VMs "unknown inaccessible" and can't browse datastore after normal reboot

[cospence]

Hello VMWare Community,

I know the thread I am posting has been a popular topic however after reading many of your posts I am was unable to determine what I needed to do in order to fix my exact situation. This is a production enviroment server for a medical clinic so we are REALLY feeling the pain of this problem, any help you can provide will be GREATLY appreciated.

My Hardware:

Tyan Tank GT20 1U Rackmount Server

Two physical Intel Xeon 2.0GHz Quad Core E5335 CPUs

Two datastores: One is a single SATA drive. The other datastore is a RAID5 array managed by a 3Ware 9650SE SATA controller

What Happened:

Our server demands have grown to where I needed to gracefully power the server off and install a second physical CPU. Prior to the upgrade we were only using one socket. After finally getting the server BIOS to recognize both physical processors, I booted the server up all the way and was ready to power on my VMs = job finished. (Or so I thought...)

The host had no problem booting back and up coming back online. The hypervisior (which is ESXi 4.0.0) automatically detected the second CPU and no problems were found. I then tried to turn on my VM's and all VM's that are on the RAID controller datastore ("datastore2") are displayed as "unknown inaccessible", however the vm on the single SATA hardrive (not on the raid card - "datastore3") is showing up perfectly fine.

I thought maybe I bumped the 3Ware card and it was not seated correctly, however VSphere sees the adapter with no problem, it just can't browse my datastore. I then rebooted the server and accessed the 3Ware bios to make sure all the drives in the array are fine - sure enough, I have no errors and all drives are reporting healthy.

Can ANYONE help me figure out how to get this datastore back online?

I apppreciate any guidance you can provide,

Chris

[rebootuser]

Hi Chris, have you tried rescanning for VMFS datastores?

My Blog: http://rebootuser.com

If you found this or any other post helpful please consider the use of the Helpful/Correct buttons to award points.

[cospence]

When I rescan it correctly finds my singular HD and my 3Ware controller. The only datastore present after rescan is the one that resides on the single HD. I can click on the "add storage" button and it sees my 3Ware controller as an available device, and it even correctly sees my RAID5 array size (1.83 TB), however if I click on "next" to add this storage device I get the warning:

"All available partitions will be used. This configuration will delete the current disk layout.

All file systems and data will be permanently lost."

Needless to say I then hit the big CANCEL button. 😃

It is as if the system wants to reformat the RAID5 array like I had just installed it. I just wish it would let me add the controller back and use the data that is already on there.

[rebootuser]

These links may be useful as sounds similar to your issue:

http://vmetc.com/2007/11/08/the-current-disk-layout-will-be-destroyed-all-file-systems-and-data-will...

http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&docType=kc&externalId=6482648&sl...

Regards,

Owen

My Blog: http://rebootuser.com

If you found this or any other post helpful please consider the use of the Helpful/Correct buttons to award points.

[cospence]

Just so future readers know, I found my problem.  I had installed a second CPU and while this second CPU was the same model, type, and speed, it was a different sSpec number.  I didn't even know what a sSpec number was prior to this problem.  Nonetheless, I could repeat this problem every time I moved the mismatched CPUs around (between primary and secondary motherboard slots).  The only soultion was to either:

a) Run the server with one physical CPU

b) Purchase another CPU that had the same sSpec as the one currently installed.

I went for Option B and everything has worked fine since.

Hope this helps somebody else in the future.