Kestra

Use it when

• Scheduled scripts and data pipelines are scattered across servers as cron jobs with no central visibility, retries, or backfill when one fails
• You need workflows triggered by real-time events such as file arrivals or messages on Kafka, SQS, or Pub/Sub, defined declaratively in YAML and version controlled in Git

What it solves

Not the fit when

• Not a stream processing engine like Flink or Spark Streaming (though it can launch Spark jobs via plugins)
• Runs as a server platform in Docker; not an embeddable library
• Not an AI agent framework
• The README's `server local` quickstart runs as root and mounts `/var/run/docker.sock` plus host `/tmp`; treat it as a privileged local evaluation command, not a production deployment template, and use a hardened deployment from the installation guide for production.
• stream processing engine for continuous data
• LLM agent orchestration
• in-process job queue embedded in application code

Install

docker run --pull=always -it -p 8080:8080 --user=root --name kestra --restart=always -v kestra_data:/app/storage -v /var/run/docker.sock:/var/run/docker.sock -v /tmp:/tmp kestra/kestra:latest server local

Invoke

Open http://localhost:8080, create a flow in the built-in editor as YAML (id, namespace, tasks with plugin types such as io.kestra.plugin.core.log.Log), add a trigger for schedules or events, and run it from the UI.

Alternatives

No reviewed alternatives recorded yet.