We are seeing a number of crashes, either "#PF Exception 14" or "Recursive Panics w/ VSCSI timeouts" on the ESXi 6.0 hosts running in VMs in on 5.5 VMs (VM Version 9).
In 5 out of 5 cases I looked at, there was a VM migration slightly before the crash. I'm guessing that the 6.0 host has short I/O timeouts for swapping, etc., and those are not met on migration.
Is there any way to reduce the frequency of these events?
Thanks
PS I've also observed these messages just before some of the crashes, and I couldn't find any references to this message anywhere.
2016-03-23T04:01:02.023Z cpu0:323721)WARNING: HeapMgr: 1189: Unable to map MPN ranges for heap growth.