The shared data pattern does not address any kind of consolidation of reply, because this is a pattern with the focus on storage. However, we need to access our microservices. To do that, we'll modify some of our Nginx configurations.
We'll totally modify the instances of upstream configuration. As you can see, we modified the upstream name and added more server instances to the direction of our microservices. In this setting, each microservice has four instances:
upstream proxy_servers {
server bookproject_userservice_1:3000;
server bookproject_userservice_2:3000;
server bookproject_userservice_3:3000;
server bookproject_userservice_4:3000;
server bookproject_famous_news_service_1:5000;
server bookproject_famous_news_service_2:5000;
server bookproject_famous_news_service_3:5000;
server bookproject_famous_news_service_4:5000;
server bookproject_politics_news_service_1:5000;
...