We have an 8 node cluster running Server 2016 Data center configured for Hyper V that has been in place for the past year. During this time we have experienced random OS reboots across all nodes. Idrac simply reports an OEM S event and event logs an unexpected shutdown. Based on other user threads of a similar nature the following changes have been made to system profile but to no avail. CPU: Max performanceMemory Frequency: Max PerformanceC1E: DisabledC States: Disabled Server drivers and firmware are up to date.Dell have analysed our logs and have been unable to identify any hardware issues. Hyper V is integral to our business and we are desperate to resolve and stabilise, to the point we are considering buying new hardware and rebuilding. Any suggestions would be gratefully received.
With it affecting all 8 nodes, I wouldn't expect it to be a hardware failure. Have you tried running Microsoft's BPA to make sure your design conforms to their recommendations?