DataStore Gone :( due to removing LUNs: pls help

[yvv47]

Hi,

Pls see the below steps I did :

• I created two luns 850 GB & 76 GB on EMC CX300 and assigned it to ESX and expanded/increased the datastore
• Then I deleted those luns directly from Storage – EMC CX 300 (I did not removed the LUNs from the datastore) – I think this is the only thing which I missed and problem created.
• After seeing the ESX datastore again, it was grayed out and nothing was happening.
• When trying to take console of those VMs it was not letting me to see the OS window
• And when tried RDP it was going

Please help me to restore the datastore again to the normal state. I am worried about those 5 VMs (Servers) that I have build 6 months ago

I need to add 1 TB to the datastore but it is not functioning at all.

Please help me...

THanks,

Yash

[mcowger]

If you deleted those LUNs from the storage side, your data and virtual machines are gone (I'm sorry to be the bearer of bad news).

You'll need to create new LUNs and new datastores, then restore form backups (hopefully you have them).

[yvv47]

HI Matt, thanx for your reply. But I have deleted those 2 luns which i have used for expanding the datastore. which was the part of the datastore for only 2 minutes. And still I'm able to take RDP to those 5 servers :smileyconfused:

datastore was created through 3 LUNS

and then I was adding 2 LUNs and I deleted last two luns. So is it lost?

[weinstein5]

Yes if you expanded the dtastore by the use of extents than yes the data is lost - 

[yvv47]

here comes the pain

Whethere it was just a part of the datastore for 2 minutes ? can it be restorable (I don't have any backups)

I think tomorrow will be my last day there @ my org.

[yvv47]

thanks all for your suggestions/answers.

But would like to hear something positive as I'm in danger now

I have a question if the data is lost then why  still I'm able to log into those servers through RDP ? :smileyconfused: :smileyshocked:

Please show me some hope for my safety...

thanks & appretiation

yash...

[mcowger]

You can RDP to them because Windows and the kernel are already in memory.  As soon as you try to access anything that requires access to the parts of the disks that aren't in memory, or you reboot those VMs, you will have lost everything.

You can try copying some data off via RDP or CIFS to get what you can, but anything you can't get that way is permanently gone without some VERY expensive data recovery techniques.

[yvv47]

Thanks Matt for your help!

I lost the game. I have to rebuild all the servers, don't know what my client will say just pray for me

There is no problem of data loss, the only problem is time. We were about to use this as a DR site, and we have copied all the production data to this. Now again we have to do the same.

Any ways...Thanks all for helping me and providing the answer.

[yvv47]

Hi...

I powered off all the VMs and then by re-opening the vmware client I got my datasote back

Is there any reason why the data is still safe after shuting down several times and also tried to copy the database files?

Or something else happned to the LUN/Datastore? any ideas gurus?

[mcowger]

Basically you got lucky because the new extents hadn't been used much (if at all ) yet.

I personally would NOT trust this datastore at all, at this point, and would migrate to a new ASAP.

[yvv47]

oh, then I would shift all the VMs one by one to another one but m again worried as there is no other source to create another datastore...

WIll have to plan something new.

thanks matt for really great suggesiton.

[yvv47]

Hi,

Here is the screenshot of the datastore properties, it seems fine, I'm not able to take the backup as we dont have enough space (3tb)

Please see the screenshot and suggest me if we can continue with the same or not? :smileyconfused:

[mcowger]

Like I said, I personally wouldn't trust it at all and would move my VMs to some other location long enough to recreate this datastore.