Discord Bot Boilerplate

Pre-requisite

Bot Creation

• Create Discord Bot Application
• Add Bot at 'Bot' tab
• Toggle 'Public bot' to off
• Copy Token

Join bot to server

• Select 'URL Generator' at 'OAuth2' tab
• Under 'Scopes', select bot and application.commands
• Select Send Messages Embed Links Read Messages/View Channels in 'Bot Permissions' selections
• Copy URL into the browser

Create MongoDB

• Create any MongoDB atlas
• Make sure to whitelist the Cloud Run IP

Environment Values

• Create .env file
• Copy and Paste, then fill variables

BOT_TOKEN="Bot Token ID"

MONGO_CLUSTER="mongodb+srv://<username>:<password>@(URL of MongoDB Cluster)"
MONGO_USER="MongoDB Username"
MONGO_PASSWORD="MongoDB Password"
MONGO_DB_NAME="MongoDB DBName"

APPLICATION_ID="Application ID of Discord Application"
GUILD_ID="Channel ID that can be found in Discord Settings Tab. Developer setting must be on in your personal settings (Only for private bot)"

DISCORD_CHANNEL_ID_MAIN="Channel ID that the bot will send a message (Only for sending message from bot)"

• remove from .gitignore if it needs to be run on cloud

Run

• yarn install
• for build, yarn build, or just yarn start

GCP with Cloud Run using Github Action (Option - Est. Minimum cost: $50/Month)

• IMPORTANT: You MUST enable "Always-On" CPU allocation to use this method, because Discord bot does not start HTTP server that listens on an HTTP port (8080)
• Enable Artifact Registry API Cloud Run API
• Create Service Account with Cloud Run Service Agent role
• Create Key (JSON) and make it as Secret(Action) at github repo settings. Then name it GCP_SA_KEY
• IAM > Add roles for the service account: Artifact Registry Administrator Cloud Run Admin
• Edit cd_production.yml for Continuous Deployment. Replace PROJECT_ID CLOUD_RUN_NAME GCP_REGION
• Create Repository of Artifact Registry and name it as CLOUD_RUN_NAME value and select region that matches to GCP_REGION value.

GCP with Compute Engine using Github Action (Option - Est. Minimum cost: $7~/Month)

• IMPORTANT: You MUST create Cloud Storage in order to use terraform setting and MUST match bucket and prefix at main.tf
• IAM > Add roles for the service account
• 
• Follow above Cloud Run settings to create credential file for GCP_SERVICE_ACCOUNT secret.
• Follow above Artifact Registry setting.
• Edit cd_gcp_ce_terraform.yml for Continuous Deployment. Replace variables.
• Create Compute Engine in GCP
• Set Docker option

Digital Ocean Droplet with Github Action (Option - Est. Minimum cost: $4~$10/Month)

• Create Container Registery (free version might work, but 5G is recommended as garbage collector will take the space)
• Create Droplet with SSH Authentication option
• Connect to Droplet console, then create ssh key using below commands (More info)
  ssh-keygen -t ed25519 -a 200 -C "[email protected]"
cat .ssh/id_ed25519.pub > .ssh/authorized_keys and copy this text into Settings > Security tab > add SSH Key on the left down corner at DO dashboard.
  cat .ssh/id_ed25519 and copy the text as github secret SSHKEY
• Create github secrets
  HOST is droplet ip
  DIGITALOCEAN_ACCESS_TOKEN can be created at API menu of DO dashboard
  PASSPHRASE is what you entered as passphrase for ssh creation
  USERNAME is droplet username. usually root
• Change env values in yml file
• Copy / Paste any .env into the Droplet
• NOTE: Setting env file needs some efforts than GCP.