Over the last few years i'm experiencing more and more error or warning type events in the logs that do not point at actual problems but according to support "can safely be ignored". Not needing to spend time searching the internet or calling techsupport would save time & money.
If Microsoft knows of those why should the Windows event system not?

A cloud service that tells the admin of on-premise servers and/or IaaS VMs what’s wrong and how to remediate the problem. It would do this by finding patterns in the machine's diagnostic data (Event Logs, ETL logs, txt logs, etc.). This is not a monitoring scenario, but rather a deep analysis scenario.

help

When something breaks or goes wrong, what are your biggest pain points for finding a resolution/root cause? I'm thinking about process, people, or technology.