Datastore not show in the Storage list

[Hojjat]

Hi Everyone.

We have an ESXi 5.1 server with non-Raided two hard disks and some VMs inside in (Datastore 1 & Datastore 2). After a power failure, some of VMs renamed to "Unknown (inaccessible)" which are in the specified hard disk and now we cannot use them. After some tests, I see that Datastore 1 not show in the Storage list. I tried to add storage and recreate that datastore but this error was occurred :

"Either the selected disk already has a VMFS datastore or the host cannot perform a partition table conversion. Select another disk."

How can i recover my VMs without any data lost ?

Any suggestion will be appreciated.

Regards,

[a_p_]

Please take a look at the "Command Line" section in https://kb.vmware.com/s/article/1011387​ to see whether this helps.

André

[RajeevVCP4]

If there is power outage then you required reboot all hosts ,

All VM will be online or you required power on 1 by 1

[Hojjat]

Thanks for replying,

Reboot did not help me also i cant power on my VMs because power options are grayed.

[nileshm]

Reboot your host once.

[continuum]

Please check if the datastore still has a partitiontable with partedUtil.
If you want me to check the VMFS-volume please read
Create a VMFS-Header-dump using an ESXi-Host in production | VM-Sickbay
and provide a downloadlink.
Ulli

[Hojjat]

Hi continuum​

I have tried to create dump file and download it via WinSCP and attach here for you. Let me know if i have any mistake.

[Hojjat]

Sorry, when i want attach file in forum the upload percentage was not show.

Here is download link :   https://ufile.io/hk4f2

[continuum]

Hello
I dont think that you created the dump for the correct partition.
The dd command finished after just 3 MBs - so you selected the boot partition.
Please create a new dump file for the correct partition.

[Hojjat]

Thanks for replying,

Which one is the correct partition in screenshot?

My disk capacity : 2 TB

[continuum]

It should be partition 3

[Hojjat]

This error was appeared when Partition 3 was selected :

[continuum]

This means that there is no way to fix the volume so that it can be mounted again.
But it may be possible to still extract some or all of the VMs.
If the VMs are important enough we can use a Linux VM to create the dump.
Please download my Linux LiveCD from here
https://sanbarrow.com/livecds/moa64-nogui/MOA64-nogui-incl-src-111014-efi.iso
Install Anydesk and call me via skype tomorrow.
Ulli

[Hojjat]

Thanks a lot.

What time can i call you tomorrow ?

Best Regards,

[continuum]

You can call after 11:00 in the morning (Berlin Time)

[continuum]

Update:
we are extracting the VMs now.
A first VM is already back in production.
The others will follow later ... copying 1.7 TB of data takes its time when both read and write has to be done from Linux via sshfs.
Ulli

[mdavilateldat]

Hi all,  @continuum

I have read your comments regarding this issue with datastore.

I believe i have similar case, but unsure because are showing partitions as naa.

I cannot make just yet the dump file but as per you can see on the screenshots, there is anyway to extract the VMs? These are important production VMs, unfortunately no raid or backups were scheduled.

Thanks in advance

Moisés

[mdavilateldat]

Hi again,

If anyone else beside continuum please can help me out, i summarize the issue this the datastore:

I am in critical situation with old vSphere 5.5 unfortunately support by VMWare ended septembre 2018.

In my company we have 11 VM on single local hard disk (no backup) under ESXI 5.5.0 build 1131820

8 of those 11 VM were up and running but there was power outage. So the esxi was not shutdown properly.

First i had issues to access vSphere via vcenter (i restarted the agent management services via ssh) but the behaviour of the exsi was not normal, i got on logs randomly hosd non responsive messages every 5 min aprox.

Then after new reboot i got access again via vcenter (not url as this version is not compatible as you know) but datastore was lost, no persistent message datastore under storage tab.

See attached errors and current host partitions.

I am in the point that i added new local disk and created new datastore named datastorerescue and the VMFS-Header-dump

Thanks in advance!

Mosés

[continuum]

looking into it now
Ulli

[MikeStoica]

Your datastore still not available?

Can you take a screenshot of your datastore and VM?

[Hojjat]

Yes, My datastore was disappeared as shown in the screenshot ("datastore-1" is not available - capacity 1.8 TB)

[Hojjat]

ScreenShot

[Hojjat]

Sorry

Correct Screenshot

[MikeStoica]

1

List all volumes that have been detected as snapshots.

esxcli <conn_options> storage filesystem list

2

Run esxcli storage filesystem mount with the volume label or volume UUID.

By default, the volume is mounted persistently, use --no-persist to mount persistently.

esxcli <conn_options> storage filesystem volume mount --volume-label=<label>|--volume-uuid=<VMFS-UUID>

[Hojjat]

Thanks,

After run esxcli storage filesystem mount, this error was appeared (screenshot) :

[MikeStoica]

Run the command like this:

esxli storage filesystem volume mount --volume-uuid=(your UID)

[Hojjat]

The error was appeared again :

[MikeStoica]

I see that it shows as mounted.

If you do a storage rescan does it show up?

[Hojjat]

I've done this already but it does not show up.

The size of volume and free space is equal in the last screenshot, this means that all data have been deleted ???!!!

[a_p_]

The size of volume and free space is equal in the last screenshot, this means that all data have been deleted ???!!!

With the huge size it shows, it's more likely an issue with the partition table, and/or the file VMFS system.

I'd suggest you follow continuum​'s post. He's an expert when it comes to data recovery.

André