Troubleshooting VM Problems

Citrix provides two forms of support: free, self-help support on the Citrix Support[Citrix] website and paid-for Support Services, which you can purchase from the Support Site. With Citrix Technical Support, you can open a Support Case online or contact the support center by phone if you experience technical difficulties.

The Citrix Support site hosts various resources that may be helpful to you if you experience unusual behavior, crashes, or other problems. Resources include: Support Forums, Knowledge Base articles, and product documentation.

If you experience unusual VM behavior, this article aims to help you solve the problem describes where application logs are located and other information that can help your XenServer Solution Provider and Citrix track and resolve the issue.

Troubleshooting of installation issues is covered in the XenServer Installation Guide. Troubleshooting of XenServer host issues is covered in the XenServer Administrator’s Guide.

Note

Citrix recommends that you follow the troubleshooting information in this article solely under the guidance of your XenServer Solution Provider or Citrix Support.

Vendor Updates: Citrix recommends that VMs are kept up-to-date with operating system vendor-supplied updates. VM crashed and other failures, may have been fixed by the vendor.

VM Crashes

If you are experiencing VM crashes, it is possible that a kernel crash dump can help identify the problem. If the crash is reproducible, follow this procedure and consult your guest OS vendor for further investigation on this issue.

Controlling Linux VM Crashdump Behavior

Troubleshooting Linux VM general problemsFor Linux VMs, the crashdump behavior can be controlled through the actions-after-crash parameter. The following are the possible values:

Value Description
preserve leave the VM in a paused state (for analysis)
restart no core dump, just reboot VM (this is the default)
destroy no coredump, leave VM halted

To enable saving of Linux VM crash dumps

  1. On the XenServer host, determine the UUID of the desired VM by running the following command:

    xe vm-list name-label=name params=uuid --minimal
    
  2. Change the actions-after-crash value using xe vm-param-set. For example, run the following command on dom0:

    xe vm-param-set uuid=vm_uuid actions-after-crash=preserve
    
  3. Crash the VM.

    For PV guests, run the following command on the VM:

    echo c | sudo tee /proc/sysrq-trigger
    
  4. Execute the dump core on dom0. For example, run:

    xl dump-core domid filename
    

Controlling Windows VM Crashdump Behavior

Troubleshooting Windows VM general problemsFor Windows VMs, the core dump behavior cannot be controlled by the actions-after-crash parameter. By default Windows crash dumps are put into %SystemRoot%\Minidump in the Windows VM itself.

You can configure the VMs dump level by following the menu path My Computer > Properties > Advanced > Startup and Recovery.

Troubleshooting Boot Problems on Linux VMs

Troubleshooting Linux VM boot problemsThere is a utility script named xe-edit-bootloader in the XenServer host control domain which can be used to edit the bootloader configuration of a shutdown Linux VM. This can be used to fix problems which are preventing it from booting.

To use this script:

  1. To ensure that the VM in question is shut down, run the command

    xe vm-list
    

    The value of power-state will be halted

  2. You can use the UUID as follows:

    xe-edit-bootloader -u linux_vm_uuid -p partition_number
    

    or the name-label as follows:

    xe-edit-bootloader -n linux_vm_name_label -p partition_number
    

    The partition number represents the slice of the disk which has the filesystem. In the case of the default Debian template, this is 1 since it is the first partition.

  3. You will be dropped into an editor with the grub.conf file for the specified VM loaded. Make the changes to fix it, and save the file, exit the editor, and start the VM.