Hi all,
After re-balancing a linked clone persistent pool, some (~10%) of vm UDD's are sitting at the root of the datastore rather than within the vm folder where they should be. The pool has about 90 Windows7 desktops split across 6 250GB datastores with a moderate over commit level. I don't think this has anything to do with the re-balance operation because I've just checked a test pool of 2 linked clone persistent desktops, sat in one single datastore and both the UDDs are outside of the vm folders too - this pool has never been re balanced, just refreshed and recomposed. Looking at one of the affected vmx files, you can clearly see that the UDD is referenced outside of the vm folder e.g.
scsi1:1.fileName = "/vmfs/volumes/533a27df-25bb3d4a-a487-bc305be1a54f/vmname-vdm-user-disk-D-e2a3a1e4-eb7e-4c4f-bc52-12e8d00ee926.vmdk"
A re-balance operation on the individual vm's resolves this as long as the vm is moved to a different datastore, but I would like to get to the bottom of this as this is not the first time I've seen this within our environment.
Anyone else seen similar behaviour/knows how to resolve?
Thanks in advance
View 5.2
Vcentre 5.1.0
Comoser 5.2
ESXi 5.1.0