Introduction
Midil is the control plane for event-driven systems. It traces every event from the moment it's produced to the moment it's fully handled, keeps a record of anything that failed, and gives your team the controls to act on what it finds. Pause a consumer, throttle it, or dig into why something landed in the dead-letter queue, all from one dashboard.
Who this is for
If your product is built on queues, topics, or webhooks (think orders fanning out to fulfillment, billing, and notifications, background jobs, or inter-service events), Midil gives you and your team a shared, real-time view of what's flowing through that system and what's going wrong.
What you can do
- Trace a single event across every service that touches it, from the first producer to the last consumer.
- Investigate failures: dead-lettered events, duplicate deliveries, and open incidents, grouped so you can spot patterns instead of reading logs line by line.
- Act: pause, throttle, or drain a consumer that's misbehaving, and get alerted the moment a metric crosses a threshold you define.
How it fits together
Two things work together:
- The dashboard: where you and your team look at events, traces, alerts, and incidents, and operate consumers day to day.
- An SDK: a small library you add to your own services so they report what's happening to Observatory. The Python SDK is available today, and more languages are on the way.
Where to go next
- New to Midil? Start with the Quickstart.
- Want the vocabulary first? Read Core Concepts.
- Already instrumented and just want to use the dashboard? Jump to Using the Dashboard.