VMware Communities
Benedikt31
Contributor
Contributor
Jump to solution

Are there known issues with the snapshot manager of VMware Workstation?

Hi

I am using Workstation 12.5 and had a couple of serious issues while working with the snapshot manager in the last year. In four times I ended up with a vm unwilling to start due to errors with the virtual disk. It tells me something like "the virtual disk or one of disks it depends on is not a virtualdisk"-

The last crash was last week and I have analyzed the situation to some extent.

I am working with snapshots. I have had finalized the work in the VM, shuted down the vm, snapshotted the current state, cloned the new snapshot. Shutdown the whole system. The day after I booted the system, launched Vmware Workstation and pressed play: ~ "The virtual disk is not a virtual disk"

And that was right. The disk that the *.vmx was pointing to, was a single file with no content at all. That's generally fine, because the vmx points to the latest state and this disk should only contain the modifications compared to the latest snapshot. And there were modifications, because I never have started the VM after the snapshot. However, the file was completely empty (nulls), even the disk description in the beginning was not there.

After investing some time I get rid of the latest two or three snapshots by editing some of the files in the directory of the VM and I were able to start the VM again. Of course loosing these snapshots.

If you have any ideas how I can predict this behavior or bypass, I would be more than happy.

P.S.: Just FYI. The clone of the snapshot was fine. But unfortunately I was not able to restore the VM to any snapshots. I was always complaining about the defect disk, until I deleted the two or three snapshots manually.

Reply
0 Kudos
1 Solution

Accepted Solutions
Benedikt31
Contributor
Contributor
Jump to solution

Do not create full clones of just created snapshots. At least start the VM after the snapshot was created and stop and restart Vmware Workstation.

View solution in original post

Reply
0 Kudos
5 Replies
wenwang_myheave
Enthusiast
Enthusiast
Jump to solution

Can you offer more information for deep investigation:

1. WS build

2. Host OS, Guest OS information

3. Detailed steps about how to reproduce this issue

4. ui log (should be under %TEMP%, you can get detailed log location under "Help > About VMware Workstation ")

5. vm related log file, you can get the vm log file location by choosing the VM within VM library, then check "Help > About VMware Workstation".

6. if there is a vm crash, there should be also *.dmp, *.corez file under the same location as vm log file, pls collect them as well

Reply
0 Kudos
Benedikt31
Contributor
Contributor
Jump to solution

Hi

I have only limited information, as all logfiles were overwritten by daily usage and I am not able to reproduce this error. Unfortunately I was not aware of the logfiles in %TMP%, so I only have focussed and used the logfiles in the directory of the vm for investigation. Below my answers. In the meantime, I will try to reproduce this error again.

1. WS build

is 12.5.9 build-7535481

2. Host OS, Guest OS information

Host OS is Windows 7 Professional, 64-bit 6.1.7601, Service Pack 1, Guest OS is MS Windows Server 2016 Standard, Build 14393

3. Detailed steps about how to reproduce this issue

As today I am not able to reproduce this error. I am working three days a week or so with WS. In the last 12 month I was faced four times with this error message. Typically after I have created a backup of the used VM. I listded the steps I had done that resulted in the error:

a) shutdown vm

b) create a new snapshot of the vm

c) take a full clone of the just created snapshot, to be used as backup, if primarily VM crashes 🙂

d) closed WS, after cloning finished

e) shutwdown the Host OS

f) sleep over night

g) start host OS

h) start WS

i) start VM --> error message similar to: the file is not a virtual disk

Interestingly my investigation found only an issue with the latest snapshot-vmdk created. That was the vmdk that would store store data, that would had been changed after or in addition to the latest snapshot. I find this of interest, because the VM was never started (or I was not able to start the VM), after the snapshot was created. So the VM / Windows Sever  never had the chance to generate data on top of the latest snapshot. So my impression is, that the creation of the snapshot resulted in a functional snapshot and in a non-functional latest state used by the Snapshot manager / WS. Unfortunately the snap shot manager did not allow to set the current state to the snapshot or any other snapshot. Probably WS is always checking the "correctness" of all VMDKs, regardless if they are part of the choosen snapshot or not.

4. ui log (should be under %TEMP%, you can get detailed log location under "Help > About VMware Workstation ")

the logfiles in the specific time do not exists any more. There's only a vmware-vmx-nnn.log, that seems to be related to the start of the exported full clone (used as backup VM)

5. vm related log file, you can get the vm log file location by choosing the VM within VM library, then check "Help > About VMware Workstation".

see above

6. if there is a vm crash, there should be also *.dmp, *.corez file under the same location as vm log file, pls collect them as well

the vm did not start at all, so there is no *.dmp and *.corez

Reply
0 Kudos
Benedikt31
Contributor
Contributor
Jump to solution

I was able to reproduce the error. My steps:

1. Launching VM, logging in, shutdown the guest OS

2. Goto to VM | Snapshot Manager | Create Snapshot, providing name and description.

3. Goto VM | manage | Clone | full Clone

4. After Clone was sucessfully created, I deleted the clone, while the clone as VM is still visible as Tab in WS.

5. Goto to VM | Snapshot Manager | Create a second Snapshot, providing new name and description.

Error Message

6. Starting the VM: The file specified is not a virtual disk.

According to vmware-ui.log the ....8.vmdk is not a virtual disk. 8.vmdk. is not mentioned in the vmsd file, but in the vmx-file. It has the initial size of 11.206.656 bytes. It has no content at all, even the file descriptor section at the binning is missing.

2018-04-03T13:36:54.936+02:00| vmui| I125: DISKLIB-LIB   : Failed to open 'C:\vmware\Noctua Demo System V4.1\iam-garancy-dev\iam-garancy-dev-000008.vmdk' with flags 0x17 The file specified is not a virtual disk (15).

2018-04-03T13:36:54.936+02:00| vmui| I125: VMHS: failed to open 'C:\vmware\Noctua Demo System V4.1\iam-garancy-dev\iam-garancy-dev-000008.vmdk' : The file specified is not a virtual disk (15).

2018-04-03T13:36:54.936+02:00| vmui| I125: Cmd /host2/#_client/util/disk/cmd/##4ca/op/getDiskInfo/ failed: The file specified is not a virtual disk

2018-04-03T13:38:39.819+02:00| vmui| I125: DISKLIB-LINK  : "C:\vmware\Noctua Demo System V4.1\iam-garancy-dev\iam-garancy-dev-000008.vmdk" : failed to open (The file specified is not a virtual disk). 

2018-04-03T13:38:39.819+02:00| vmui| I125: DISKLIB-CHAIN : "C:\vmware\Noctua Demo System V4.1\iam-garancy-dev\iam-garancy-dev-000008.vmdk" : failed to open (The file specified is not a virtual disk).

2018-04-03T13:38:39.819+02:00| vmui| I125: DISKLIB-LIB   : Failed to open 'C:\vmware\Noctua Demo System V4.1\iam-garancy-dev\iam-garancy-dev-000008.vmdk' with flags 0x110015 The file specified is not a virtual disk (15).

Attached you will find the complete vmware-ui-nnnn.log and two other logs from the same day. And the vmx file.

Reply
0 Kudos
Benedikt31
Contributor
Contributor
Jump to solution

I have checked the state of the vmdk.

After creating the snapshot, the new created vmdk seems to be fine.

After creating a full clone, the vmdk only consists of nulls.

Obviously the function "creating a full clone of a snapshot" is buggy.

Reply
0 Kudos
Benedikt31
Contributor
Contributor
Jump to solution

Do not create full clones of just created snapshots. At least start the VM after the snapshot was created and stop and restart Vmware Workstation.

Reply
0 Kudos