Instrumentation Demo App

This is a demo app that we will use to complete challenges for the software instrumentation course.

About this project

This project is a gRPC server, written in Go, that provides few endpoints to manage courses catalog and bookings for a fake online course marketplace. This project uses library named gRPC-gateway which provide a RESTful API endpoints proxy. So, while the implementation is actually a gRPC server, you can also access it by using curl or any other way you used to.

Getting Started

Pre-setup

I have prepared some basic setup here including the following:

• Docker compose file to start few dependencies.
• Logger initialization which will write the logs to stdout and file located on logs/app.log. Thus, you can just directly use zerolog to log your application. You can find the initialization on server.go.

Running the application

1. Make sure you have terminated all running containers from other projects if any. Run docker compose down on the other project if necessary.

2. Set these environment variable in your profile (e.g. ~/.bashrc):

   export REDIS_PASSWORD=<set to password you like>

3. From this project directory, start all dependencies with docker compose for this project:

   make bootstrap
   docker compose up -d

   Make sure you have make installed. If not, run apt install make or whatever you need to do to install make on your machine.

   If you get error when running make bootstrap because some dependencies are missing, just install them. For instance, if you got this following errors:

   no required module provides package google.golang.org/grpc/cmd/protoc-gen-go-grpc; to add it:
       go get google.golang.org/grpc/cmd/protoc-gen-go-grpc
   

   In this case, run go get google.golang.org/grpc/cmd/protoc-gen-go-grpc.

   Please check the docker-compose.yml file to see what services are started.

4. To start this API server, run:

   make course/server

   You should see that HTTP server starts at port 8800 and gRPC server starts at port 9900.

5. Seed the database by running the following command from this project directory

   make course/seed

   If necessary, you may update the data after it is seeded to database. Or you can truncate the database and re-seed it if necessary.

6. Check out list of available APIs from the swagger docs. Go to http://localhost:8800/swagger. You can try out the API from there as well if you want.

Running load generator

Load generator here is used to test your implementation.

While you may see the implementation of the load generator, I suggest you to not do that. Instead, you should treat the load generator as a black box so that you can fully understand how the API clients interacting with your API only by using instrumentation you set.

1. Make sure you have python 3.x installed.

2. Install virtualenv

   pip install virtualenv

3. Go to locust directory under scripts directory. Crete and activate a virtual environment

   cd scripts/locust
   virtualenv .venv
   source .venv/bin/activate

4. Install the required dependencies for the load generator

   pip install -r requirements.txt

5. Run the load generator

   locust -f locustfiles --class-picker --modern-ui -H http://localhost:8800

6. Open your browser and go to http://localhost:8089. You should see the locust dashboard. There are two users scenarios:

   • GeneralUser is the scenario that you can use to test several APIs. Use this for most of the time when working on the final challenges. You do not need to spawn a lot of users to test this scenario. One or just a few users are enough.

   • CompetingUser is the scenario that you can use later when working on the tracing final challenge. The scenario is designed to simulate a competition between users when making the reservation. You need to spawn a more users to test this scenario to see how it might impact the performance of the system.

7. To start the load generator, go to http://localhost:8089, fill the parameter as you need, and click Start Swarming.