I remark something strange with Vmware 6.5.
FAS are the destination of Snapmirror.
We make some tests for activate the PRA and mount the LUN to VMWARE PRA.
We make a break of SnapMirror on the volume for put the volume in writable mode.
We mount the LUN on vmware but it not visible on all ESXi (just the ESXi where we mount the LUN).
In another ESXi we see correctly the LUN in "list of LUN" with naa.600.a09..... but that's all in colum datastore instead to see the name of LUN we see "unconsumed"
I suspect something in vmware but if I create a new fresh LUN on NetApp (without SnapMirror) I have not this problem.
The LUN are mounting on all ESXi.
Thank's for your idea,
See The Solution
I believe this is due to VMware detecting the LUNS as Snapshots. Follow the instructions on this KB and you should be able to resolve the issue.
Thank you for your quick reply.
I remember hearing something about the resignature of LUNs.
I follow your vmware kb and found also a KB NetApp
Visibly we need to pass this command for LUN mounting after the SnapMirorr's break
list orphaned VMFS datastoresesxcli storage vmfs snapshot list
esxcli storage vmfs snapshot list
Resig VMFS datastoreesxcli storage vmfs snapshot resignature --volume-uuid=
esxcli storage vmfs snapshot resignature --volume-uuid=
Do you know after this commands when we do a resync and reput in production snapmirror if you need again launch resignature in Production ESX ?
So I wouldn't resignature the LUN. I would just run the following so that you don't have to resignature when you perform a revserve resync.
# esxcli storage vmfs snapshot mount -l label|-u uuid
View solution in original post
Ah ok I understand.
Just the "simple mount" for can do a giveback after.
Thank you for your explanation I check that.
Just a feedback on this problem.
Customer say me he don't choose "resign" option when he mount the datastore to vmware.
Now it's ok.
Thank you all for your quick help very helpfull !