Skip to content
Docs Pydantic Docs

Fleet: manage Collectors and SDKs from Logfire | Logfire

Change what your OpenTelemetry Collectors and applications collect, and where they send it, from Logfire: no config files to edit on each machine and no redeploys.

Your telemetry reaches Logfire from two kinds of software, and Fleet manages both:

  • Collectors: an OpenTelemetry Collector is a program that runs on your machines, receives or gathers telemetry (host metrics, log files, data from your applications), processes it, and sends it on. Fleet manages Collectors on hosts (servers, virtual machines, laptops) and in Kubernetes clusters. See Manage Collectors.
  • Applications: the SDK inside your application decides what it records and exports. Fleet changes its settings while it runs: sampling, minimum log level, redaction and scrubbing, and which signals to export. See Manage applications.

Each Collector and application takes its settings from a policy: a versioned set of settings you edit and publish in Logfire. A Collector policy holds the Collector’s whole pipeline. An SDK policy holds the SDK settings it sets, and leaves the rest to the application. A Collector whose configuration is managed elsewhere, such as by Helm or GitOps, shows in Fleet but keeps that configuration. See Change policies.

Clients connect to Fleet over the Open Agent Management Protocol (OpAMP), the OpenTelemetry standard for managing telemetry software remotely. Each client reports its health and the settings it is running, so Fleet also shows what runs where.

Find Fleet in Logfire

Open Fleet management in your project’s sidebar. The Hosts, Kubernetes and Services pages also link to it.

The Collectors tab of Fleet management, listing each Collector with its connection, health, policy status and policy

The page has four tabs:

  • Overview: Flows draws where your Collectors receive data and where they send it. Traffic compares how much each Collector and policy sent with the period before.
  • Collectors: every Collector on your hosts and in your clusters, with its connection, health, the policy it runs and whether that policy applied. Add Collector connects a host or a cluster.
  • SDKs: every application connected to Fleet. Add application connects one.
  • Policies: every Collector and SDK policy. Add policy creates one, and Rollouts appears here when you turn rollouts on.

How Fleet works opens a short tour of these ideas in the product. Settings opens your project’s Fleet settings: connection keys and the Use rollouts switch.

When to use Fleet

Use Fleet when you run Collectors or applications on more than a few machines and want to change them from one place: lower a sampling rate during an incident, stop sending debug logs, or add a log file to every host, then see which clients applied the change.

You don’t need Fleet to send data to Logfire. An application that calls logfire.configure(), or a Collector you configure yourself, sends data without it. Fleet changes only the clients you connect to it.

Limits

Your plan sets how many applications and Collector policies your organization can manage. Fleet shows your usage and the limit when you reach it.

Next steps