VMX, SVM, or other CPU flags missing after updatin

kongjx

VMX, SVM, or other CPU flags missing after updating Xen
This document (7001447) is provided subject to the
disclaimer
at the end of this document.
Environment
Novell SUSE Linux Enterprise Server 10 Service Pack 2
Novell SUSE Linux Enterprise Desktop 10 Service Pack 2
Novell ZENworks Orchestrator
PC (x86-32 or x86-64) hardware supporting full hardware virtualization (HVM).
SituationChanges
The Xen hypervisor (package xen) and/or the Xen-enabled kernel (packages kernel-xenpae and kernel-xen) were updated on a system running a SLE10 SP2 based product.
Symptoms

• The CPU capabilities stated in /proc/cpuinfo are missing flags which have been reported before, in particular the vmx flag (Intel CPUs), or the svm flag (AMD CPUs) is missing.
  
• It is not possible to install a fully virtualized system on appropriate hardware.
  
• When Xen live migration is set up, live migration may fail.
  
• When ZENworks Orchestrator is used, virtual machine images which require full virtualization may be marked as unavailable, or may be unable to migrate, with a ERROR: Host lacks HVM support being reported.
  

Resolution
This issue is fixed as of version 2.6.16.60-0.33 of the kernel-xenpae and kernel-xen packages.
Additional InformationLive migration - detailed symptoms
When live migration of a Xen VM fails due to this issue, a traceback similar to the following may be logged on the node on which the VM was running:
[2008-11-19 07:08:34 8011] INFO (XendCheckpoint:373) All memory is saved
[2008-11-19 07:08:34 8011] INFO (XendCheckpoint:373) Save exit rc=0
[2008-11-19 07:08:35 8011] ERROR (XendCheckpoint:144) Save failed on domain
vm-hagbard (22).
Traceback (most recent call last):
  File "/usr/lib64/python2.4/site-packages/xen/xend/XendCheckpoint.py", line 122, in save
    write_exact(fd, buf, "could not write device model state")
  File "/usr/lib64/python2.4/site-packages/xen/xend/XendCheckpoint.py", line 43, in write_exact
    raise XendError(errmsg)
XendError: could not write device model state
and messages similar to the following being logged on the node onto which the VM was to be migrated:
[2008-11-19 07:08:34 8251] INFO (XendCheckpoint:373) Received all pages (0 races)[2008-11-19 07:08:34 8251] INFO (XendCheckpoint:373) ERROR Internal error: error setting the HVM context[2008-11-19 07:08:35 8251] INFO (XendCheckpoint:373) Restore exit with rc=1
Document
Document ID:
7001447
Creation Date:
09-29-2008
Modified Date:
03-03-2009
Novell Product:
SUSE Linux Enterprise Desktop
Novell Product:
SUSE Linux Enterprise Server
Novell Product:
ZENworks Orchestrator
Disclaimer
The Origin of this information may be internal or external to Novell. Novell makes all reasonable efforts to verify this information. However, the information provided in this document is for your information only. Novell makes no explicit or implied claims to the validity of this information.
Any trademarks referenced in this document are the property of their respective owners. Consult your product manuals for complete trademark information.
http://www.novell.com/support/search.do?cmd=displayKC&docType=kc&externalId=7001447&sliceId=1&docTypeID=DT_TID_1_1


本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u/8549/showart_1859896.html