Symptoms:
In vpxa.log we will see below error with CTK during the storage vmotion from a snapshot lun to a production lun
Clone: 100% done.DISKLIB-CTK : ChangeTrackerCloneFD: invalid source CTK file descriptor.
DISKLIB-VMFS : "/vmfs/volumes/5be3231c-43fbc90e-3941-ecebb8975bbc/test/testclone1-flat.vmdk" : closed.
DISKLIB-VMFS : "/vmfs/volumes/5be3231c-43fbc90e-3941-ecebb8975bbc/test/testclone1-flat.vmdk" : open successful (1115137) size = 0, hd = 0. Type 3
DISKLIB-VMFS : "/vmfs/volumes/5be3231c-43fbc90e-3941-ecebb8975bbc/test/testclone1-flat.vmdk" : closed.
DISKLIB-LIB_CLONE : Failed to clone: Invalid change tracker error code (5948).
Failed to clone disk: Invalid change tracker error code (5948).
#### PRocess failed with "DISKLIB-LINK : DiskLinkNativeVmfsCloneExisting: Failed to create native clone. (24)."
VMware vSphere ESXi 6.7
the issue was a code change from 6.5 to 6.7 to no longer allow unclean CTK files to be migrated or cloned. Since this was a code change they determined it was not a bug in 6.7 but a change in behavior.
it was not a bug in 6.7 but a change in behavior
Wordaround :
Workaround:
instant recovery VMs created from storage snapshots will need to be powered on prior to them being able to be migrated using storage vMotion.