VMware Communities
continuum
Immortal
Immortal

Please popup a list of VMs that could not be powered off last time WS was killed unexpectedly

WS 12 does not handle it well when the host OS reboots itself while the user is away from the screen - it makes no difference if the reboot is unexpected or after a Windows update.
The VMs I typically run often use /tmp for what they are doing.
If you come back to your screen and realize that your host rebooted itself it is a pain because the work in /tmp will be lost and you also have to search which VMs you used last time.
Suggestion: during normal operation maintain a file with a list of all running VMs.
Remove VMs that are powered off cleanly from that list.
When ever WS starts new - look up which VMs may have been killed hard during last session.
Offer to re-launch them.
Include VMs from all users into this checks.
Sometimes I read that WS 12 is supposed to power down VMs gently when the host wants a reboot.
Please fix the documentation if it really claims that - this is not funny ...
I never regarded this as a serious problem but since the last UI-redesign I dont know a easy way to even find the VMs that were killed hard during last session.
Have a look at Firefox - it offers to restore your last tabs after a crash.

Ulli


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

Reply
0 Kudos
8 Replies
HuanguoZhong
VMware Employee
VMware Employee

Thanks for you suggestion, it's a good feature request.

Reply
0 Kudos
AlanFer
Contributor
Contributor

Hello continuum, I am looking to download the ISO of the recovery that you made to vmfs you post on an old post, but all these broken links, could you help me know if there is even this tool and so it is still possible download and from where?

The tool is published in: https://communities.vmware.com/thread/404225?start=0&tstart=0

Thank you.

Reply
0 Kudos
continuum
Immortal
Immortal

Hi Alan
the latest one can be found at
http://vm-sickbay.com
Ulli


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

Reply
0 Kudos
wila
Immortal
Immortal

Ulli, your link only works when I drop the "s" from https.

eg. http://vm-sickbay.com

works whereas your link times out.

--

Wil

| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
Reply
0 Kudos
continuum
Immortal
Immortal

Oops ...
thank you Wil


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

Reply
0 Kudos
WeigangHuang
VMware Employee
VMware Employee

Hi continuum,

Thanks very much for your great suggestion! We are working on the enhancement. Would you please answer the following questions to help us clarify the requirement?

  1. How many VMs do you usually open in Workstation? And how many of them will be powered on for use?
  2. Do you think it is acceptable to restore all VMs unpromptedly? Let’s say, vm1, vm2, vm3, vm4, vm5 were opened, among which vm1 and vm2 were running, then host rebooted and WS was killed, on the next start of WS, do you think it is OK to reopen these 5 vms (1-5) without popping up the vm list - all vms opened in WS before the unexpected exit will be reopened automatically?
  3. Suppose you just opened a new VM, and immediately the host had a reboot (maybe the scheduled update time is up), on the next start of WS, do you expect such new VM to be reopened automatically?
  4. Similar to 3, but a VM was closed, then the host got an immediate reboot, do you expect such a VM not to be opened on the next start?
  5. Do you think it is meaningful to keep “current VM”? For example, before the unexpected exit, you were operating the vm3, which had the active tab, on the next start of WS, do you expect the same vm3 to be the active tab as well? Is it acceptable we just reopen all but lose the “current vm”?
Reply
0 Kudos
continuum
Immortal
Immortal

First of all - sorry for the late reply - I am quite busy with VMFS-recoveries at the moment.
1. How many VMs do you usually open in Workstation? And how many of them will be powered on for use?


Nowadays I most of the times have one WS-instance running as a regular user and another one running as administrator - I wish I could avoid that easily but it ends like that most of the times.
Typically I have one or two ESXi VMs, 2 - 4 Linux Ubuntu-commandline VMs and one Win7 VM running.
Every instance typically has 5 - 10 tabs open.
Amount of VMs in the sidebar varies between 30 and 150.
The working host has 8Gb RAM and I typically run VMs with a total of upto 16GB RAM side by side.

2. Do you think it is acceptable to restore all VMs unpromptedly? Let’s say, vm1, vm2, vm3, vm4, vm5 were opened, among which vm1 and vm2 were running, then host rebooted and WS was killed, on the next start of WS, do you think it is OK to reopen these 5 vms (1-5) without popping up the vm list - all vms opened in WS before the unexpected exit will be reopened automatically?


I maybe spoiled by earlier WS-versions - I dont remember upto which version a pending Windows-reboot-request was efectively denied if any VM is still running.
So I am used to get away without taking written notes which of the hundreds of development Linux-VMs I am just working on.
So whenever an unexpected reboot can not be avoided I nowadays often do not find the VMs that I was working on - once the host is up again.
3. Suppose you just opened a new VM, and immediately the host had a reboot (maybe the scheduled update time is up), on the next start of WS, do you expect such new VM to be reopened automatically?
Yes - that would be great.

4.Similar to 3, but a VM was closed, then the host got an immediate reboot, do you expect such a VM not to be opened on the next start?
I would hope to find such a VM in the list of open tabs.

5.Do you think it is meaningful to keep “current VM”? For example, before the unexpected exit, you were operating the vm3, which had the active tab, on the next start of WS, do you expect the same vm3 to be the active tab as well? Is it acceptable we just reopen all but lose the “current vm”?

Lets say I have 10 tabs opened - 4 VMs actually running. If the host crashes then or forces a reboot - I would be glad if I had 4 tabs with those VMs that were running before the unexpected reboot.
To sum it up:
first priority: have a procedure that will reliably produce a list of VMs that were running before something unexpected occured. This could be helped if WS maintains a list of active vmware-vmx.exe processes and their related vmx-path.
The reg-key HKEY_CURRENT_USER\Software\VMware, Inc.\Running VM List almost does that already but it only lists VMs launched by the user.
Looking up VMs launched by the administrator using registry is a pain.
So I would rather vote for a solution that adds an open VM to config.ini and removes the item from config.ini after the VM has been shutdown cleanly.
I do not need a super-easy-to-use solution but rather one that is reliable - even if it requires to look up a ini-file.
By far the best solution for my taste: just deny a reboot-request of the host  if any VM is running !


Thanks Ulli


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

Reply
0 Kudos
WeigangHuang
VMware Employee
VMware Employee

Thanks a million! Your answering is very helpful to us!

Reply
0 Kudos