ESXi 5 Purple Screen of Death in VMware Workstation 8

While I’m traveling, I carry a beefy laptop as a mobile test lab. I fired it up recently after several weeks of no lab work and to my chagrin, looked up and saw the dreaded purple screen of death for the first time. I saw this as a good opportunity to research this particular PSoD and post my findings.

Interpreting the Errors

I was able to find an excellent VMware knowledge base article on how to interpret purple screen errors. You can find it here: Interpreting an ESX host purple diagnostic screen

When I returned to make Workstation the active window, I was confronted with the dialogue box below, glowing purple from what awaited me in the background.

Hmm. So from line 2 of the screen shot below, we can see that physical CPU 0 (pCPU 0) had no heartbeats. Then a bit further down the message, we can see that pCPU 0 “didn’t have a heartbeat for 2728 seconds,” and that it “*may* be locked up.” This tells me that the virtualized ESXi box was basically unresponsive for about 45 minutes.

What’s the problem?

I couldn’t find anything specific that told me exactly what the problem was, but I came to the conclusion that it was the laptop power settings causing the issue. When working in the test lab in the past, I was usually active in the lab – never letting the computer go into power saving mode. Apparently, I left the laptop idle too long and it went to sleep and stopped the hard disks.

I’m using Windows 7 Ultimate 64-bit. You can change the power settings easily by going to Control Panel > Power Settings > Change when the computer sleeps, and changing the appropriate options to never put the computer to sleep.

Google