Documentation previewThese docs are actively being built. Some pages may change as the framework and examples are finalized.
GoForjv0.9

Deployment Basics ​

A generated GoForj App is a Go application with explicit runtime commands.

Deployment is mostly deciding which binary to run, which runtime commands to start, and which environment variables select infrastructure.

Build ​

Build the App:

bash
forj build

forj build refreshes generated components, runs Wire, indexes APIs, and builds the binary.

Runtime Commands ​

Common production commands:

bash
./bin/app run
./bin/app api
./bin/app worker
./bin/app scheduler
./bin/app migrate

Use run for standalone topology. Use specific runtime commands when HTTP, workers, and scheduler should be separate processes.

Environment ​

Set runtime environment through process env, .env files, or your deployment platform.

Important areas:

  • APP_ENV
  • APP_SHUTDOWN_TIMEOUT
  • RUNTIME_MODE
  • HTTP host and port
  • supported and selected drivers
  • secrets
  • metrics ports
  • readiness token

Deployment Order ​

Typical order:

  1. build the App
  2. configure environment
  3. run migrations when database schema changed
  4. start runtime processes
  5. verify health and readiness
  6. verify metrics and logs

Common Mistakes ​

Common mistakes

  • Do not deploy a binary after changing generated drivers without running forj build.
  • Do not run every runtime in one process just because local development does.
  • Do not start multiple scheduler processes unless the schedules support it.
  • Do not make required configuration appear optional.

Next Steps ​