Best practice of log query to monitor the health of VMs

Occasional Contributor


Is there a best practice or sample log query of log analytics to monitor the health/performance of virtual machines? I know there are many articles on how to create or use existing log queries but I'm looking for a ready log queries to monitor the health of Azure VMs (like in metrics we have the default metrics of CPU, Disk, Memory, Network with their recommended threshold).


Any advice ?




2 Replies
The challenge with this is normally the "recommended threshold" part. No two computers are the same, so for one it maybe ok to use 90% CPU - for the next just 10%

@Clive_Watson thank you for your reply and you are right about the threshold. But my question was more about the best way to monitor/alert the health of Azure VMs. for example, I can monitor the VMs using platform metrics or guest metrics using diagnostic logs. Or send the OS metrics to log analytics workspace and use the log query or metrics ( in case of individual VM). Also I could enable VM insights and use its default health monitoring and alerts ... so many options. I'm wondering which once I should use ?