VMFS Corruption (No Partition table... all the VMFS is corrupted) - URGENT Help Please

[Merovingio]

Hi all,

I have a Openfiler iSCSI publishing a 1,85TB LUN. Five Servers (1 ESX 3.5 and 4 ESXi) with 12 VMs on this datastore.

Yesterday morning the iSCSI was connected but there is no data over the LUN. I couldn't browse the LUN and 5 of the VMs was shutdown. Other 4 was working with problems and the last 3 was working OK.

I can't do a migrate becouse I have no access to the LUN. I tried to recover the partition table as documented, creating it with fdisk, changing the type to fb and the begining to 128, but the problem persists.

I tried to resignalling the LUN but the problem persists. As I can see in the VI Client the data is on becouse it recognize the size and the space ocupied and free space.

The VMs that keeps working continue working after several reboots of the VMs and the Storage Server. We have save some data from the working servers, but we have to get data from the server that failed becouse there is no Backup. (I now this is stupid). We are going to try to migrate the working servers using converter this morning.

The ESX log says...

VMKernel LOG

Feb 16 08:04:36 esx35 vmkernel: 0:21:31:43.912 cpu1:1025)StorageMonitor: 196: vmhba32:4:0:0 status = 24/0 0x0 0x0 0x0

VMKWarning

Feb 15 17:31:31 esx35 vmkernel: Invalid totalResources 0 (cluster 0).[type 1] Invalid nextFreeIdx 0 (cluster 0).0:06:58:02.463 cpu0:1035)WARNING: Res3: 1930: FS 4a48fe66-e245beba-8553-00144f20b764 may be damaged. Resource cluster metadata corruption detected

I have been googling and searching the communities the last 24 hours and I can't get any way to get into the LUN to get the VMs or to restore de complete LUN.

Any Idea?

Thanks a lot in advance.

[depping]

CAL SUPPORT,

Don't fiddle around, it will only get worse.

Duncan

VMware Communities User Moderator | VCP | VCDX

-

---

Now available: Paper - vSphere 4.0 Quick Start Guide (via amazon.com) | PDF (via lulu.com)

Blogging: | Twitter:

[Merovingio]

Hi,

I know support is the best option, but I have no support service and the system is on pre-explotation so support is not an option.

Greetings

[Merovingio]

An Update

###> vmkfstools -P -h /vmfs/volumes/ofl1

VMFS-3.31 file system spanning 1 partitions.

File system label (if any): ofl1

Mode: public

Capacity 1.8T, 691G available, file block size 1.0M

UUID: 4a48fe66-e245beba-8553-00144f20b764

Partitions spanned (on "lvm"):

vmhba32:6:0:1

It says the block size is 1.0M but the LUN has 1.8TB. As I know it must be 8MB to reach the 2TB, isn't it?

Is there any way to change this without format the unit?

[kastlr]

Hi,

check the following VMware Knowledge Base article which might help you.

Recovering a lost partition table on a VMFS volume

Hope this helps a bit.

Greetings from Germany. (CET)

[Merovingio]

Hi,

I have done that:

esxcfg-vmhbadevs -q

vmhba32:6:0 /dev/sdg

fdisk -lu /dev/sdg

Disk /dev/sdg: 1999.2 GB, 1999273721856 bytes

255 heads, 63 sectors/track, 243064 cylinders, total 3904831488 sectors

Units = sectors of 1 * 512 = 512 bytes

Device Boot Start End Blocks Id System

/dev/sdg1 128 -390144137 1952411516 fb Unknown

vmkfstools -V

No Response

But Still no luck, the iSCSI is connected but I can't browse the LUN.

Thanks