Wasp bot is a constantly running nodejs server that serves different purposes:
- Acting as a Discord bot for our Discord community.
- Sending Wasp analytics reports.
Copy env.example to .env and fill it with your PostHog API key and Discord bot token.
Both are needed to run bot locally, but only PostHog API key is needed to run just analytics locally.
Run npm install for the first time or after any changes in npm deps.
Also, run npm run prepare to set up pre-commit hooks (for stuff like code formatting and linting!).
When developing, run the bot server locally with npm run buildAndStartBot. This will actually connect it to the Discord server, so you will have two bots running at the same time -> production one, and your development one. So test quickly what you need and then shut it down.
Run npm run buildAndCalcAnalytics to run analytics manually and get the report in the CLI.
The bot is deployed to Fly. The fly.toml file contains the configuration for the deployment. The Fly deployoment uses the Dockerfile to build the image.
The bot is deployed automatically on every push to the production branch. You can also deploy with the Fly CLI by running fly deploy.
The Fly.io server on which wasp-bot is deployed has a persistent volume attached to it called wasp_bot_storage mounted at /data dir, in order to persist the cached analytics events from Posthog between deployments.
Our Wasp-bot app provides the WASP_ANALYTICS_CACHED_EVENTS_JSON_PATH environment variable, which in this case we set to point to /data/wasp-analytics-cached-events.json.
You can check production app logs with fly logs and SSH into the app container with fly ssh console.