chore: add firing alert count and system/k8s metric existance status (#11730)

pkg/modules/inframonitoring/implinframonitoring/collect.go

@@ -0,0 +1,32 @@
+package implinframonitoring
+
+import (
+	"context"
+	"fmt"
+
+	"github.com/SigNoz/signoz/pkg/errors"
+	"github.com/SigNoz/signoz/pkg/telemetrymetrics"
+	"github.com/SigNoz/signoz/pkg/valuer"
+)
+
+func (m *module) Collect(ctx context.Context, _ valuer.UUID) (map[string]any, error) {
+	stats := make(map[string]any)
+
+	metadataTable := fmt.Sprintf("%s.%s", telemetrymetrics.DBName, telemetrymetrics.AttributesMetadataTableName)
+	var (
+		systemMetricCount uint64
+		k8sMetricCount    uint64
+	)
+	query := fmt.Sprintf(
+		"SELECT (SELECT count() FROM (SELECT 1 FROM %s WHERE metric_name LIKE 'system.%%' LIMIT 1)), (SELECT count() FROM (SELECT 1 FROM %s WHERE metric_name LIKE 'k8s.%%' LIMIT 1))",
+		metadataTable, metadataTable,
+	)
+	if err := m.telemetryStore.ClickhouseDB().QueryRow(ctx, query).Scan(&systemMetricCount, &k8sMetricCount); err == nil {
+		stats["telemetry.metrics.system.exists"] = systemMetricCount > 0
+		stats["telemetry.metrics.k8s.exists"] = k8sMetricCount > 0
+	} else {
+		m.logger.DebugContext(ctx, "failed to collect metrics namespace existence stats", errors.Attr(err))
+	}
+
+	return stats, nil
+}

pkg/modules/inframonitoring/inframonitoring.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"net/http"
 
+	"github.com/SigNoz/signoz/pkg/statsreporter"
 	"github.com/SigNoz/signoz/pkg/types/inframonitoringtypes"
 	"github.com/SigNoz/signoz/pkg/valuer"
 )
@@ -22,6 +23,7 @@ type Handler interface {
 }
 
 type Module interface {
+	statsreporter.StatsCollector
 	ListHosts(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableHosts) (*inframonitoringtypes.Hosts, error)
 	ListPods(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostablePods) (*inframonitoringtypes.Pods, error)
 	ListNodes(ctx context.Context, orgID valuer.UUID, req *inframonitoringtypes.PostableNodes) (*inframonitoringtypes.Nodes, error)

pkg/query-service/rules/manager.go

@@ -744,6 +744,32 @@ func (m *Manager) TriggeredAlerts() []*ruletypes.NamedAlert {
 	return namedAlerts
 }
 
+type AlertStats struct {
+	FiringRules int64
+	LastFiredAt time.Time
+}
+
+func (m *Manager) AlertStats(ctx context.Context) AlertStats {
+	m.mtx.RLock()
+	defer m.mtx.RUnlock()
+
+	// TODO(therealpandey): scope these stats per org; rules for all orgs are aggregated here.
+	stats := AlertStats{}
+	for _, r := range m.rules {
+		if r.State() == ruletypes.StateFiring {
+			stats.FiringRules++
+		}
+
+		for _, alert := range r.ActiveAlerts() {
+			if alert.FiredAt.After(stats.LastFiredAt) {
+				stats.LastFiredAt = alert.FiredAt
+			}
+		}
+	}
+
+	return stats
+}
+
 // NotifyFunc sends notifications about a set of alerts generated by the given expression.
 type NotifyFunc func(ctx context.Context, orgID string, alerts ...*ruletypes.Alert)
 

pkg/ruler/signozruler/provider.go

@@ -100,7 +100,16 @@ func (provider *provider) Collect(ctx context.Context, orgID valuer.UUID) (map[s
 		return nil, err
 	}
 
-	return ruletypes.NewStatsFromRules(rules), nil
+	stats := ruletypes.NewStatsFromRules(rules)
+
+	alertStats := provider.manager.AlertStats(ctx)
+	stats["alert.firing.count"] = alertStats.FiringRules
+	if !alertStats.LastFiredAt.IsZero() {
+		stats["alert.last_fired.time"] = alertStats.LastFiredAt.UTC()
+		stats["alert.last_fired.time_unix"] = alertStats.LastFiredAt.Unix()
+	}
+
+	return stats, nil
 }
 
 func (provider *provider) ListRuleStates(ctx context.Context) (*ruletypes.GettableRules, error) {

pkg/signoz/signoz.go

@@ -499,6 +499,7 @@ func New(
 		serviceAccount,
 		cloudIntegrationModule,
 		modules.LogsPipeline,
+		modules.InfraMonitoring,
 		querier,
 	}