VMware Cloud Community
Squibz
Contributor
Contributor
Jump to solution

Server 2012 Converted with VMware Converter will not boot in ESXi "Cannot open the disk or one of the snapshot disks it depends on"

A very trigger happy ex-employee performed the conversion to a NAS and then wiped the Server to install ESXi and import the VM. Trying my best to get this to boot but it is giving me the error listed in the Title. See below for the full ESXi console text and attached for the Log. I've poured through articles for the last 3 days but have exhausted all resources I've been able to find. It may be worth noting that I can boot the VM in VirtualBox with Hirens BootCD and access the entire directory - So i believe theres something broken in terms of finding a file to boot it?

Thanks for any help provided!

Power On VM

Key haTask-7-vim.VirtualMachine.powerOn-179973191

Description

Power On this virtual machine

Virtual machine

Server02.MSMETZCO.LOCAL

State

Failed - File system specific implementation of LookupAndOpen[file] failed

  Errors

  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • File system specific implementation of LookupAndOpen[file] failed
  • Failed to lock the file
  • Cannot open the disk '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' or one of the snapshot disks it depends on.
  • Module 'Disk' power on failed.
  • Failed to start the virtual machine.
Reply
0 Kudos
1 Solution

Accepted Solutions
a_p_
Leadership
Leadership
Jump to solution

Failed - Unsupported or invalid disk type 2 for 'scsi0:0'. Ensure that the disk has been imported.

The log file shows the disk format as "monolithicSparse", which is not a supported format on ESXi (i.e. only supported on e.g. VMware Workstation, Player, or Fusion).

You mentioned that an "ex-employee performed the conversion to a NAS", so what you may want to do is to use VMware Converter, and convert the VM back to the ESXi host, so that the virtual disk will be converted to a supported format.


André

View solution in original post

Reply
0 Kudos
9 Replies
SupreetK
Commander
Commander
Jump to solution

From within the VM folder, does the disk report 'Device or Resource Busy' if the command <touch *> is run? If yes, run the command <vmkfstools -D "Server02.MSMETZCO.LOCAL-flat.vmdk"> and identify the MAC address of the host that is holding a read-only lock on the disk. Is it a VMFS-6 datastore?

Cheers,

Supreet

Reply
0 Kudos
bluefirestorm
Champion
Champion
Jump to solution

The VM has two virtual disks pointing to the same VMDK file.

2018-10-11T12:45:29.965Z| vmx| I125: DICT          scsi0:0.fileName = "Server02.MSMETZCO.LOCAL.vmdk"

2018-10-11T12:45:29.965Z| vmx| I125: DICT          scsi0:1.fileName = "Server02.MSMETZCO.LOCAL.vmdk"

That is the likely cause of the error. Either the VM is suppose to have only 1 virtual disk or one of the virtual disk is pointing to the incorrect VMDK.

Reply
0 Kudos
Squibz
Contributor
Contributor
Jump to solution

Hey there, first off thanks so much for your reply. I noticed that as well. There is a C drive and a D drive along with an E drive in the server.

C - OS

D - Recovery Part

E - Data Drive

Also there is only a Single VMDK that was created with the converted.

I did try to remove the second one from the VMX file but it gave a configuration error. What would you propose as a fix for that?

Thanks!

Reply
0 Kudos
Squibz
Contributor
Contributor
Jump to solution

Thanks for your reply and attempt to help me! Much appreciated.

So I am SSH'd into the server and changed directory to the Volume where my VM is located. I used the command as follows: touch <filepath.vmdk> and it doesnt really do anything. Just accepts it, no output and brings me back to a new line to enter my next command..

Hope that helps?

Reply
0 Kudos
bluefirestorm
Champion
Champion
Jump to solution

It is possible to have only one virtual disk with several volumes inside, not much different from a physical disk. So having drives C, D, E in the same virtual disk is possible.

It is hard to tell what is the configuration error is without seeing the vmware.log when you tried to it power up after removing the extra virtual disk. I assumed you removed the second virtual disk from the VM using the ESXi host client? Or is the configuration error that you are referring shows up when you try to remove the second virtual disk from the VM?

From the log file, it appears that the virtual disk is suppose to be 1.8TB

2018-10-11T12:45:30.204Z| Worker#0| I125: DISKLIB-DSCPTR: Opened [0]: "Server02.MSMETZCO.LOCAL.vmdk" (0x10a)

2018-10-11T12:45:30.204Z| Worker#0| I125: DISKLIB-LINK  : Opened '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' (0x10a): monolithicSparse, 3907115008 sectors / 1.8 TB.

2018-10-11T12:45:30.204Z| Worker#0| I125: DISKLIB-LIB   : Opened "/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk" (flags 0x10a, type monolithicSparse).

Since the VM configuration was referring to the same virtual disk with two different virtual disk controller ports, one of them will fail. It is the equivalent of attaching two drive cables to the same physical drive at the same time; although that is impossible in the physical world. Obviously something has gone wrong with the vmx configuration file as powering up the VM results in attempts to open the same virtual disk twice on different virtual scsi ports.

2018-10-11T12:45:30.195Z| vmx| I125: DISK: OPEN scsi0:1 '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' persistent R[]

2018-10-11T12:45:30.195Z| Worker#0| I125: DISK: OPEN scsi0:0 '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' persistent R[]

Which in turn caused the locking to fail and power up of the VM to fail. Because of the retries it took over 4.5 seconds to do so and it looks like scsi0:1 failed.

2018-10-11T12:45:34.738Z| vmx| I125: DISK: Opening disks took 4543 ms.

2018-10-11T12:45:34.738Z| vmx| I125: Module 'Disk' power on failed.

2018-10-11T12:45:34.738Z| vmx| I125: VMX_PowerOn: ModuleTable_PowerOn = 0

2018-10-11T12:45:34.738Z| vmx| I125: SVMotion_PowerOff: Not running Storage vMotion. Nothing to do

2018-10-11T12:45:34.738Z| vmx| I125: scsi0:0: numIOs = 0 numMergedIOs = 0 numSplitIOs = 0 ( 0.0%)

2018-10-11T12:45:34.738Z| vmx| I125: Closing disk 'scsi0:0'

2018-10-11T12:45:33.223Z| vmx| I125: OBJLIB-FILEBE : FileBEOpen: can't open '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' : Failed to lock the file (262146).

2018-10-11T12:45:33.223Z| vmx| I125: DISKLIB-SPARSE: "/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk" : failed to open (16392): ObjLib_Open failed

2018-10-11T12:45:33.223Z| vmx| I125: DISKLIB-LINK  : "/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk" : failed to open (Failed to lock the file). 

2018-10-11T12:45:33.223Z| vmx| I125: DISKLIB-CHAIN : "/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk" : failed to open (Failed to lock the file).

2018-10-11T12:45:33.223Z| vmx| I125: AIOMGR: AIOMgr_OpenWithRetry: Descriptor file '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' locked (try 0)

2018-10-11T12:45:33.527Z| vmx| I125: AIOMGR: AIOMgr_OpenWithRetry: Descriptor file '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' locked (try 1)

2018-10-11T12:45:33.829Z| vmx| I125: AIOMGR: AIOMgr_OpenWithRetry: Descriptor file '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' locked (try 2)

2018-10-11T12:45:34.131Z| vmx| I125: AIOMGR: AIOMgr_OpenWithRetry: Descriptor file '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' locked (try 3)

2018-10-11T12:45:34.433Z| vmx| I125: AIOMGR: AIOMgr_OpenWithRetry: Descriptor file '/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk' locked (try 4)

Squibz
Contributor
Contributor
Jump to solution

So I removed the second reference to the same VMDK in the VMX file by opening it with WinSCP's built in editor. Not sure if that's the correct way to do it, but it gave a glaring error. See the log attached for the bootup from when I attempted that. Also attached is the VMX file.

I'm not sure what information is relevant, but figured it wouldn't hurt to mention the previous configuration was one 2tb hard drive partitioned to those 3 partitions listed (C,D,E) and 1.8tb is the TOTAL size of that disk, where 1,05 TB is what was used. is that potentially the problem? The ESXi host had 1.5tb available in its Datastore, but I figured that would be ok with the actual usage being only 1.05 or so.

Here's the info that the ESXi provides when i try to start it with the below lines removed:

scsi0:1.present = "TRUE"

scsi0:1.fileName = "Server02.MSMETZCO.LOCAL.vmdk"

ESXi Console info:

haTask-7-vim.VirtualMachine.powerOn-179973574

Description

Power On this virtual machine

Virtual machine

Server02.MSMETZCO.LOCAL

State

Failed - Unsupported or invalid disk type 2 for 'scsi0:0'. Ensure that the disk has been imported.

Errors

    Unsupported or invalid disk type 2 for 'scsi0:0'. Ensure that the disk has been imported.

    Unable to create virtual SCSI device for 'scsi0:0' (/vmfs/volumes/595e5775-5570aed0-76ea-20474787e3d0/Metz/Server02.MSMETZCO.LOCAL/Server02.MSMETZCO.LOCAL.vmdk).

    Module 'DevicePowerOn' power on failed.

    Failed to start the virtual machine.

Thank you so much!!

Reply
0 Kudos
bluefirestorm
Champion
Champion
Jump to solution

The vmware.log you attached still has those lines and still has the same errors as the first vmware.log. It is only less than 2 hours later than the first vmware.log you attached. So it does not match the "Failed - Unsupported or invalid disk type 2 for 'scsi0:0'. Ensure that the disk has been imported." error that you pasted.

2018-10-11T14:19:07.873Z| vmx| I125: DICT           scsi0:0.present = "TRUE"

2018-10-11T14:19:07.873Z| vmx| I125: DICT          scsi0:0.fileName = "Server02.MSMETZCO.LOCAL.vmdk"

2018-10-11T14:19:07.873Z| vmx| I125: DICT           scsi0:1.present = "TRUE"

2018-10-11T14:19:07.873Z| vmx| I125: DICT          scsi0:1.fileName = "Server02.MSMETZCO.LOCAL.vmdk"

The vmware.log files will roll over and earlier tries would have a number in the filename.

It would be best to edit the VM configuration using the ESXi host client.

Anyway, have a look at this KB https://kb.vmware.com/s/article/1028943

It is IMPORTANT to make a backup of the VM though BEFORE taking action based on this KB.

a_p_
Leadership
Leadership
Jump to solution

Failed - Unsupported or invalid disk type 2 for 'scsi0:0'. Ensure that the disk has been imported.

The log file shows the disk format as "monolithicSparse", which is not a supported format on ESXi (i.e. only supported on e.g. VMware Workstation, Player, or Fusion).

You mentioned that an "ex-employee performed the conversion to a NAS", so what you may want to do is to use VMware Converter, and convert the VM back to the ESXi host, so that the virtual disk will be converted to a supported format.


André

Reply
0 Kudos
Squibz
Contributor
Contributor
Jump to solution

This was it! Thank you so much! Converted it again from Workstation to ESXi and actually noticed that I could then remove that second reference to the same VMDK at that point. Oddly enough it failed at 98% due to a BCD error, but i ran Hiren's on the machine and fixed the BCD just fine and it worked flawlessly.

I cannot thank both of you enough!

Bob

Reply
0 Kudos