Invalid Disk Chain created on virtual machine has "snapshot.alwaysAllowNative" set in NFS datastore with VAAI plugin

Products

• VMware vSphere ESXi

Issue/Introduction

Symptoms:
Virtual Machines stored on NFS datastores and configured with "snapshot.alwaysAllowNative" may encounter Invalid Disk Chain. For more information about "snapshot.alwaysAllowNative", see Enable NAS Native Snapshots on Virtual Machines.

"Invalid Disk Chain" means vmdk description file for some disks in the whole chain contain both parentFileNameHint and ddb.nativeParentHint settings.

After invalid disk chain created, you may encounter below symptoms:

• Some files may be left on storage after VM delete from vCenter.
• Remove all snapshots may result in corrupted VM disk.

Environment

VMware vSphere ESXi 8.0.0
VMware vSphere ESXi 7.0.1
VMware vSphere ESXi 6.7
VMware vSphere ESXi 6.5

Cause

snapshot.alwaysAllowNative = "TRUE" setting would allow the virtual machine in NFS datastore with VAAI plugin to be able to create Native snapshots ignoring its base disk is flat one or not.
Once the VAAI plugin doesn't work any more in the NFS datastore, the subsequent snapshots would become redo-log type,Thus invalid disk chain like: redo-log...->native, native->...->redo-log->...->native chain, even redo-log...->native...->redo-log would be created.

Resolution

Currently there is no resolution.

Workaround:
To workaround this issue, please use full cloning for the virtual machine. For more information, see Clone an Existing Virtual Machine.

For the existing invalid disk chain, using full cloning to fix the disk chain issue; And also remove the snapshot.alwaysAllowNative settings to make sure only redo-log chain was created which won't depend on the VAAI plugin.