Build and deploy

We will build, deploy and scale using the docker-compose command

docker-compose up --build --scale master=2

This will tell docker-compose to :

  • Read the docker-compose.yml file
  • Build all the container images
  • Run all the containers in detached mode
  • Scale the master image to 2 instances

Start all containers

PS C:\Users\aniru\workspace\github\python-nginx-microservice> docker compose up
[+] Running 5/5
 - Network python-nginx-microservice_default      Created                                                                                                                                          1.1s 
 - Container python-nginx-microservice-news-1     Created                                                                                                                                          0.8s 
 - Container python-nginx-microservice-master-1   Created                                                                                                                                          0.8s 
 - Container python-nginx-microservice-weather-1  Created                                                                                                                                          0.8s 
 - Container nginx                                Created                                                                                                                                          0.2s 
Attaching to nginx, python-nginx-microservice-master-1, python-nginx-microservice-news-1, python-nginx-microservice-weather-1

We can now test if the containers are running. 

PS C:\Users\aniru\workspace\github\python-nginx-microservice> docker compose ps
NAME                                  COMMAND                  SERVICE             STATUS              PORTS
nginx                                 "/docker-entrypoint.…"   nginx               running             0.0.0.0:80->80/tcp
python-nginx-microservice-master-1    "python master_assis…"   master              running             3001/tcp
python-nginx-microservice-master-2    "python master_assis…"   master              running             3001/tcp
python-nginx-microservice-news-1      "python news.py"         news                running             0.0.0.0:3003->3003/tcp
python-nginx-microservice-weather-1   "python weather.py"      weather             running             0.0.0.0:3002->3002/tcp

As you can see, this indicates :

  • nginx service is running as nginx on port:80
  • master service is running as python-nginx-microservice-master-1 and python-nginx-microservice-master-1 on internal port:3001
  • news service is running as python-nginx-microservice-news-1 on port:3002
  • weather service is running as python-nginx-microservice-weather-1 on port:3003

Test the application

Now we will test the application by invoking 3 URLs:

  • Index page : In this case the following code is executed in the master service Screenshot

  • News page : We will now query all news for the day in US 

http://localhost/news?country=us

Let's check the logs by using command docker logs <container-name> -f

PS C:\Users\aniru\workspace\github\python-nginx-microservice> docker logs python-nginx-microservice-news-1 -f
 * Serving Flask app "news" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: on
 * Running on http://0.0.0.0:3003/ (Press CTRL+C to quit)
 * Restarting with stat
 * Debugger is active!
 * Debugger PIN: 329-461-268
172.28.0.3 - - [18/Jan/2023 20:22:42] "GET /news?country=de HTTP/1.1" 200 -
172.28.0.3 - - [18/Jan/2023 20:55:28] "GET /news?country=us HTTP/1.1" 200 -

Screenshot

  • Weather page : 
http://localhost/weather?city=chicago

Let's check the logs by using command docker logs <container-name> -f

PS C:\Users\aniru\workspace\github\python-nginx-microservice> docker logs python-nginx-microservice-weather-1 -f
 * Serving Flask app "weather" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: on
 * Running on http://0.0.0.0:3002/ (Press CTRL+C to quit)
 * Restarting with stat
 * Debugger is active!
 * Debugger PIN: 858-141-160
172.28.0.2 - - [18/Jan/2023 20:22:45] "GET /weather?city=amsterdam HTTP/1.1" 200 -
172.28.0.2 - - [18/Jan/2023 20:59:38] "GET /weather?city=amsterdam HTTP/1.1" 200 -

Screenshot

Test the load-balancer

Open two separate terminals and run:

docker logs python-nginx-microservice-master-1 -f
docker logs python-nginx-microservice-master-2 -f

Now invoke these URLs in random while observing the logs: