Skip to content

Configuration & Deployment

Use this page to wire environment variables, TOML configuration, and deployment workflows for the Genji Shimada bot.

Environment configuration

Runtime settings come from a combination of environment variables and a TOML file loaded at startup.

Core variables

Required environment variables in .env:

  • DISCORD_TOKEN – Required by bot.start(). Obtain from Discord Developer Portal.
  • APP_ENVIRONMENT – Controls the command prefix and which TOML file is loaded:
  • "production" → loads configs/prod.toml, uses "?" prefix
  • Any other value → loads configs/dev.toml, uses "!" prefix
  • API_KEY – Forwarded to the APIService for authenticated requests to the API.
  • API hostnames are derived from APP_ENVIRONMENT (genjishimada-api-dev for development, genjishimada-api for production).

RabbitMQ variables

Required for message queue integration:

  • RABBITMQ_USER – RabbitMQ username
  • RABBITMQ_PASS – RabbitMQ password
  • RABBITMQ_HOST – RabbitMQ host (e.g., localhost or genjishimada-rabbitmq in Docker)

Optional observability variables

For error tracking and monitoring:

  • SENTRY_DSN – Sentry DSN for error tracking
  • SENTRY_AUTH_TOKEN – Sentry auth token (optional)
  • SENTRY_FEEDBACK_URL – Custom feedback URL (optional)

TOML configuration

The TOML schema is defined in utilities/config.py and covers guild, role, and channel identifiers.

Development configuration

Edit apps/bot/configs/dev.toml for development IDs:

[guild]
id = 1234567890  # Your development Discord server ID

[channels]
newsfeed = 1234567890
completions_verification = 1234567890
completions_upvote = 1234567890
playtest = 1234567890
xp = 1234567890
logs = 1234567890

[roles]
admin = 1234567890
moderator = 1234567890
verified = 1234567890

Production configuration

Edit apps/bot/configs/prod.toml for production IDs:

[guild]
id = 9876543210  # Production Discord server ID

[channels]
newsfeed = 9876543210
completions_verification = 9876543210
completions_upvote = 9876543210
playtest = 9876543210
xp = 9876543210
logs = 9876543210

[roles]
admin = 9876543210
moderator = 9876543210
verified = 9876543210

The Genji constructor reads the appropriate file on startup based on APP_ENVIRONMENT.

Local development workflow

  1. Install dependencies: 

    just setup

  2. Configure environment: 

    cp .env.local.example .env.local
# Edit .env.local with your Discord token and settings

  3. Edit development config: 

    # Edit apps/bot/configs/dev.toml with your Discord IDs

  4. Start infrastructure: 

    docker compose -f docker-compose.local.yml up -d

This starts PostgreSQL, RabbitMQ, and MinIO for local development.

  1. Run the bot: 
    just run-bot

The bot automatically loads .env.local and connects to the API at localhost:8000.

  1. Lint before committing: 
    just lint-bot

Docker deployment

Development

If you run the bot inside Docker:

docker compose -f docker-compose.dev.yml up -d genjishimada-bot-dev

Production

Use the production compose file:

docker compose -f docker-compose.prod.yml up -d genjishimada-bot

Deployment checklist

Before deploying to production:

  • Update configs/prod.toml with production Discord IDs
  • Set all required environment variables in production .env
  • Ensure RabbitMQ and the Genji API are reachable
  • Build and deploy the container image (or restart the process)
  • Monitor Discord logs and Sentry events after rollout
  • Verify bot appears online in Discord
  • Test key commands and queue consumers

Observability

Logging

setup_logging() in main.py configures log levels and filters:

  • Filters noisy Discord messages
  • Enables DEBUG logs for internal packages when APP_ENVIRONMENT is "development"
  • Logs to console by default

View logs:

# Local development
just run-bot

# Docker
docker compose -f docker-compose.prod.yml logs -f genjishimada-bot

Sentry

main() initializes Sentry with trace and profile sampling when SENTRY_DSN is set:

import sentry_sdk
from sentry_sdk.integrations.asyncio import AsyncioIntegration

sentry_sdk.init(
    dsn=SENTRY_DSN,
    environment=APP_ENVIRONMENT,
    integrations=[AsyncioIntegration()],
)

Benefits: - Automatic exception capture - Performance traces - User context (Discord user info)

View errors at sentry.io.

Troubleshooting

Bot Won't Start

Check Discord token: 

# Verify DISCORD_TOKEN is set
echo $DISCORD_TOKEN

Check permissions: - Ensure bot has required intents enabled in Discord Developer Portal - Verify bot is invited to the server with correct permissions

Check logs: 

docker compose -f docker-compose.prod.yml logs genjishimada-bot

Queue Messages Not Processing

For local development: 

# Check RabbitMQ is running
docker compose -f docker-compose.local.yml ps rabbitmq-local

# Check bot logs for connection errors (if running with just run-bot)
# Logs will appear in your terminal

# Check RabbitMQ management UI
open http://localhost:15672

For remote deployments: 

# Check RabbitMQ is running
docker compose -f docker-compose.dev.yml ps genjishimada-rabbitmq-dev

# Check bot logs
docker compose -f docker-compose.dev.yml logs genjishimada-bot-dev | grep -i rabbitmq

Check queue consumers: - Inspect bot logs for queue bindings and errors

API Requests Failing

Verify API key: 

# Check API_KEY is set
echo $API_KEY

Check API availability: 

curl -H "X-API-KEY: $API_KEY" http://localhost:8000/healthcheck

Check bot logs: 

docker compose -f docker-compose.prod.yml logs genjishimada-bot | grep -i api

Next Steps