Vmalert metrics update (#1580)

* vmalert: remove `vmalert_execution_duration_seconds` metric

The summary for `vmalert_execution_duration_seconds` metric gives no additional
value comparing to `vmalert_iteration_duration_seconds` metric.

* vmalert: update config reload success metric properly

Previously, if there was unsuccessfull attempt to reload config and then
rollback to previous version - the metric remained set to 0.

* vmalert: add Grafana dashboard to overview application metrics

* docker: include vmalert target into list for scraping

* vmalert: extend notifier metrics with addr label

The change adds an `addr` label to metrics for alerts_sent and alerts_send_errors
to identify which exact address is having issues.
The according change was made to vmalert dashboard.

* vmalert: update documentation and docker environment for vmalert's dashboard

Mention Grafana's dashboard in vmalert's README in a new section #Monitoring.

Update docker-compose env to automatically add vmalert's dashboard.
Update docker-compose README with additional info about services.

app/vmalert/Makefile

@@ -56,6 +56,7 @@ test-vmalert:
 	go test -v -race -cover ./app/vmalert/datasource
 	go test -v -race -cover ./app/vmalert/notifier
 	go test -v -race -cover ./app/vmalert/config
+	go test -v -race -cover ./app/vmalert/remotewrite
 
 run-vmalert: vmalert
 	./bin/vmalert -rule=app/vmalert/config/testdata/rules2-good.rules \

app/vmalert/README.md

@@ -340,6 +340,17 @@ See full description for these flags in `./vmalert --help`.
 * `query` template function is disabled for performance reasons (might be changed in future);
 
 
+## Monitoring
+
+`vmalert` exports various metrics in Prometheus exposition format at `http://vmalert-host:8880/metrics` page. 
+We recommend setting up regular scraping of this page either through `vmagent` or by Prometheus so that the exported 
+metrics may be analyzed later.
+
+Use official [Grafana dashboard](https://grafana.com/grafana/dashboards/14950) for `vmalert` overview.
+If you have suggestions for improvements or have found a bug - please open an issue on github or add 
+a review to the dashboard.
+
+
 ## Configuration
 
 Pass `-help` to `vmalert` in order to see the full list of supported

app/vmalert/group.go

@@ -174,12 +174,6 @@ func (g *Group) updateWith(newGroup *Group) error {
 	return nil
 }
 
-var (
-	alertsFired      = metrics.NewCounter(`vmalert_alerts_fired_total`)
-	alertsSent       = metrics.NewCounter(`vmalert_alerts_sent_total`)
-	alertsSendErrors = metrics.NewCounter(`vmalert_alerts_send_errors_total`)
-)
-
 func (g *Group) close() {
 	if g.doneCh == nil {
 		return
@@ -220,7 +214,16 @@ func (g *Group) start(ctx context.Context, nts []notifier.Notifier, rw *remotewr
 	}
 
 	logger.Infof("group %q started; interval=%v; concurrency=%d", g.Name, g.Interval, g.Concurrency)
-	e := &executor{nts, rw}
+	e := &executor{rw: rw}
+	for _, nt := range nts {
+		ent := eNotifier{
+			Notifier:         nt,
+			alertsSent:       getOrCreateCounter(fmt.Sprintf("vmalert_alerts_sent_total{addr=%q}", nt.Addr())),
+			alertsSendErrors: getOrCreateCounter(fmt.Sprintf("vmalert_alerts_send_errors_total{addr=%q}", nt.Addr())),
+		}
+		e.notifiers = append(e.notifiers, ent)
+	}
+
 	t := time.NewTicker(g.Interval)
 	defer t.Stop()
 	for {
@@ -263,10 +266,16 @@ func (g *Group) start(ctx context.Context, nts []notifier.Notifier, rw *remotewr
 }
 
 type executor struct {
-	notifiers []notifier.Notifier
+	notifiers []eNotifier
 	rw        *remotewrite.Client
 }
 
+type eNotifier struct {
+	notifier.Notifier
+	alertsSent       *counter
+	alertsSendErrors *counter
+}
+
 func (e *executor) execConcurrently(ctx context.Context, rules []Rule, concurrency int, interval time.Duration) chan error {
 	res := make(chan error, len(rules))
 	if concurrency == 1 {
@@ -297,19 +306,16 @@ func (e *executor) execConcurrently(ctx context.Context, rules []Rule, concurren
 }
 
 var (
+	alertsFired = metrics.NewCounter(`vmalert_alerts_fired_total`)
+
 	execTotal  = metrics.NewCounter(`vmalert_execution_total`)
 	execErrors = metrics.NewCounter(`vmalert_execution_errors_total`)
-	execDuration = metrics.NewSummary(`vmalert_execution_duration_seconds`)
 
 	remoteWriteErrors = metrics.NewCounter(`vmalert_remotewrite_errors_total`)
 )
 
 func (e *executor) exec(ctx context.Context, rule Rule, interval time.Duration) error {
 	execTotal.Inc()
-	execStart := time.Now()
-	defer func() {
-		execDuration.UpdateDuration(execStart)
-	}()
 
 	tss, err := rule.Exec(ctx)
 	if err != nil {
@@ -350,11 +356,11 @@ func (e *executor) exec(ctx context.Context, rule Rule, interval time.Duration)
 		return nil
 	}
 
-	alertsSent.Add(len(alerts))
 	errGr := new(utils.ErrGroup)
 	for _, nt := range e.notifiers {
+		nt.alertsSent.Add(len(alerts))
 		if err := nt.Send(ctx, alerts); err != nil {
-			alertsSendErrors.Inc()
+			nt.alertsSendErrors.Inc()
 			errGr.Add(fmt.Errorf("rule %q: failed to send alerts: %w", rule, err))
 		}
 	}

app/vmalert/helpers_test.go

@@ -63,6 +63,7 @@ type fakeNotifier struct {
 	alerts []notifier.Alert
 }
 
+func (*fakeNotifier) Addr() string { return "" }
 func (fn *fakeNotifier) Send(_ context.Context, alerts []notifier.Alert) error {
 	fn.Lock()
 	defer fn.Unlock()

app/vmalert/main.go

@@ -274,6 +274,9 @@ func configReload(ctx context.Context, m *manager, groupsCfg []config.Group) {
 			continue
 		}
 		if configsEqual(newGroupsCfg, groupsCfg) {
+			// set success to 1 since previous reload
+			// could have been unsuccessful
+			configSuccess.Set(1)
 			// config didn't change - skip it
 			continue
 		}

app/vmalert/notifier/alertmanager.go

@@ -12,6 +12,7 @@ import (
 // AlertManager represents integration provider with Prometheus alert manager
 // https://github.com/prometheus/alertmanager
 type AlertManager struct {
+	addr          string
 	alertURL      string
 	basicAuthUser string
 	basicAuthPass string
@@ -19,6 +20,9 @@ type AlertManager struct {
 	client        *http.Client
 }
 
+// Addr returns address where alerts are sent.
+func (am AlertManager) Addr() string { return am.addr }
+
 // Send an alert or resolve message
 func (am *AlertManager) Send(ctx context.Context, alerts []Alert) error {
 	b := &bytes.Buffer{}
@@ -57,9 +61,10 @@ const alertManagerPath = "/api/v2/alerts"
 
 // NewAlertManager is a constructor for AlertManager
 func NewAlertManager(alertManagerURL, user, pass string, fn AlertURLGenerator, c *http.Client) *AlertManager {
-	addr := strings.TrimSuffix(alertManagerURL, "/") + alertManagerPath
+	url := strings.TrimSuffix(alertManagerURL, "/") + alertManagerPath
 	return &AlertManager{
-		alertURL:      addr,
+		addr:          alertManagerURL,
+		alertURL:      url,
 		argFunc:       fn,
 		client:        c,
 		basicAuthUser: user,

app/vmalert/notifier/alertmanager_test.go

@@ -10,6 +10,14 @@ import (
 	"time"
 )
 
+func TestAlertManager_Addr(t *testing.T) {
+	const addr = "http://localhost"
+	am := NewAlertManager(addr, "", "", nil, nil)
+	if am.Addr() != addr {
+		t.Errorf("expected to have %q; got %q", addr, am.Addr())
+	}
+}
+
 func TestAlertManager_Send(t *testing.T) {
 	const baUser, baPass = "foo", "bar"
 	mux := http.NewServeMux()

app/vmalert/notifier/notifier.go

@@ -2,7 +2,12 @@ package notifier
 
 import "context"
 
-// Notifier is common interface for alert manager provider
+// Notifier is a common interface for alert manager provider
 type Notifier interface {
+	// Send sends the given list of alerts.
+	// Returns an error if fails to send the alerts.
+	// Must unblock if the given ctx is cancelled.
 	Send(ctx context.Context, alerts []Alert) error
+	// Addr returns address where alerts are sent.
+	Addr() string
 }

deployment/docker/prometheus.yml

@@ -6,15 +6,15 @@ scrape_configs:
   - job_name: 'vmagent'
     static_configs:
       - targets: ['vmagent:8429']
-
+  - job_name: 'vmalert'
+    static_configs:
+      - targets: ['vmalert:8880']
   - job_name: 'vminsert'
     static_configs:
       - targets: ['vminsert:8480']
-
   - job_name: 'vmselect'
     static_configs:
       - targets: ['vmselect:8481']
-
   - job_name: 'vmstorage'
     static_configs:
       - targets: ['vmstorage:8482']