经过几轮测试，

今天终于在公司上海机房正式升级主ESX服务器。十多个SERVER运行在上面，就算先有重要SERVER的备份，也要不容有失了。

开始VCENTER5.1的UPDATE MANAGER一切顺利。相关ESX5.1的ISO拷贝就位。

执行REMEDIATE。。。

服务器重启。紫红屏幕故障。PSOD。

http://enterpriseadmins.org/blog/virtualization/vmware-esxi-psod-on-dell-server/

http://www.perfectcloud.org/fix-it/dell-esxi-psod-lint1-motherboard-interrupt/

Last week I was informed of a customer who had a Purple Screen of Death on a Dell PowerEdge R810 running ESXi 4.1 Update 1.  The stop screen showed the following reason:

LINT1 motherboard interrupt. This is a hardware problem: please contact your hardware vendor

After working with Dell tech support, the customer was directed to disable the C States and C1E settings in the BIOS.  I was interested in this setting as I have a cluster using the same hardware with the same version of ESXi.

The following article describes C States and specifically the C1E setting; it is a method to reduce power consumption by powering off cores when not in use:  http://www.delltechcenter.com/page/Impact+of+C1E+on+PowerEdge+11G+Servers+–+HPBD+100909. 

The VMware Performance Best Practices Guide (available here: http://www.vmware.com/pdf/Perf_Best_Practices_vSphere4.0.pdf) specifically states on page 15 to “Disable C1E halt state in the BIOS.” 

I don’t spend a lot of time changing settings in the BIOS, but with the possible impact of this setting thought this was worth sharing.

 

这个CASE和公司的较为靠近。DELL R815与810的BIOS也不是完全一样的。只有C1E setting，没有 C States 。DISABLE掉这个PROCESSOR的特性。然后，OK啦。

好惊险。。。。