Memory usage for Openshift Microservices/Pod

Products

• CA Application Performance Management Agent (APM / Wily / Introscope)
• CA Application Performance Management (APM / Wily / Introscope)
• INTROSCOPE
• DX Application Performance Management

Issue/Introduction

• Facing issues with data collected by agent for GC per Openshift Pod.
• In the APM Investigator/Webview the memory utilization for the Pod is going up to 10+ GB.
• But the maximum memory allocation per pod is up to 8GB only.

 

 

Environment

• Release: 10.7.0
• Component: DX SaaS .NET Agent

Cause

GC Heap Bytes in Use metric is based on System.GC.GetTotalMemory(false) and GC Heap Total Bytes metric is based on 

System.Diagnostics.Process.GetCurrentProcess().PrivateMemorySize64.

Below are the links to their Microsoft API docs respectively:

GC.GetTotalMemory(Boolean) Method

Process.PrivateMemorySize64 Property

 

 

Resolution

• Per Engineering, working as designed.

Additional Information

As for monitoring memory usage by the pod and thread usage by the process, you may consider looking at the container performance metrics as part of the UMA docker monitoring and other process monitoring metrics instead of the GC Heap metrics and the thread pool metrics being used currently.