feat: introduce health check endpoint (#2257)

* feat(query-service): ✨ Add health check route and handler

* chore(install-script): 🔧 use health endpoint with instead of services list

---------

Signed-off-by: Prashant Shahi <prashant@signoz.io>

deploy/install.sh

@@ -228,7 +228,7 @@ wait_for_containers_start() {
 
     # The while loop is important because for-loops don't work for dynamic values
     while [[ $timeout -gt 0 ]]; do
-        status_code="$(curl -s -o /dev/null -w "%{http_code}" http://localhost:3301/api/v1/services/list || true)"
+        status_code="$(curl -s -o /dev/null -w "%{http_code}" "http://localhost:3301/api/v1/health?live=1" || true)"
         if [[ status_code -eq 200 ]]; then
             break
         else

pkg/query-service/app/clickhouseReader/reader.go

@@ -3640,3 +3640,13 @@ func (r *ClickHouseReader) QueryDashboardVars(ctx context.Context, query string)
 	}
 	return &result, nil
 }
+
+func (r *ClickHouseReader) CheckClickHouse(ctx context.Context) error {
+	rows, err := r.db.Query(ctx, "SELECT 1")
+	if err != nil {
+		return err
+	}
+	defer rows.Close()
+
+	return nil
+}

pkg/query-service/app/http_handler.go

@@ -351,7 +351,7 @@ func (aH *APIHandler) RegisterRoutes(router *mux.Router) {
 	router.HandleFunc("/api/v1/feedback", OpenAccess(aH.submitFeedback)).Methods(http.MethodPost)
 	// router.HandleFunc("/api/v1/get_percentiles", aH.getApplicationPercentiles).Methods(http.MethodGet)
 	router.HandleFunc("/api/v1/services", ViewAccess(aH.getServices)).Methods(http.MethodPost)
-	router.HandleFunc("/api/v1/services/list", aH.getServicesList).Methods(http.MethodGet)
+	router.HandleFunc("/api/v1/services/list", ViewAccess(aH.getServicesList)).Methods(http.MethodGet)
 	router.HandleFunc("/api/v1/service/overview", ViewAccess(aH.getServiceOverview)).Methods(http.MethodPost)
 	router.HandleFunc("/api/v1/service/top_operations", ViewAccess(aH.getTopOperations)).Methods(http.MethodPost)
 	router.HandleFunc("/api/v1/service/top_level_operations", ViewAccess(aH.getServicesTopLevelOps)).Methods(http.MethodPost)
@@ -364,6 +364,7 @@ func (aH *APIHandler) RegisterRoutes(router *mux.Router) {
 	router.HandleFunc("/api/v1/version", OpenAccess(aH.getVersion)).Methods(http.MethodGet)
 	router.HandleFunc("/api/v1/featureFlags", OpenAccess(aH.getFeatureFlags)).Methods(http.MethodGet)
 	router.HandleFunc("/api/v1/configs", OpenAccess(aH.getConfigs)).Methods(http.MethodGet)
+	router.HandleFunc("/api/v1/health", OpenAccess(aH.getHealth)).Methods(http.MethodGet)
 
 	router.HandleFunc("/api/v1/getSpanFilters", ViewAccess(aH.getSpanFilters)).Methods(http.MethodPost)
 	router.HandleFunc("/api/v1/getTagFilters", ViewAccess(aH.getTagFilters)).Methods(http.MethodPost)
@@ -1671,6 +1672,22 @@ func (aH *APIHandler) getConfigs(w http.ResponseWriter, r *http.Request) {
 	aH.Respond(w, configs)
 }
 
+// getHealth is used to check the health of the service.
+// 'live' query param can be used to check liveliness of
+// the service by checking the database connection.
+func (aH *APIHandler) getHealth(w http.ResponseWriter, r *http.Request) {
+	_, ok := r.URL.Query()["live"]
+	if ok {
+		err := aH.reader.CheckClickHouse(r.Context())
+		if err != nil {
+			aH.HandleError(w, err, http.StatusServiceUnavailable)
+			return
+		}
+	}
+
+	aH.WriteJSON(w, r, map[string]string{"status": "ok"})
+}
+
 // inviteUser is used to invite a user. It is used by an admin api.
 func (aH *APIHandler) inviteUser(w http.ResponseWriter, r *http.Request) {
 	req, err := parseInviteRequest(r)

pkg/query-service/interfaces/interface.go

@@ -77,4 +77,5 @@ type Reader interface {
 	GetFanoutStorage() *storage.Storage
 
 	QueryDashboardVars(ctx context.Context, query string) (*model.DashboardVar, error)
+	CheckClickHouse(ctx context.Context) error
 }