VMware Cloud Community
dm73
Contributor
Contributor

Please help.

Hi Community

I hope someone will be able to give me a bit of advice or help.

I would even consider paying for someone to assist.

We have 2x hp dl380g8 servers (esx 5.5 update 2 installed on sdcard) and a SAS attached netapp 2724 with 12x 1.2tb drives split into 2x 6tb LUNs. The installers of our enviroment initialized one LUN

Which is where our datastore is... or was. Someone then instead of formatting the second LUN with NTFS formatted the VMFS LUN. There is too much speculation to know exactly who or why.

To cut a very long story short I have asked Kroll to investigate but there recovery fees are just way out of any budget I will ever get. Have tried getdataback NTFS which manages to see the vmdk files as volumes but the files are all muddled and in the wrong folders and when extracted are unreadable.

So I am now scanning the drive with diskinternals vmfs recovery. it picks up only 24 folders and 324 files which is promising as that would be roughly the amount of folders and data I remember having on the datastore. I calculate the scan will take another 12 hours, is this worth the wait and possibly the $700 price of the license (that's more manageable than the +-$20000 kroll wants and I could never afford.)

I need to have this critical data back before Monday and I have exhausted all my options except asking the community for assistance.

Anyone please if you have any ideas or could help me out of this pickle I would be very gratefull

Thanks in advance.

0 Kudos
6 Replies
a_p_
Leadership
Leadership

Welcome to the Community,

you may try to contact continuum and ask him for advice and/or help. He's a data recovery specialist and might be able to help.

André

0 Kudos
podrush
Contributor
Contributor

You have my sympathy, nasty situation.

the diskinternals tools look pretty good, and given that it appears to have found roughly the amount of data you think you have then this sounds like a promising option let it finish the scan and see what the results are.

If it was my situation this is what I would do. I assume you have the ablility to 'restore' this data onto another vmfs partition in order to then test that it has retained its integrity post recovery? Not to be negative but often recovery appears successful until you try and run an application against the data or database only to find that its unusable.

I guess the other question is what other write activity there has been on the disk subsequent to it being reformatted/remade

if the datastore is a single partition, and has only been formatted by another application or host then effectively rebuilding it should** bring the partition data and the datastore itself back.

I am sure you have seen this but this like my be helpful. Recovering a lost partition table with a VMFS datastore - vInfrastructure Blog

good luck - if you can please let us know the outcome...

cheers

graham

0 Kudos
dm73
Contributor
Contributor

Thanks Andre for your assistance.

I have however just joined the community (noob) so i dont have 10 points to be able to contact him.

I will however try and figure out how to get hold of h8m and let you know how it goes.

thanks again.

D

0 Kudos
a_p_
Leadership
Leadership

I just checked Skype, and saw he's online. You can find his Skype name on his profile page.


André

0 Kudos
dm73
Contributor
Contributor

Hi Graham.

Thanks for the reply. To answer nothing has been written writtento the datastore since the crash. I am waiting for the scan completion (around 10 hours remaining) and i will most definately post the results.

if you hear of anything else please let me know.

thanks

D

0 Kudos
dm73
Contributor
Contributor

Thanks Andre have just sent him a msg now. Waiting in anticipation. Fingers crossed.

D

0 Kudos