fix: Add health check endpoint (#116)

* CHANGELOG

* fix: add a health check endpoint

* CHANGELOG

* fix: return wrapped error

Author: Vvaradinov

CHANGELOG.md

@@ -39,8 +39,10 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)
 
 ### Improvements
 
+* [#116](https://github.com/babylonlabs-io/covenant-emulator/pull/116) Add health check to prometheus server.
 * [#117](https://github.com/babylonlabs-io/covenant-emulator/pull/117) Add health check on startup, increase gas adjustment and update docs.
 
+
 ## v0.14.0
 
 * [#114](https://github.com/babylonlabs-io/covenant-emulator/pull/114) bump babylon to v1.0.0-rc.8

covenant/covenant.go

@@ -610,3 +610,23 @@ func (ce *CovenantEmulator) Stop() error {
 	})
 	return stopErr
 }
+
+// CheckReadiness checks if the covenant emulator is ready to serve requests.
+// It verifies internal state (if the main loop is running) and connectivity to
+// dependencies (remote signer).
+func (ce *CovenantEmulator) CheckReadiness() error {
+	select {
+	case <-ce.quit:
+		return fmt.Errorf("emulator is not running")
+	default:
+		// Emulator is running
+	}
+
+	// Check connectivity to the remote signer by calling its PubKey endpoint
+	_, err := ce.signer.PubKey()
+	if err != nil {
+		return fmt.Errorf("failed to get public key from covenant signer: %w", err)
+	}
+
+	return nil
+}

covenant/service/prometheus.go

@@ -6,10 +6,17 @@ import (
 	"net/http"
 	"time"
 
+	logger "github.com/rs/zerolog"
+
 	"github.com/prometheus/client_golang/prometheus/promhttp"
 	"go.uber.org/zap"
 )
 
+// ReadinessChecker defines the interface for checking readiness status
+type ReadinessChecker interface {
+	CheckReadiness() error
+}
+
 type PrometheusServer struct {
 	svr *http.Server
 
@@ -18,6 +25,8 @@ type PrometheusServer struct {
 	interval time.Duration
 
 	quit chan struct{}
+
+	readinessChecker ReadinessChecker
 }
 
 func NewPrometheusServer(addr string, interval time.Duration, logger *zap.Logger) *PrometheusServer {
@@ -39,6 +48,37 @@ func NewPrometheusServer(addr string, interval time.Duration, logger *zap.Logger
 	}
 }
 
+// SetReadinessChecker sets the readiness checker implementation
+func (ps *PrometheusServer) SetReadinessChecker(rc ReadinessChecker) {
+	ps.readinessChecker = rc
+
+	// Add the readiness endpoint only after the checker is set
+	if mux, ok := ps.svr.Handler.(*http.ServeMux); ok {
+		mux.HandleFunc("/health", ps.readinessHandler)
+		ps.logger.Info("Readiness endpoint registered at /health")
+	}
+}
+
+// readinessHandler handles the /health endpoint requests
+func (ps *PrometheusServer) readinessHandler(w http.ResponseWriter, r *http.Request) {
+	if ps.readinessChecker == nil {
+		ps.logger.Error("Readiness checker not configured")
+		http.Error(w, "Readiness checker not configured", http.StatusInternalServerError)
+		return
+	}
+
+	if err := ps.readinessChecker.CheckReadiness(); err != nil {
+		ps.logger.Error("Readiness check failed", zap.Error(err))
+		http.Error(w, err.Error(), http.StatusServiceUnavailable)
+		return
+	}
+
+	w.WriteHeader(http.StatusOK)
+	if _, err := w.Write([]byte("Healthy")); err != nil {
+		logger.Ctx(r.Context()).Err(err).Msg("failed to write response")
+	}
+}
+
 func (ps *PrometheusServer) Start() {
 	ps.logger.Info("Starting Prometheus server",
 		zap.String("address", ps.svr.Addr))

covenant/service/server.go

@@ -44,6 +44,8 @@ func (s *CovenantServer) RunUntilShutdown() error {
 	}
 
 	ps := NewPrometheusServer(promAddr, metricsCfg.UpdateInterval, s.logger)
+	// Set the covenant emulator as the readiness checker
+	ps.SetReadinessChecker(s.ce)
 
 	defer func() {
 		ps.Stop()

docs/covenant-emulator-setup.md

@@ -258,4 +258,35 @@ covd start
 ```
 
 All the available CLI options can be viewed using the `--help` flag. These
-options can also be set in the configuration file.
+options can also be set in the configuration file.
+
+## 6. Monitoring and Health Checks
+
+### 6.1. Readiness Check Endpoint
+
+The covenant emulator provides a health check endpoint at `/health` that can be used 
+by monitoring systems, to determine if the service is ready to handle requests.
+
+The endpoint can be accessed through the same address as the Prometheus metrics endpoint, 
+which is configured in `covd.conf`:
+
+```
+http://<prometheusAddr>/health
+```
+
+The endpoint returns:
+- `HTTP 200 OK` with response body "Healthy" when the emulator is running properly
+and can connect to its dependencies
+- `HTTP 503 Service Unavailable` with an error message when the emulator is not ready
+
+The readiness check verifies:
+1. Whether the emulator's main loop is running
+2. Connectivity to the remote signer by calling its PubKey endpoint
+
+#### Example Usage
+
+To check the emulator's readiness:
+
+```bash
+curl -v http://127.0.0.1:2112/health
+```