violet BSOD on ESXI 5.5 Host

[Mark2000]

Hello, I have a Big Problem 😞 About 10-15 minutes i get an violet BSOD on my ESXI 5.5 Host for the first Time... 😞 What can i do? I dont have any Idea.. the server has not yet restarted.. Here is a Screenshot from the HP ILO Remote Management. Can somebody help me? - What is the Problem? Where does the Problem come? Sry for my bad english - i am a German User 😃 Best Regards, Mark

[JarryG]

Maybe this can help you a little (though PSODs are not easy to decipher):

VMware KB: Interpreting an ESX/ESXi host purple diagnostic screen

VMware KB: Understanding a Failed to ack TLB invalidate purple diagnostic screen

BTW, the first few lines are the most important, so nex time please try to upload complete screenshot.

[Wh33ly]

As far as I can see through the bar I think the message is  :

PCPU1 locked up. Failed to ack TLB invalidate. (Total of 2 locked up, PCPU(s). 1,2)

[Mark2000]

thanks for the fast answer! i wil check this.. here is a full screenshot - (sry):smileyblush:

[Mark2000]

@wh33ly - what does this mean? or what can i do? :smileyconfused:

[JarryG]

Read please the second link I posted. Exactly this message is discussed there...

[SATHISHVIJAY]

Hi Mark, have you gone through the Vmkernal log file and found out which has caused this PSOD.

Hi Jarry, could you tell us what are the possible causes for this PSoD and their Troubleshooting & FIXES.

Also, Do anyone come across the below one:

u

[CedricAnto]

It appears like you have a faulty hardware. NMI are are interrupts triggered in an attempt to regain control of your CPU that have been non-responsive, sometimes they dont respond to NMIs as well, these typically manifest as PCPU locked up. We also see PCPU locked up as 1 & 2(most likely one dual core CPU out of order), this is stronger indication that one of your CPU/socket that may be faulty.

Run thorough hardware diagnostics. Dont add this server in production and ensure the server is part of the VMware Hardware compatibility list

[Wh33ly]

I agree with virtual_knight, one out of some new machines we had delivered got this error had indeed some hardware failures. Luckily we could return the unit as DOA and was confirmed afterwards that the machine had hardware related problems.

[Rubeck]

Also, Do anyone come across the below one:

u

Yes... I have.

Called HP... The support guy said "LINT 1 motherboard interrupt, eh?. New motherboard it is then" A guy showed up 2 hours later and replaced the board.... no problem since.

This was on a DL380p Gen8 too..

/Rubeck