Corruption in VMFS datastore

[JasonGillis]

HI all,

I've got a VMFS datastore that has a directory that ESX can't seem to delete, even though it tries every 2 minutes. From the contents of the directory, it looks like something related to the directory index has become corrupted. The directory looks like this:

# ls -l 1020

total 64

-rw------- 1 root root 353 Jul 2 2010 current.png

?----

---

? ? ? ? ? ???<?g?fΟ?i???

?----

---

? ? ? ? ? ???????hA+?????????KJ? \w#?s????????????

?----

---

? ? ? ? ? IHDR

I can't delete any of the files with the ?'s. I searched around and couldn't find any solid leads on a file system check tool for VMFS. Some posts mentioned vmkfstools -R, but that option doesn't seem to be listed in my version (ESX 4).

Any thoughts on how I can fix this?

Thanks,

Jason

[wila]

Hello Jason,

There's no fsck for vmfs. There's a 3rd party tool which might for this in the future, but I would not trust it at this moment. See: http://planetvm.net/blog/?p=1592

The best thing to do is to get all your VM's off of this LUN, destroy the entire LUN and recreate it.

Ofcourse the most important thing first is to find out what happened with your vmfs storage as this is highly suspicious...

--
Wil
_____________________________________________________
VI-Toolkit & scripts wiki at http://www.vi-toolkit.com

Contributing author at blog www.planetvm.net

Twitter: @wilva