Mastodon instances can slow down or become unresponsive without warning. You might notice the federated timeline loading slowly, posts failing to send, or the web interface timing out. These issues often stem from server resource strain, database bottlenecks, or failing background jobs. This article explains how to use Mastodon’s built-in health check endpoints and admin tools to audit instance performance and identify problems before they affect users.
Health checks provide a structured way to monitor your instance. Mastodon exposes several endpoints that report the status of critical services: the web server, database, Redis cache, and Sidekiq job queues. By running these checks regularly, you can catch configuration errors, resource exhaustion, and service outages early. This guide covers the main health checks, how to interpret their output, and what actions to take when a check fails.
You will learn how to access the health check endpoints, understand the status codes and JSON responses, and set up basic monitoring with command-line tools or scripts. The steps apply to self-hosted instances running Mastodon v3.5 or later. If you administer an instance on a managed host, check whether the provider exposes similar endpoints through their dashboard.