Vector Observability API

The Vector Observability API provides HTTP endpoints for health monitoring of running Vector instances and gRPC endpoints for component inspection and event streaming. Enable via api.enabled: true in Vector configuration. Binds to 127.0.0.1:8686 by default. Note: the API does not support authentication and should only be used in isolated environments.

OpenAPI Specification

vector-observability-api-openapi.yml Raw ↑
openapi: 3.0.3
info:
  title: Vector Observability API
  description: >-
    Vector is a high-performance observability data pipeline for collecting,
    transforming, and routing logs, metrics, and traces. The Vector Observability
    API provides HTTP endpoints for health checks and gRPC endpoints for component
    management, event streaming, and pipeline observability. The API must be
    explicitly enabled in Vector configuration (api.enabled: true) and binds
    to 127.0.0.1:8686 by default. The API does not currently support
    authentication and should only be enabled in isolated environments.
  version: "0.55.0"
  contact:
    name: Vector Community
    url: https://vector.dev/community/
  license:
    name: MPL-2.0
    url: https://github.com/vectordotdev/vector/blob/master/LICENSE
  x-generated-from: documentation

externalDocs:
  description: Vector API Reference
  url: https://vector.dev/docs/reference/api/

servers:
  - url: http://127.0.0.1:8686
    description: Local Vector Instance (default API address)

tags:
  - name: Health
    description: Health check endpoints for load balancers and Kubernetes probes.
  - name: Components
    description: List and inspect Vector pipeline components (sources, transforms, sinks).
  - name: Events
    description: Stream output events from Vector components for debugging and monitoring.

paths:
  /health:
    get:
      operationId: getHealth
      summary: Vector Get Health Status
      description: >-
        Returns the health status of the running Vector instance. Returns HTTP 200
        with {"ok": true} when Vector is running normally. Returns HTTP 200 with
        {"ok": false} when Vector is shutting down. Used by load balancers and
        Kubernetes liveness/readiness probes.
      tags:
        - Health
      security: []
      responses:
        '200':
          description: Vector health status response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HealthResponse'
              examples:
                getHealth200Example:
                  summary: Default getHealth 200 response
                  x-microcks-default: true
                  value:
                    ok: true
      x-microcks-operation:
        delay: 0
        dispatcher: FALLBACK
    head:
      operationId: headHealth
      summary: Vector Head Health Check
      description: >-
        Same semantics as GET /health but returns no response body. Used for
        lightweight health probes where the response body is not needed.
      tags:
        - Health
      security: []
      responses:
        '200':
          description: Vector is healthy
      x-microcks-operation:
        delay: 0
        dispatcher: FALLBACK

components:
  schemas:
    HealthResponse:
      type: object
      properties:
        ok:
          type: boolean
          description: >-
            Whether Vector is running normally. True when operational, false
            during shutdown.
          example: true