Exposing the Desired Metrics

Explore how to expose Go runtime metrics such as goroutines count and memory usage to Prometheus for effective monitoring. Learn to implement HTTP handlers that serve these metrics, use Go modules and external packages, and run the application with Docker. This lesson guides you through running and testing the metrics endpoint locally and preparing the environment for Prometheus integration.

We'll cover the following...

Coding example
Reading metrics
- Let’s try our code
Enabling Prometheus
- Dockerfile
- Let’s build our Dockerfile

This lesson illustrates how to expose metrics from the runtime/metrics package to Prometheus. In our case, we use /sched/goroutines:goroutines and /memory/classes/total:bytes. We already know about the former, which is the total number of goroutines. The latter metric is the amount of memory mapped by the Go runtime into the current process as read-write.

Note: Because the presented code uses an external package, it should be put inside ~/go/src and Go modules should be enabled using go mod init.

Coding example

The Go code of prometheus.go is as ...

1.Course Introduction

2.Introduction to Go

3. Basic Go Data Types

4.Composite Data Types

5.Reflection and Interfaces

6.Go Packages and Functions

7.Telling a UNIX System What to Do

8.Go Concurrency

9.Building Web Services

10.Working with TCP/IP and WebSocket

11.Working with REST APIs

12.Code Testing and Profiling

13.Working with gRPC

14.Go Generics

15.Conclusion

16.Appendix A—Go Garbage Collector

Exposing the Desired Metrics

Coding example