Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

*tl;dr: VMWare WS is oddly resilient to rm -rf. Help?*


OS: Linux Mint 14

The problem:

  • VMWare WS 9 reports that an ACE VM is running
  • An instance of VMWare VMX is running, somewhere.
  • The UI for VMWare WS 9 isn't even open

Suggested VMWare W 9 deinstall procedure:

  1. Stop vmware service
  2. Check if vmware service has stopped
  3. All guidelines assume that the vmware service has stopped.

1) Stop the service:

/etc/init.d/vmware stop

Stopping VMware services:
   VMware Authentication Daemon                                        done
At least one instance of VMware VMX is still running.
Please stop all running instances of VMware VMX first.

   VMware Authentication Daemon                                        done

1.1) Well, that didn't work. I'll try to stop another service.

/etc/init.d/vmware-workstation-server stop

Stopping Workstation Server:                                      done

2) Did it work? (no).

lsmod | grep vm

vmnet                  55800  13 
vmci                   87583  1 vsock
vmmon                  76095  1 
kvm_intel             132759  0 
kvm                   414070  1 kvm_intel

The kernel modules for vmmon, and vmnet are still active (bad).


Maybe I can just use the uninstaller?

/usr/bin/vmware-installer -u vmware-workstation 

The VMWare Installer could not shut down all running virtual machines. If you have ACE VMs open, please shut them down or suspend them now and press 'Retry' to continue.


As far as I know, I do not have any VMs open. VMWare Workstation 9 isn't even open. Maybe the purported ACE VM is a manifestation of the vmware daemon that I can't kill.


killall vmware && killall vmware-vmx

{vmware, vmware-vmx} no process found

/var/run:

acpid.pid               motd                    sshd.pid
acpid.socket            mount                   udev
atd.pid                 mysqld                  udisks2
avahi-daemon            network                 user
console                 NetworkManager.pid      utmp
ConsoleKit              nm-dhclient-wlan0.conf  vmblock-fuse
console-kit-daemon.pid  nm-dns-dnsmasq.conf     vmnat.4087
crond.pid               plymouth                vmnet-bridge-0.pid
crond.reboot            pm-utils                vmnet-dhcpd-vmnet1.pid
cups                    pppconfig               vmnet-dhcpd-vmnet8.pid
dbus                    resolvconf              vmnet-natd-8.pid
initramfs               rsyslogd.pid            vmnet-netifup-vmnet1.pid
libvirt                 samba                   vmnet-netifup-vmnet8.pid
libvirtd.pid            sdp                     vmware
lock                    sendsigs.omit.d         wpa_supplicant
mdm.pid                 shm
mdm_socket              sshd

Maybe vmblock-fuse contains something interesting?

vmblock-fuse
├── blockdir
└── dev

1 directory, 1 file

less /var/run/vmblock-fuse/dev

I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am VMBLOCK-FUSE^@I am


This is getting silly

Any suggestions?

Thank you.

share|improve this question
1  
killall vmware-vmx? Or check /var/run for any vmware PIDs and kill -9 them. –  Xyon Jan 24 '13 at 7:14
    
Updated my question. Is there a canonical means of determining the associated host application? I don't want to remove anything required by VirtualBox. Vmnet is probably associated with VMWare, but I could be wrong. If I'm wrong, that would be bad. –  Delete Jan 24 '13 at 13:25
1  
I just looked in /var/run while running a VM in virtualbox, and there's nothing there beginning with a V - which leads me to suspect that those pidfiles you have there are all vmware. It looks like you've got a lot of vmware networking devices active - what happens when you disable / kill all of those and stop vmware? –  Xyon Jan 24 '13 at 13:37
    
Sorry, I'm in the middle of a lecture. Literally nothing changed. I've updated my answer. Maybe the peer-processes are somehow resistant to SIGKILL? –  Delete Jan 24 '13 at 14:19
1  
Right, so that's a stale pid, which indicates that the vmware process isn't running. Can you verify that with the other vmware pids in there? Hopefully you'll find the one which is blocking your uninstall. –  Xyon Jan 24 '13 at 15:21

1 Answer 1

I had the same issue and I tried this:

1- Download this patch to fix bug for Linux 3.5+

2- Exec the patch

3-sudo rm /lib/modules/$(uname -r)/misc/vm*

4- Reboot

5- sudo vmware-modconfig --console --install-all

6- Try again

I hope you'll find this usefull. It worked for me.

Regards.

share|improve this answer
1  
What exact patch do you mean here? –  vonbrand Jan 30 '13 at 10:34
    
By try again, what do I try again? Everything? –  Delete Jan 30 '13 at 19:07
    
Sorry I forgot the link... By try again, I mean, try again your uninstall procedure. –  Djamy Jan 30 '13 at 22:55

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.