[vSphere] add triggered alarm to existing metricsets.

CHANGELOG.next.asciidoc

@@ -331,6 +331,7 @@ https://github.com/elastic/beats/compare/v8.8.1\...main[Check the HEAD diff]
 - Add new metrics for the vSphere Virtualmachine metricset. {pull}40485[40485]
 - Add support for snapshot in vSphere virtualmachine metricset {pull}40683[40683]
 - Update fields to use mapstr in vSphere virtualmachine metricset  {pull}40707[40707]
+- Add metrics related to alert in all the vSphere metricsets. {pull}40714[40714]
 - Add support for period based intervalID in vSphere host and datastore metricsets {pull}40678[40678]
 
 *Metricbeat*

metricbeat/docs/fields.asciidoc

@@ -66955,6 +66955,16 @@ type: keyword
 
 --
 
+*`vsphere.cluster.triggerd_alarms.*`*::
++
+--
+List of all the triggerd alarms.
+
+
+type: object
+
+--
+
 [float]
 === datastore
 
@@ -67092,6 +67102,16 @@ type: keyword
 
 --
 
+*`vsphere.datastore.triggerd_alarms.*`*::
++
+--
+List of all the triggerd alarms.
+
+
+type: object
+
+--
+
 *`vsphere.datastore.vm.count`*::
 +
 --
@@ -67195,6 +67215,16 @@ type: long
 
 --
 
+*`vsphere.datastorecluster.triggerd_alarms.*`*::
++
+--
+List of all the triggerd alarms.
+
+
+type: object
+
+--
+
 [float]
 === host
 
@@ -67552,6 +67582,16 @@ type: keyword
 
 --
 
+*`vsphere.host.triggerd_alarms.*`*::
++
+--
+List of all the triggerd alarms.
+
+
+type: object
+
+--
+
 *`vsphere.host.uptime`*::
 +
 --
@@ -67681,6 +67721,16 @@ type: long
 
 --
 
+*`vsphere.network.triggerd_alarms.*`*::
++
+--
+List of all the triggerd alarms.
+
+
+type: object
+
+--
+
 [float]
 === resourcepool
 
@@ -67907,6 +67957,16 @@ type: keyword
 
 --
 
+*`vsphere.resourcepool.triggerd_alarms.*`*::
++
+--
+List of all the triggerd alarms.
+
+
+type: object
+
+--
+
 [float]
 === virtualmachine
 
@@ -68134,6 +68194,16 @@ type: long
 
 --
 
+*`vsphere.virtualmachine.triggerd_alarms.*`*::
++
+--
+List of all the triggerd alarms.
+
+
+type: object
+
+--
+
 [[exported-fields-windows]]
 == Windows fields
 

metricbeat/module/vsphere/cluster/_meta/fields.yml

@@ -51,4 +51,9 @@
         - name: names
           type: keyword
           description: >
-            List of all the Network names associated with the cluster.
+            List of all the Network names associated with the cluster.
+    - name: triggerd_alarms.*
+      type: object
+      object_type: keyword
+      description: >
+        List of all the triggerd alarms.

metricbeat/module/vsphere/cluster/cluster.go

@@ -21,6 +21,7 @@ import (
 	"context"
 	"fmt"
 	"strings"
+	"time"
 
 	"github.com/elastic/beats/v7/metricbeat/mb"
 	"github.com/elastic/beats/v7/metricbeat/module/vsphere"
@@ -29,6 +30,7 @@ import (
 	"github.com/vmware/govmomi/property"
 	"github.com/vmware/govmomi/view"
 	"github.com/vmware/govmomi/vim25/mo"
+	"github.com/vmware/govmomi/vim25/types"
 )
 
 // init registers the MetricSet with the central registry as soon as the program
@@ -53,6 +55,19 @@ type assetNames struct {
 	outputHostNames      []string
 }
 
+type triggerdAlarm struct {
+	Name        string    `json:"name"`
+	ID          string    `json:"id"`
+	Status      string    `json:"status"`
+	Time        time.Time `json:"time"`
+	Description string    `json:"description"`
+}
+
+type metricData struct {
+	assetNames     assetNames
+	triggerdAlarms []triggerdAlarm
+}
+
 // New creates a new instance of the MetricSet.
 func New(base mb.BaseMetricSet) (mb.MetricSet, error) {
 	ms, err := vsphere.NewMetricSet(base)
@@ -97,7 +112,7 @@ func (m *ClusterMetricSet) Fetch(ctx context.Context, reporter mb.ReporterV2) er
 
 	// Retrieve summary property for all Clusters
 	var clt []mo.ClusterComputeResource
-	err = v.Retrieve(ctx, []string{"ClusterComputeResource"}, []string{"name", "host", "network", "datastore", "configuration"}, &clt)
+	err = v.Retrieve(ctx, []string{"ClusterComputeResource"}, []string{"name", "host", "network", "datastore", "configuration", "triggeredAlarmState"}, &clt)
 	if err != nil {
 		return fmt.Errorf("error in Retrieve: %w", err)
 	}
@@ -120,24 +135,29 @@ func (m *ClusterMetricSet) Fetch(ctx context.Context, reporter mb.ReporterV2) er
 				m.Logger().Warn("Metric das_config.enabled not found")
 			}
 
+			triggerdAlarm, err := getTriggerdAlarm(ctx, pc, clt[i].TriggeredAlarmState)
+			if err != nil {
+				m.Logger().Errorf("Failed to retrieve alerts from cluster %s: %w", clt[i].Name, err)
+			}
+
 			reporter.Event(mb.Event{
-				MetricSetFields: m.mapEvent(clt[i], assetNames),
+				MetricSetFields: m.mapEvent(clt[i], &metricData{assetNames: assetNames, triggerdAlarms: triggerdAlarm}),
 			})
 		}
 	}
 	return nil
 
 }
 
-func getAssetNames(ctx context.Context, pc *property.Collector, cl *mo.ClusterComputeResource) (*assetNames, error) {
+func getAssetNames(ctx context.Context, pc *property.Collector, cl *mo.ClusterComputeResource) (assetNames, error) {
 	referenceList := append(cl.Datastore, cl.Host...)
 
 	outputDatastoreNames := make([]string, 0, len(cl.Datastore))
 	outputHostNames := make([]string, 0, len(cl.Host))
 	if len(referenceList) > 0 {
 		var objects []mo.ManagedEntity
 		if err := pc.Retrieve(ctx, referenceList, []string{"name"}, &objects); err != nil {
-			return nil, fmt.Errorf("failed to retrieve managed entities: %w", err)
+			return assetNames{}, fmt.Errorf("failed to retrieve managed entities: %w", err)
 		}
 
 		for _, ob := range objects {
@@ -157,7 +177,7 @@ func getAssetNames(ctx context.Context, pc *property.Collector, cl *mo.ClusterCo
 	if len(cl.Network) > 0 {
 		var netObjects []mo.Network
 		if err := pc.Retrieve(ctx, cl.Network, []string{"name"}, &netObjects); err != nil {
-			return nil, fmt.Errorf("failed to retrieve network objects: %w", err)
+			return assetNames{}, fmt.Errorf("failed to retrieve network objects: %w", err)
 		}
 
 		for _, ob := range netObjects {
@@ -166,9 +186,30 @@ func getAssetNames(ctx context.Context, pc *property.Collector, cl *mo.ClusterCo
 		}
 	}
 
-	return &assetNames{
+	return assetNames{
 		outputNetworkNames:   outputNetworkNames,
 		outputDatastoreNames: outputDatastoreNames,
 		outputHostNames:      outputHostNames,
 	}, nil
 }
+
+func getTriggerdAlarm(ctx context.Context, pc *property.Collector, triggeredAlarmState []types.AlarmState) ([]triggerdAlarm, error) {
+	var triggeredAlarms []triggerdAlarm
+	for _, alarmState := range triggeredAlarmState {
+		var triggeredAlarm triggerdAlarm
+		var alarm mo.Alarm
+		err := pc.RetrieveOne(ctx, alarmState.Alarm, nil, &alarm)
+		if err != nil {
+			return nil, err
+		}
+		triggeredAlarm.Name = alarm.Info.Name
+		triggeredAlarm.Description = alarm.Info.Description
+		triggeredAlarm.ID = alarmState.Key
+		triggeredAlarm.Status = string(alarmState.OverallStatus)
+		triggeredAlarm.Time = alarmState.Time
+
+		triggeredAlarms = append(triggeredAlarms, triggeredAlarm)
+	}
+
+	return triggeredAlarms, nil
+}

metricbeat/module/vsphere/cluster/data.go

@@ -23,23 +23,27 @@ import (
 	"github.com/elastic/elastic-agent-libs/mapstr"
 )
 
-func (m *ClusterMetricSet) mapEvent(cl mo.ClusterComputeResource, data *assetNames) mapstr.M {
+func (m *ClusterMetricSet) mapEvent(cl mo.ClusterComputeResource, data *metricData) mapstr.M {
 	event := mapstr.M{
 		"host": mapstr.M{
-			"count": len(data.outputHostNames),
-			"names": data.outputHostNames,
+			"count": len(data.assetNames.outputHostNames),
+			"names": data.assetNames.outputHostNames,
 		},
 		"datastore": mapstr.M{
-			"count": len(data.outputDatastoreNames),
-			"names": data.outputDatastoreNames,
+			"count": len(data.assetNames.outputDatastoreNames),
+			"names": data.assetNames.outputDatastoreNames,
 		},
 		"network": mapstr.M{
-			"count": len(data.outputNetworkNames),
-			"names": data.outputNetworkNames,
+			"count": len(data.assetNames.outputNetworkNames),
+			"names": data.assetNames.outputNetworkNames,
 		},
 		"name": cl.Name,
 	}
 
+	if len(data.triggerdAlarms) > 0 {
+		event.Put("triggerd_alarms", data.triggerdAlarms)
+	}
+
 	if cl.Configuration.DasConfig.Enabled != nil {
 		event.Put("das_config.enabled", *cl.Configuration.DasConfig.Enabled)
 	}

metricbeat/module/vsphere/cluster/data_test.go

@@ -49,7 +49,11 @@ func TestEventMapping(t *testing.T) {
 		outputNetworkNames:   []string{"Network_0"},
 	}
 
-	outputEvent := m.mapEvent(ClusterTest, &assetNames)
+	var metricDataTest = metricData{
+		assetNames: assetNames,
+	}
+
+	outputEvent := m.mapEvent(ClusterTest, &metricDataTest)
 	testEvent := mapstr.M{
 		"das_config": mapstr.M{
 			"enabled": false,

metricbeat/module/vsphere/datastore/_meta/fields.yml

@@ -57,6 +57,11 @@
       type: keyword
       description: >
         Status of the datastore.
+    - name: triggerd_alarms.*
+      type: object
+      object_type: keyword
+      description: >
+        List of all the triggerd alarms.
     - name: vm.count
       type: long
       description: >

metricbeat/module/vsphere/datastore/data.go

@@ -49,6 +49,10 @@ func (m *DataStoreMetricSet) mapEvent(ds mo.Datastore, data *metricData) mapstr.
 		},
 	}
 
+	if len(data.triggerdAlarms) > 0 {
+		event.Put("triggerd_alarms", data.triggerdAlarms)
+	}
+
 	if ds.Summary.Capacity > 0 {
 		usedSpacePercent := float64(ds.Summary.Capacity-ds.Summary.FreeSpace) / float64(ds.Summary.Capacity)
 		event.Put("capacity.used.pct", usedSpacePercent)