Hi,
We're running VMware ESXi 4.1. All our datastores are on various filesystems that are connected to our EMC Celerra SAN via iSCSI.
We'd like to delete one of our datastores and reclaim the space it's using in our storage pool.
I did some research and was planning on just deleting the datastore, unpresenting / removing the LUN, and then deleting the filesystem, but then I came across this article:
All Paths Down definitely does not sound like a fun thing. Is this the procedure we need to follow?
Assuming I'm following the directions listed in the article correctly, our next step is to mask the LUN.
For the filesystem I want to delete, I show for esxcfg-scsidevs --vmfs:
naa.6006048cb8eda001c5baaaac35aa0679:1 /vmfs/devices/disks/naa.6006048cb8eda001c5baaaac35aa0679:1 4b426a01-292fbaf0-a8d3-00237d6291d6 0 Email Xtender
For esxcfg-mpath -L | grep -i<naaid>I get:
vmhba37:C0:T0:L3 state:active naa.6006048cb8eda001c5baaaac35aa0679 vmhba37 0 0 3 NMP active san iqn.1998-01.com.vmware:Avanti26-48233d72 00023d000001,iqn.1992-05.com.emc:apm000925033680000-3,t,1
So according to this data, the datastore was on vmhba37, Contoller 0, Target 0, LUN 3. Is that correct?
So if I want to mask the LUN containing this datastore, I believe I need to perform:
esxcli corestorage claimrule add --rule 192 -t location -A vmhba37 -L 3 -P MASK_PATH
Is that also correct?
Thanks in advance,
Erik