Skip to content
Plume docs

Running Plume

Plume is a single server process. There are no subcommands: running the binary starts the HTTP server, and everything else happens inside that one process. There is no plume serve, plume migrate, plume import, or plume worker.

What happens on startup

Section titled “What happens on startup”

When you run Plume (the binary or docker compose up), it:

  • Starts the HTTP server on port 8080.
  • Auto-runs the embedded goose database migrations.
  • Bootstraps the admin user from the environment (see Configuration).
  • Launches the send worker and the automation worker as in-process goroutines.

CSV import is not a command — it’s the API endpoint POST /api/lists/{listId}/import (or the importer in the UI). See the REST API reference.

How to run it

Section titled “How to run it”
Section titled “Docker (recommended)”
Terminal window
docker compose up -d

Prebuilt binary

Section titled “Prebuilt binary”

Download a release binary, set the environment variables, and run it:

Terminal window
./plume

From source (development)

Section titled “From source (development)”
Terminal window
go run ./cmd/plume

Configuration

Section titled “Configuration”

All configuration is through environment variables — there is no config file and no .env.example. See Configuration for the full list of PLUME_* variables.

Health check

Section titled “Health check”

Plume serves a health endpoint at GET /health, suitable for load-balancer and container health probes.