Article number
000003037
Affected Versions
All
Source Hypervisor
All
Target Hypervisor
All

VM which is a part of a cluster (MSCS or Always On) with RDM or Thick Eager Zero VMDK disks on production site will not load during Failover or Failover test

Viewed 459 times

Root Cause

Multiwrite disks in clustering setups are required to either be set on an RDM disk or a ThickThin Eager Zero disk. However, Zerto will only create a thick or thin VMKD on the recovery site - which means that vSphere will stop the VM from booting on the DR Site.





 

Symptoms

A Failover test VM or Failover VM will not turn on in vSphere. A similar error regarding one or more of the VM disks will appear:

"An error was received from the ESX host while powering on VM XXXX - testing recovery.
Failed to start the virtual machine.
Cannot open the disk '/vmfs/volumes/....../XXXX.vmdk' or one of the snapshot disks it depends on. 
Thin/TBZ/Sparse disks cannot be opened in multiwriter mode"

Solution

This VPG will need to be removed (please allow the "Keep Recovery Disks" option when deleting the VPG -  it is checked by default) and the recovery preseed disks should be changed to a Thick Eager Zero provisioning.
This can be done with an ESXi command line tool named vmfkstools.

The needed command, which runs on the ESXi host is:
vmkfstools --eagerzero /vmfs/volumes/DatastoreName/VMName/VMName.vmdk

This subject and the remediation method (with vmkfstools) is covered in the following VMware KB:
https://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1035823

Once completed, please create the VPG with preseeding.

**NOTE: Even if you uncheck the "Thin" checkbox in the "Storage" tab of the VPG Wizard while creating the VPG- Zerto will create Thick Lazy Zeroed disks, while for us to work with MSCS - we must have Thick Eager Zeroed disks on the recovery site.**