Unable to increase VMFS5 on local storage (using vmkfstools)

[tuemadsen]

Hi.

I'm seeing an annoying issue which apparantly I share with several other people.

I'm trying to increase the capacity of a local VMFS5 datastore as per: KB2002461 (http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=200246...

My host (5.1) is booting from an USB pen drive, and i had a slightly less than 2Tb large LUN on a local SmartArray Controller.formatted with one large (almost) 2Tb VMFS5. I added a disk to my raidset and expandend the LUN on which my 2Tb VMFS5 is located. After that I used partedUtil (partedUtil -getUsableSectors) to get the new ending sector number to which I could expand my partition with the "partedUtil resize" command.

All of this presented no problem and I can now see my partition as a 2.73Tb extend both from the commandline as well as the VI Client.

My problem is I can't complete the last part of the proces which is growing the (almost) 2Tb VMFS5 to the now 2.73Tb space I have available in the partition.

I run the command "vmkfstools --growfs /vmfs/devices/disks/naa.600508b1001052395658363

03752001b:1 /vmfs/devices/disks/naa.600508b100105239565836303752001b:1" and get this in return:

Not Found

Error: No such file or directory

I believe i'm issuing the correct command and judging from how may others that are experiencing this issue, I'm guessing this is a bug? But I could be wrong so now I'm asking you guys.

Listed below are a few commands and their output to document my troubles (and to prove i'm reffering to the correct disk/partition/vmfs in my commands):

partedUtil get /vmfs/devices/disks/naa.600508b100105239565836303752001b

364791 255 63 5860375984

1 2048 5860375950 0 0

partedUtil getptbl /vmfs/devices/disks/naa.600508b100105239565836303752001b

gpt

364791 255 63 5860375984

1 2048 5860375950 AA31E02A400F11DB9590000C2911D1B8 vmfs 0

vmkfstools -P /vmfs/volumes/HDD_R5

VMFS-5.54 file system spanning 1 partitions.

File system label (if any): HDD_R5

Mode: public

Capacity 2198754820096 (2096896 file blocks * 1048576), 624961781760 (596010 blocks) avail

UUID: 4f4382ca-f160bd50-5e77-b499baa49a66

Partitions spanned (on "lvm"):

        naa.600508b100105239565836303752001b:1

Is Native Snapshot Capable: YES

Please help...

[a_p_]

Since you mentioned a "a local SmartArray Controller" I'm afraid you are hit by http://kb.vmware.com/kb/2006942

Until VMware/HP provide a fixed driver, you'll have to stay with logical volumes smaller than ~2TB.

André

[tuemadsen]

I am aware of the 2Tb issue with the cciss driver for Smart Array P400 series of controllers. But my server is equiped with a Smart Array P420 that uses the HPSA driver which does not suffer the mentioned issue. I'm absolutely sure this is not the issue as ESXi correctly detects the LUN and my expanded partition as a 2.73Tb extent.

My only issue is the inability to grow the actual VMFS 5 filesystem inside my 2.73Tb partition.

I have seen several other posts on the net with people suffering the same issue, and have yet to discover a solution. I believe the bug might have something to do with my controller being upgraded from a earlier Smart Array some weeks ago and hence I imported the VMFS volumes choosing the option "do not resignature volume" as I did not want to go through adding all my VM's to inventory again.

I found a post on the web with a guy having what appears to be the same issue, and after he choose to reimport his volumes again but resignaturing them this time, he could grow his filesystem.

I could just evacuate the datastore, remove the VMFS and recreate from new to solve this, but I don't have the extra needed space, nor do i have the time.