GoFib-Docker-APM-Sandbox

Dockerized Fibonacci Calculator

This sandbox spins up a Go api that interacts within a containerized environment and communicates MySQL, Redis, and can be utilized via the command line or using the React client application.

The code is made up of 3 separate subprojects which run on separate containers and is meant to help illustrate tracing, span generation, tagging, and correlated logs in the Datadog APM.

What's containerized:

• Datadog Agent enables APM and Logs
• Go API server
  • Debug mode optional
  • Automatic instrumentation traces API calls
  • Includes Redis client that publishes requests to Redis
• Go DB worker
  • Debug mode optional
  • Manual instrumentation traces work done by the server
  • Subscribed to Redis and calculates Fibonacci values
• React client app
  • Provides a UI to easily calculate, delete, and view all values
• Redis server
• MySQL server

Sections

• Step 1: Create a custom .env file
• Step 2: Build images
• Step 3: Spin up containers
• Step 4: Curling endpoints
• Step 5: Spin down containers

Step 1: Create a custom .env file

Make sure that in your ~ directory, you have a file called sandbox.docker.env that contains:

DD_API_KEY=<Your API Key>

Step 2: Build images

1. Add your API Key to docker-compose.yml
2. Have Docker running, then

docker-compose build --no-cache

Step 3: Spin up containers

docker-compose up

Step 4: Curling endpoints

You can use the browser to access these endpoints. Here are the endpoints you can use if you want to use the terminal:

1. GET a list of all values

curl localhost:80/api/all

2. POST a new value

curl -X POST localhost:80/api/fib -d '{"value": <fib-value>}'

3. DELETE a value

curl -X DELETE localhost:80/api/fib/<fib-value>

4. GET a value

curl localhost:80/api/fib/<fib-value>

At this point, you should have traces in the Datadog UI outlining how some recursive functions can have overlapping subproblems and how to optimize them.

Step 5: Spin down containers

docker-compose down