessible) .
Is there any 3rd party backups running in your environment by any chance?
Yes there is.But backup does not work when this error is received.
As I’ve seen in another environment where they have backups running and after the backup completes and re-attach the drives back and since the UUID gets changed after the detaching the drive. The ZVM doesn’t see the old UUID and that’s causing the error xxx.vmdk cannot be synced because it is inaccessible.