Add metrics

pkg/scheduler/metrics.go

@@ -0,0 +1,29 @@
+package scheduler
+
+import (
+	"time"
+
+	"github.com/zeta-chain/node/zetaclient/metrics"
+)
+
+// Note that currently the hard-coded "global" metrics are used.
+func recordMetrics(task *Task, startedAt time.Time, err error, skipped bool) {
+	var status string
+	switch {
+	case skipped:
+		status = "skipped"
+	case err != nil:
+		status = "failed"
+	default:
+		status = "ok"
+	}
+
+	var (
+		group = string(task.group)
+		name  = task.name
+		dur   = time.Since(startedAt).Seconds()
+	)
+
+	metrics.SchedulerTaskInvocationCounter.WithLabelValues(status, group, name).Inc()
+	metrics.SchedulerTaskExecutionDuration.WithLabelValues(status, group, name).Observe(dur)
+}

pkg/scheduler/scheduler.go

@@ -163,22 +163,17 @@ func (t *Task) Stop() {
 
 // execute executes Task with additional logging and metrics.
 func (t *Task) execute(ctx context.Context) error {
+	startedAt := time.Now().UTC()
+
 	// skip tick
 	if t.skipper != nil && t.skipper() {
+		recordMetrics(t, startedAt, nil, true)
 		return nil
 	}
 
-	t.logger.Debug().Msg("Invoking task")
-
 	err := t.exec(ctx)
 
-	// todo metrics (TBD)
-	//   - duration (time taken)
-	//   - outcome (skip, err, ok)
-	//   - bump invocation counter
-	//   - "last invoked at" timestamp (?)
-	//   - chain_id
-	//   - metrics cardinality: "task_group (?)" "task_name", "status", "chain_id"
+	recordMetrics(t, startedAt, err, false)
 
 	return err
 }

zetaclient/metrics/metrics.go

@@ -170,6 +170,27 @@ var (
 		Name:      "num_connected_peers",
 		Help:      "The number of connected peers (authenticated keygen peers)",
 	})
+
+	// SchedulerTaskInvocationCounter tracks invocations categorized by status, group, and name
+	SchedulerTaskInvocationCounter = promauto.NewCounterVec(
+		prometheus.CounterOpts{
+			Namespace: ZetaClientNamespace,
+			Name:      "scheduler_task_invocations_total",
+			Help:      "Total number of task invocations",
+		},
+		[]string{"status", "task_group", "task_name"},
+	)
+
+	// SchedulerTaskExecutionDuration measures the execution duration of tasks
+	SchedulerTaskExecutionDuration = promauto.NewHistogramVec(
+		prometheus.HistogramOpts{
+			Namespace: ZetaClientNamespace,
+			Name:      "scheduler_task_duration_seconds",
+			Help:      "Histogram of task execution duration in seconds",
+			Buckets:   []float64{0.05, 0.1, 0.2, 0.3, 0.5, 1, 1.5, 2, 3, 5, 7.5, 10, 15}, // 50ms to 15s
+		},
+		[]string{"status", "task_group", "task_name"},
+	)
 )
 
 // NewMetrics creates a new Metrics instance