README.md

Chat server deployment guide

This directory contains the files, configurations and scripts.

Setup

• Clone the repository using git.

  Note: for the docker based deployement instead of cloning the complete repository, use sparse-checkout.

  To clone only the deployment directory.

    # Create an empty directory on the server using 
    mkdir <dir-name>
    cd <dir-name>
    # Initialize git inside the directory
    git init
  
    # Enable sparse checkout
    git config core.sparseCheckout true
  
    # Add deployment directory path for checkout
    echo deployment >> .git/info/sparse-checkout
  
    # Add remote origin in local git repo
    git remote add -f origin https://www.github.com/ramank775/chat-server
  
    # Pull the files
    git pull origin master

• Create configuration/enviroment files

  • Create .env file from the sample .env.tmpl file.
  • Create nginx.conf file from the provided sample file based on the setup your are running at config/nginx directory.
  • Copy service.json file of firebase project at config/firebase directory.
  • Create services.json file from the provided sample file at config/discovery_service. Presently for service discovery of the gateway server a file based service discovery is used. For every gateway server add an entry in the services.json file

    {
      [gateway-name-1]: 'http://endpoint-gateway-1:port'
      [gateway-name-2]: 'http://endpoint-gateway-2:port'
      ...
      [gateway-name-n]: 'http://endpoint-gateway-n:port'
    }
    

    Note: gateway name should be same as --gateway-name=gateway-1 option supplied to start the gateway server.

Docker based setup (Recommanded)

• Create docker-compose.local.yml file from docker-compose.infra.yml and change the neccesary options as per the requirement
• Start the services using docker-compose -f docker-compose.yml -f docker-compose.local.yml up -d