You can configure CPU rules to temporarily mitigate a high CPU issue until the real cause for the unexpected issue is found.

Once you are inside Diagnostic Tools, click Proactive CPU Monitoring.

Proactive CPU Monitoring offers 4 different modes listed below.

The size of the dump generated is also directly proportional to the memory consumed by the process.

The Maximum Actions is not honored and monitoring stops automatically after Maximum Duration.

Kill is a forceful termination of the process and not a graceful exit.

All requests that the current worker process is processing will be terminated, and end users may see 502 errors.

If system terminates the child process, the error message may even be 500 or some form of application exception based on the stack.

No analysis is performed but after the session finishes, you have an option to analyze the memory dump after the session ends by clicking the Analyze button.

In addition, CPU dumps are analyzed by DumpAnalyzer tool, and an analysis report is generated.

Analyzing a dump with this tool has some CPU impact on the instance and takes roughly 5-7 minutes at minimum to analyze one dump file.

For example, if there are 10 instances and say the monitoring collected 5 dumps, all instances in parallel can analyze memory dumps.

If there is 1 instance and 5 dumps are collected, the dump analysis will happen one after the other.

This new experience, currently in preview, deploys a PostgreSQL or Azu.

Appx High Cpu