Skip to main content

Module health

zebrad::components

Module health 

Source
Expand description

HTTP health and readiness endpoints for zebrad.

Overview

  • This module exposes two small HTTP/1.1 endpoints for basic liveness/readiness checks, suitable for Kubernetes probes and load balancers.
  • Endpoints are opt-in, disabled by default. Enable by setting a listen_addr in the health config section.
  • Plain-text responses and small responses keep the checks fast and safe.

Endpoints

  • GET /healthy — returns 200 OK if the process is up and the node has at least min_connected_peers recently live peers (default: 1). Otherwise 503 Service Unavailable.
  • GET /ready — returns 200 OK if the node is near the chain tip, the estimated block lag is within ready_max_blocks_behind, and the latest committed block is recent. On regtest/testnet, readiness returns 200 unless enforce_on_test_networks is set.

Security

  • Endpoints are unauthenticated by design. Bind to internal network interfaces, and restrict exposure using network policy, firewall rules, and service configuration.
  • The server does not access or return private data. It only summarises coarse node state.

Configuration and examples

Modules§

config 🔒

Structs§

ChainTipMetrics
Metrics tracking how long it’s been since
Config
Health server configuration.
HealthCtx 🔒

Constants§

MAX_RECENT_REQUESTS 🔒
The maximum number of requests that will be handled in a given time interval before requests are dropped.
METHOD_NOT_ALLOWED_MSG 🔒
NOT_FOUND_MSG 🔒
PEER_METRICS_REFRESH_INTERVAL 🔒
RECENT_REQUEST_INTERVAL 🔒

Functions§

handle_request 🔒
healthy 🔒
init
Starts the health server if listen_addr is configured.
num_live_peers 🔒
peer_metrics_refresh_task 🔒
ready 🔒
run_health_server 🔒
simple_response 🔒