How to Gather Data for a Virtual Box Virtual Machine That is Unresponsive.

Maclean Liu(刘相兵

How to Gather Data for a Virtual Box Virtual Machine That is Unresponsive.

Applies to:
Oracle Virtual Desktop Infrastructure - Version: 3.2.2 and later   [Release: 3.0 and later ]
Information in this document applies to any platform.
Goal
The goal of this document is to help a user collect enough data that a cause can be determined why their Virtual Box virtual machine might have hung.
Solution

    Identify either the uuid or the name of the virtual machine in question.  The VBoxManage command can be used to collect these:
    bash-3.00# VBoxManage list runningvms
    "caissa1 (caissa)" {a4ffa123-5209-41f7-98ec-4ed9f141175d}

    Collect the Virtual Box logs from a hung session.  http://blogs.oracle.com/fatbloke/entry/virtualbox_log_files
    Note the times and dates when these sessions are started and when they seem to go into a hung state.
    Try to save the state of a hung machine. This can be done with the following command :

# VBoxManage controlvm savestate <uuid>|<name>


        and archive all of the files in the Virtual Machine's folder.


    4.  Inject a non-maskable interrupt into the virtual machine using the command:

# VBoxManage debugvm injectnmi <uuid>|<name>


and record if and how the virtual machine reacts.  If it reacts at all then it is the virtual machine that is hung and is not a problem on the Virtual Box end.

     5.  Explain all tests run on the guest to verify that it was unresponsive (ping, etc.).