The project includes a command-line interface tool to check available Platformatic runtimes.
After installing the project dependencies, you can run the tool directly:
./cli.jsTo run the CLI on a custom port, you can just pass it as an argument with ./cli.js --port 4321.
To record a flamegraph session, and choose to profile either the cpu or the heap, you can run the CLI with ./cli.js --record --profile heap. Once you will stop the process, an HTML one-file bundle will be auto-generated, and you will be able to navigate (even offline) the watt-admin app, looking at the flamegraph and at the metrics stored for the whole time you have run the CLI.
The tool is also available as a binary when installed globally or linked:
npm link # Link the package locally
watt-runtime # Run the commandThe tool automatically discovers all available Platformatic runtimes. If multiple runtimes are found, it will prompt you to select one using an interactive menu.
$ watt-runtime
Select a runtime: (Use arrow keys)
❯ runtime-name (PID: 12345) (Started at 3/10/2025, 10:00:00 AM)
another-runtime (PID: 54321) (Started at 3/10/2025, 9:30:00 AM)
If only one runtime is available, it will be automatically selected.
npm install && cp .env.sample .env
npm run build
npm run start
This will auto-reload both the backend and frontend when changes are made.
npm run dev
- check you started
watt-admin - you should now see an info log like
Platformatic is now listening at http://127.0.0.1:{PORT}(open the local URL) - click on link above to navigate the frontend app
- you can call the backend by prefixing the local URL with
/api - call the
/api/runtimesendpoint (you should receive the PID) - call the
/api/runtimes/{pid}/metricsendpoint
