I've discovered some inconsistencies in a vSphere environment (ESX 4.0 hosts) I'm looking to help migrate from.
There are a number of RDMs configured where 3 out of 4 hosts have one LUN ID assigned, and the fourth has a different LUN ID.
The result of this is an error when trying to vMotion an affected VM we get an error along the lines of "Virtual Disk 'Hard disk x' is a mapped direct-access LUN that is not accessible".
I've seen a few articles (e.g. 1029786) that refer to removing disks etc. but that article is a little more confusing to me than I'd like it to be when potentially "destroying the mapping file" so I want to ensure I get things right!
As per that article, I've verified that the 5th & 6th characters of the VML file aren't consistent but this is due to the storage having been originally provisioned with an incorrect LUN ID.
(Storage is provided via iSCSI from physical FalconStor NSS appliance cluster.)
I'm hoping the following should work to correct this issue for each affected VM:
Should it be this simple or am I missing something critical?
Note: Discussion successfully moved from VMware ESX™ 4 to VMware vSphere™ Storage
>>There are a number of RDMs configured where 3 out of 4 hosts have one LUN ID assigned, and the fourth has a different LUN ID.
Check with your storage guy or who allocated the disk from the array to make sure all the hosts groups are the same.
Guess who has become "the storage guy" ?
I've already been all through them and found a few inconsistencies which will need resolving but these two are the first ones on the list as I need to get at the host for maintenance (therefore need to get these VMs off it temporarily).