API Reference

The Octopus gateway exposes a handful of HTTP APIs alongside the traffic it proxies. They fall into three groups:

Admin REST API01

The Admin API (served by the octopus-admin crate) is the management surface: list and inspect discovered services, routes, and upstreams; read and reload configuration; and query analytics. It is documented in full — every endpoint, request, and response — on the Observability page.

It is not repeated here. See Admin REST API.

FARP Registration API02

The FARP (Forge API Gateway Registration Protocol) API is how services tell the gateway they exist. Services push a manifest to register, send periodic heartbeats, and deregister on shutdown; the gateway fetches their schemas, generates routes, and serves a federated API surface (OpenAPI / AsyncAPI / GraphQL) plus Swagger UI and ReDoc.

See the sibling page: FARP Registration API. For the concepts behind it, see Service Discovery & FARP and Schema federation.

Health and probe endpoints03

The gateway serves Kubernetes-style process probes (/livez, /readyz, /startz, plus the back-compat aliases /health and /healthz) on the main listen port. These are for liveness, readiness, and connection-draining — not for management.

See Health probes for the full path list, status-code semantics, and drain behaviour.

Metrics endpoint04

Prometheus metrics are exposed in text format on the listen port at /metrics (with the /__metrics alias):

curl http://localhost:8080/metrics

Next steps05

• Admin REST API — the management API, documented in full.

• FARP Registration API — service registration over HTTP.

• Health probes — liveness, readiness, startup, and drain.

• Metrics — the Prometheus endpoint and metric reference.