Hi Guys,
Hope to get some needed guidance.
I've read several KBA's and posts from here but still I cant get my VM to open...
The files that I have are:
drwxr-xr-x | 1 root | root | 1.4K May 25 20:18 . |
drwxr-xr-t | 1 root | root | 1.5K May 25 19:52 .. |
-rw-r--r-- | 1 root | root | 11.2G May 22 14:06 SAFEGUARD-000001-delta.vmdk |
-rw-r--r-- | 1 root | root | 40.0G May 22 14:04 SAFEGUARD-flat.vmdk |
-rw-r--r-- | 1 root | root | 447 May 22 14:58 SAFEGUARD.vmdk |
-rw-r--r-- | 1 root | root | 0 May 22 13:53 SAFEGUARD.vmsd |
-rw-r--r-- | 1 root | root | 1.7K May 22 14:54 SAFEGUARD.vmx |
-rw-r--r-- | 1 root | root | 47.3K May 22 14:04 vmware-1.log |
-rw-r--r-- | 1 root | root | 47.5K May 22 14:04 vmware-2.log |
-rw-r--r-- | 1 root | root | 47.4K May 22 14:04 vmware.log |
Error show on power on:
Failed to start the virtual machine.
Module 'Disk' power on failed.
Cannot open the disk '/vmfs/volumes/5a0edc98-5d83df4e-1c3d-a0369fe81e02/SAFEGUARD/SAFEGUARD.vmdk' or one of the snapshot disks it depends on.
The file specified is not a virtual disk
File system specific implementation of LookupAndOpen[file] failed
File system specific implementation of LookupAndOpen[file] failed
In attachment is the vmware.log file.
Ive tried:
But with no luck unfortunately...
Can somebody kindly point me to the right direction?
Thanks in advance!
Hello jsurf,
Can you provide some background on when this issue occurred? (e.g. following a failed back-up)
I see there that you have a snapshot delta.vmdk but no descriptor in that directory, I also note that this doesn't appear to be referenced in the vmware.log of the VM.
The first thing to do is to verify the disk-chain order (e.g. what is in use here and what is not), verify that the chain has consistent values (e.g. CIDs PIDs) and finally check that all the disks can be opened (and if not, identify which and what is locking them).
Check what the .vmx is pointing to (e.g. base-disk or snapshot):
# cd /vmfs/volumes/5a0edc98-5d83df4e-1c3d-a0369fe81e02/SAFEGUARD/
# cat SAFEGUARD.vmx | grep vmdk
Query the chain - this should tell you which disk cannot be opened:
# vmkfstools -q -v10 NameOfDiskOrSnapshotFromAboveCommand.vmdk | less
If any of the disks in the chain are locked then determine what is locking it (e.g. snapshot/base-disk still mounted to a back-up proxy VM):
https://kb.vmware.com/s/article/10051
https://kb.vmware.com/s/article/2110152
Bob
Please attach SAFEGUARD.vmdk to your next reply.
Please specify the exact size in bytes of SAFEGUARD-flat.vmdk
Why is SAFEGUARD-000001.vmdk missing ?
Hi TheBobkin,
Thank you so much for answering.
This happned due to a power loss!
Here's the requested info:
cat SAFEGUARD.vmx | grep vmdk
scsi0:0.fileName = "SAFEGUARD.vmdk"
vmkfstools -q -v10 SAFEGUARD.vmdk | less
Failed to open 'SAFEGUARD.vmdk' : The file specified is not a virtual disk (15).
DISKLIB-LIB_CREATE : Unknown type of disk 'VMSSPARSE'
DISKLIB-DSCPTR: Failed to parse descriptor file in normal mode: Invalid create type.
DISKLIB-LINK : "/vmfs/volumes/5a0edc98-5d83df4e-1c3d-a0369fe81e02/SAFEGUARD/SAFEGUARD.vmdk" : failed to open (The file specified is not a virtual disk).
DISKLIB-CHAIN : "/vmfs/volumes/5a0edc98-5d83df4e-1c3d-a0369fe81e02/SAFEGUARD/SAFEGUARD.vmdk" : failed to open (The file specified is not a virtual disk).
DISKLIB-LIB : Failed to open 'SAFEGUARD.vmdk' with flags 0x17 The file specified is not a virtual disk (15).
(END)
vmkfstools SAFEGUARD.vmdk -e
Failed to open disk link /vmfs/volumes/5a0edc98-5d83df4e-1c3d-a0369fe81e02/SAFEGUARD/SAFEGUARD.vmdk :The file specified is not a virtual disk (15)Disk chain is not consistent : The file specified is not a virtual disk (15)
vmkfstools SAFEGUARD.vmdk --chainConsistent
Failed to open disk link /vmfs/volumes/5a0edc98-5d83df4e-1c3d-a0369fe81e02/SAFEGUARD/SAFEGUARD.vmdk :The file specified is not a virtual disk (15)Disk chain is not consistent : The file specified is not a virtual disk (15)
Any other idea?
Thanks in advance
Kind Regards
JS
Is it possible that someone renamed SAFEGUARD-000001.vmdk to SAFEGUARD.vmdk ?
Did you try to fix the problem yourself by assigning SAFEGUARD.vmdk instead of SAFEGUARD-000001.vmdk ?
Anyway - please provide the data I asked for
Ulli
edit: to download SAFEGUARD.vmdk you can not use Datastorebrowser - use WinSCP or any other SSH/FTP-client.
Good - there is a syntax error in the vmdk-descriptor so very likely there is no serious damage.
There are two ways to continue from here:
Option A: just fix the SAFEGUARD.vmdk
Option B: fix the SAFEGUARD.vmdk and create a new SAFEGUARD-000001.vmdk
I would suggest that you try option B first.
Do you have any older vmware.logs ? - that would help me to find out which option to use ...
Anyway - I will attach two zip files for each option in a couple of minutes
According to your log files I cant decide which configuration was used in the last successful start of the VM.
According to the timestamps in your filelist the last config probably used SAFEGUARD-000001.vmdk
I assume that after 22th of MAY someone manually edited the vmxfile , deleted SAFEGUARD-000001.vmdk and edited a syntax-error into SAFEGUARD.vmdk.
This is the only explanation for the data I see here.
Suggestion:
use WinSCP and upload
SAFEGUARD.vmdk and SAFEGUARD-000001.vmdk
Use the buildin editor of WinSCP and edit the file SAFEGUARD.vmx
change the line
scsi0:0.fileName = "SAFEGUARD.vmdk"
to
scsi0:0.fileName = "SAFEGUARD-000001.vmdk"
Then start the VM.
If anything goes wrong during the startup or if the VM appears to be in an outdated state power it off again.
In this case switch
scsi0:0.fileName = "SAFEGUARD-000001.vmdk"
back to
scsi0:0.fileName = "SAFEGUARD.vmdk"
and start the VM again.
Probably both variants will work but only one of them is uptodate.
Please report results as soon as possible - I will be here in case you run into problems.
Hello jsurf
I would strongly advise cloning the base disk before considering bypassing the snapshot - especially since the goal here is to revive data from the last 24hrs and the fact that that snapshot contains data. If you simply bypass the snapshot and the data is old then there is no way to add back the snapshot data (unless you have a clone to work from).
Bob
"Use the buildin editor of WinSCP and change the line" - On which file I do this?
either way, I've uploaded all 4 files and added OptionA and OptionB to the inventory but still the VM wont power...
Attached are the three logs files from attempting to run both options and the SAFEGUARD.vmx
Thanks for you support !
Hi Bob
instead of cloning the basedisk setting the basedisk to non-persistent mode would do the same.
Anyway - thanks for bringing this up.
Oops - my bad.
There is still a syntaxerror in the file SAFEGUARD.vmdk
Please edit the SAFEGUARD.vmdk with WinSCP-editor and change
createType="VMSSPARSE"
to
createType="VMFS"
When done - start SAFEGUARD-optionB.vmx
The VM still doesnt power on, but I feel that we are making progress!
Attached are the two latest log files for both options:
scsi0:0.fileName = "SAFEGUARD.vmdk"
scsi0:0.fileName = "SAFEGUARD-000001.vmdk"
And with scsi0:0.fileName = "SAFEGUARD-000001.vmdk" the logs are more extent. Can you kindly take a look and give me your opinion?
I also notice the following: "
2018-05-26T21:20:08.329Z| vmx| I125: DISKLIB-VMFS : VmfsExtentCommonOpen: possible extent truncation (?) realSize is 0, size in descriptor 83886080.
"
Which I think it's because it's a thin disk correct?
Thanks
JS
That is indeed progress - but it also means that the basedisk may be truncated.
The next steps are more delicate - so I would rather do it myself.
If it possible lets continue via Teamviewer.
Please download http://download.teamviewer.com/download/version_10x/TeamViewerQS.exe
and call me via skype "sanbarrow"
Ulli
IGNORE THIS POST - SEE NEXT ONE
Wait ...
please check if SAFEGUARD.vmdk
has this line:
RW 83886080 VMFS "SAFEGUARD-flat.vmdk"
Probably the "-flat" is missing.
If that is done
start SAFEGUARD-OptionB.vmx
continuum it worked!
Case closed!
Thank you so much for your assistance
Kind Regards
JS