Skip to main content

Deploy with an agent

Copy agent prompt

Before you start

Connect your AWS account and choose a project. Review the upstream deployment instructions for the release you plan to pin. See Kestra runtime and storage configuration.

Data dependencies

PostgreSQL for the repository and queue, plus durable internal storage. See SQL databases for private writer access and connection details.

Deployment order

  1. Create a private PostgreSQL database and wait for the writer. Create a private S3 bucket before starting Kestra if you use S3 internal storage. Check task credentials permit access to that bucket.
  2. Set datasources.postgres.url in KESTRA_CONFIGURATION to the writer JDBC URL. Set its username and password through runtime configuration. Set kestra.repository.type and kestra.queue.type to postgres. Configure kestra.storage.type and the storage backend.
  3. Deploy the pinned Kestra standalone image using the release’s documented command. Configure a task runner compatible with the hosting environment. Use a separate supported execution environment for jobs that require a Docker daemon.
  4. Start Kestra after PostgreSQL and storage are ready. Wait for startup before enabling schedules or connecting clients.
Use Runtime variables for runtime credentials. For a Spring Winter web server, set an app-supported health path that returns HTTP 200 without credentials. Check History before running the application check.

Verify the deployment

Run a simple flow, then verify execution history and read back an output stored by the flow.