feat: Tracking cluster image update progress via Rest

cmd/collectors/rest/plugins/clustersoftware/clustersoftware.go

@@ -0,0 +1,260 @@
+package clustersoftware
+
+import (
+	"github.com/netapp/harvest/v2/cmd/poller/plugin"
+	"github.com/netapp/harvest/v2/pkg/conf"
+	"github.com/netapp/harvest/v2/pkg/matrix"
+	"github.com/netapp/harvest/v2/pkg/slogx"
+	"github.com/netapp/harvest/v2/pkg/tree/node"
+	"github.com/netapp/harvest/v2/pkg/util"
+	"github.com/netapp/harvest/v2/third_party/tidwall/gjson"
+	"log/slog"
+)
+
+const (
+	clusterSoftware  = "cluster_software"
+	updateMatrix     = "update"
+	statusMatrix     = "status"
+	validationMatrix = "validation"
+)
+
+type ClusterSoftware struct {
+	*plugin.AbstractPlugin
+	data map[string]*matrix.Matrix
+}
+
+func New(p *plugin.AbstractPlugin) plugin.Plugin {
+	return &ClusterSoftware{AbstractPlugin: p}
+}
+
+func (c *ClusterSoftware) Init(conf.Remote) error {
+	if err := c.InitAbc(); err != nil {
+		return err
+	}
+
+	c.data = make(map[string]*matrix.Matrix)
+	if err := c.createUpdateMetrics(); err != nil {
+		return err
+	}
+	if err := c.createStatusMetrics(); err != nil {
+		return err
+	}
+	if err := c.createValidationMetrics(); err != nil {
+		return err
+	}
+
+	return nil
+}
+
+func (c *ClusterSoftware) Run(dataMap map[string]*matrix.Matrix) ([]*matrix.Matrix, *util.Metadata, error) {
+	globalLabels := dataMap[c.Object].GetGlobalLabels()
+
+	for _, instance := range dataMap[c.Object].GetInstances() {
+		instance.SetExportable(false)
+		// generate update details metrics
+		updateDetails := instance.GetLabel("update_details")
+		updateDetailsJSON := gjson.Result{Type: gjson.JSON, Raw: "[" + updateDetails + "]"}
+		c.handleUpdateDetails(updateDetailsJSON, globalLabels)
+
+		// generate status details metrics
+		statusDetails := instance.GetLabel("status_details")
+		statusDetailsJSON := gjson.Result{Type: gjson.JSON, Raw: "[" + statusDetails + "]"}
+		c.handleStatusDetails(statusDetailsJSON, globalLabels)
+
+		// generate update details metrics
+		validationResults := instance.GetLabel("validation_results")
+		validationResultsJSON := gjson.Result{Type: gjson.JSON, Raw: "[" + validationResults + "]"}
+		c.handleValidationDetails(validationResultsJSON, globalLabels)
+	}
+
+	softwareMetrics := make([]*matrix.Matrix, 0, len(c.data))
+	for _, val := range c.data {
+		softwareMetrics = append(softwareMetrics, val)
+	}
+
+	return softwareMetrics, nil, nil
+}
+
+func (c *ClusterSoftware) createUpdateMetrics() error {
+	mat := matrix.New(c.Parent+"."+updateMatrix, clusterSoftware, clusterSoftware)
+	exportOptions := node.NewS("export_options")
+	instanceKeys := exportOptions.NewChildS("instance_keys", "")
+	instanceKeys.NewChildS("", "phase")
+	instanceKeys.NewChildS("", "state")
+	instanceKeys.NewChildS("", "node")
+	instanceKeys.NewChildS("", "elapsed_duration")
+
+	mat.SetExportOptions(exportOptions)
+
+	if _, err := mat.NewMetricFloat64(updateMatrix); err != nil {
+		c.SLogger.Error("Failed to create metric", slogx.Err(err), slog.String("metric", updateMatrix))
+		return err
+	}
+
+	c.data[updateMatrix] = mat
+	return nil
+}
+
+func (c *ClusterSoftware) createStatusMetrics() error {
+	mat := matrix.New(c.Parent+"."+statusMatrix, clusterSoftware, clusterSoftware)
+	exportOptions := node.NewS("export_options")
+	instanceKeys := exportOptions.NewChildS("instance_keys", "")
+	instanceKeys.NewChildS("", "state")
+	instanceKeys.NewChildS("", "node")
+	instanceKeys.NewChildS("", "name")
+
+	mat.SetExportOptions(exportOptions)
+
+	if _, err := mat.NewMetricFloat64(statusMatrix); err != nil {
+		c.SLogger.Error("Failed to create metric", slogx.Err(err), slog.String("metric", statusMatrix))
+		return err
+	}
+
+	c.data[statusMatrix] = mat
+	return nil
+}
+
+func (c *ClusterSoftware) createValidationMetrics() error {
+	mat := matrix.New(c.Parent+"."+validationMatrix, clusterSoftware, clusterSoftware)
+	exportOptions := node.NewS("export_options")
+	instanceKeys := exportOptions.NewChildS("instance_keys", "")
+	instanceKeys.NewChildS("", "status")
+	instanceKeys.NewChildS("", "update_check")
+
+	mat.SetExportOptions(exportOptions)
+
+	if _, err := mat.NewMetricFloat64(validationMatrix); err != nil {
+		c.SLogger.Error("Failed to create metric", slogx.Err(err), slog.String("metric", validationMatrix))
+		return err
+	}
+
+	c.data[validationMatrix] = mat
+	return nil
+}
+
+func (c *ClusterSoftware) handleUpdateDetails(updateDetailsJSON gjson.Result, globalLabels map[string]string) {
+	var (
+		clusterUpdateInstance *matrix.Instance
+		key                   string
+		err                   error
+	)
+	// Purge and reset data
+	c.data[updateMatrix].PurgeInstances()
+	c.data[updateMatrix].Reset()
+
+	// Set all global labels
+	c.data[updateMatrix].SetGlobalLabels(globalLabels)
+
+	for _, updateDetail := range updateDetailsJSON.Array() {
+		phase := updateDetail.Get("phase").ClonedString()
+		state := updateDetail.Get("state").ClonedString()
+		elapsedDuration := updateDetail.Get("elapsed_duration").ClonedString()
+		nodeName := updateDetail.Get("node.name").ClonedString()
+		key = phase + state + nodeName
+
+		if clusterUpdateInstance, err = c.data[updateMatrix].NewInstance(key); err != nil {
+			c.SLogger.Error("Failed to create instance", slogx.Err(err), slog.String("key", key))
+			continue
+		}
+		clusterUpdateInstance.SetLabel("node", nodeName)
+		clusterUpdateInstance.SetLabel("state", state)
+		clusterUpdateInstance.SetLabel("phase", phase)
+		clusterUpdateInstance.SetLabel("elapsed_duration", elapsedDuration)
+
+		// populate numeric data
+		value := 0.0
+		if state == "completed" {
+			value = 1.0
+		}
+
+		met := c.data[updateMatrix].GetMetric(updateMatrix)
+		if err := met.SetValueFloat64(clusterUpdateInstance, value); err != nil {
+			c.SLogger.Error("Failed to parse value", slogx.Err(err), slog.Float64("value", value))
+		} else {
+			c.SLogger.Debug("added value", slog.Float64("value", value))
+		}
+	}
+}
+
+func (c *ClusterSoftware) handleStatusDetails(statusDetailsJSON gjson.Result, globalLabels map[string]string) {
+	var (
+		clusterStatusInstance *matrix.Instance
+		key                   string
+		err                   error
+	)
+	// Purge and reset data
+	c.data[statusMatrix].PurgeInstances()
+	c.data[statusMatrix].Reset()
+
+	// Set all global labels
+	c.data[statusMatrix].SetGlobalLabels(globalLabels)
+
+	for _, updateDetail := range statusDetailsJSON.Array() {
+		name := updateDetail.Get("name").ClonedString()
+		state := updateDetail.Get("state").ClonedString()
+		nodeName := updateDetail.Get("node.name").ClonedString()
+		key = name + state + nodeName
+
+		if clusterStatusInstance, err = c.data[statusMatrix].NewInstance(key); err != nil {
+			c.SLogger.Error("Failed to create instance", slogx.Err(err), slog.String("key", key))
+			continue
+		}
+		clusterStatusInstance.SetLabel("node", nodeName)
+		clusterStatusInstance.SetLabel("state", state)
+		clusterStatusInstance.SetLabel("name", name)
+
+		// populate numeric data
+		value := 0.0
+		if state == "completed" {
+			value = 1.0
+		}
+
+		met := c.data[statusMatrix].GetMetric(statusMatrix)
+		if err := met.SetValueFloat64(clusterStatusInstance, value); err != nil {
+			c.SLogger.Error("Failed to parse value", slogx.Err(err), slog.Float64("value", value))
+		} else {
+			c.SLogger.Debug("added value", slog.Float64("value", value))
+		}
+	}
+}
+
+func (c *ClusterSoftware) handleValidationDetails(validationDetailsJSON gjson.Result, globalLabels map[string]string) {
+	var (
+		clusterValidationInstance *matrix.Instance
+		key                       string
+		err                       error
+	)
+	// Purge and reset data
+	c.data[validationMatrix].PurgeInstances()
+	c.data[validationMatrix].Reset()
+
+	// Set all global labels
+	c.data[validationMatrix].SetGlobalLabels(globalLabels)
+
+	for _, updateDetail := range validationDetailsJSON.Array() {
+		updateCheck := updateDetail.Get("update_check").ClonedString()
+		status := updateDetail.Get("status").ClonedString()
+		key = updateCheck + status
+
+		if clusterValidationInstance, err = c.data[validationMatrix].NewInstance(key); err != nil {
+			c.SLogger.Error("Failed to create instance", slogx.Err(err), slog.String("key", key))
+			continue
+		}
+		clusterValidationInstance.SetLabel("update_check", updateCheck)
+		clusterValidationInstance.SetLabel("status", status)
+
+		// ignore all the validation result which are not in warning status
+		if status != "warning" {
+			continue
+		}
+
+		// populate numeric data
+		value := 1.0
+		met := c.data[validationMatrix].GetMetric(validationMatrix)
+		if err := met.SetValueFloat64(clusterValidationInstance, value); err != nil {
+			c.SLogger.Error("Failed to parse value", slogx.Err(err), slog.Float64("value", value))
+		} else {
+			c.SLogger.Debug("added value", slog.Float64("value", value))
+		}
+	}
+}

cmd/collectors/rest/rest.go

@@ -7,6 +7,7 @@ import (
 	"github.com/netapp/harvest/v2/cmd/collectors/rest/plugins/aggregate"
 	"github.com/netapp/harvest/v2/cmd/collectors/rest/plugins/certificate"
 	"github.com/netapp/harvest/v2/cmd/collectors/rest/plugins/cluster"
+	"github.com/netapp/harvest/v2/cmd/collectors/rest/plugins/clustersoftware"
 	"github.com/netapp/harvest/v2/cmd/collectors/rest/plugins/disk"
 	"github.com/netapp/harvest/v2/cmd/collectors/rest/plugins/health"
 	"github.com/netapp/harvest/v2/cmd/collectors/rest/plugins/metroclustercheck"
@@ -472,6 +473,8 @@ func (r *Rest) LoadPlugin(kind string, abc *plugin.AbstractPlugin) plugin.Plugin
 		return aggregate.New(abc)
 	case "Cluster":
 		return cluster.New(abc)
+	case "ClusterSoftware":
+		return clustersoftware.New(abc)
 	case "Disk":
 		return disk.New(abc)
 	case "Health":

cmd/tools/generate/counter.go

@@ -173,7 +173,6 @@ var (
 		"_labels",
 		"volume_arw_status",
 		"ALERTS",
-		"_tags",
 	}
 
 	// Exclude extra metrics for ZAPI

cmd/tools/generate/counter.yaml

@@ -1988,3 +1988,44 @@ counters:
         Endpoint: NA
         ONTAPCounter: Harvest generated
         Template: conf/rest/9.12.0/fru.yaml
+
+  - Name: cluster_tags
+    Description: This metric display tags at cluster level.
+    APIs:
+      - API: REST
+        Endpoint: NA
+        ONTAPCounter: Harvest generated
+        Template: conf/rest/9.12.0/status.yaml
+
+  - Name: volume_tags
+    Description: This metric display tags at volume level.
+    APIs:
+      - API: REST
+        Endpoint: NA
+        ONTAPCounter: Harvest generated
+        Template: conf/rest/9.12.0/volume.yaml
+
+
+  - Name: cluster_software_update
+    Description: This metric displays the software update phase with its status.
+    APIs:
+      - API: REST
+        Endpoint: NA
+        ONTAPCounter: Harvest generated
+        Template: conf/rest/9.6.0/clustersoftware.yaml
+
+  - Name: cluster_software_status
+    Description: This metric displays the software job with its status.
+    APIs:
+      - API: REST
+        Endpoint: NA
+        ONTAPCounter: Harvest generated
+        Template: conf/rest/9.6.0/clustersoftware.yaml
+
+  - Name: cluster_software_validation
+    Description: This metric displays the software pre validation checks with its status.
+    APIs:
+      - API: REST
+        Endpoint: NA
+        ONTAPCounter: Harvest generated
+        Template: conf/rest/9.6.0/clustersoftware.yaml

conf/rest/9.6.0/clustersoftware.yaml

@@ -0,0 +1,14 @@
+
+name:       ClusterSoftware
+query:      api/cluster/software
+object:     cluster_software
+
+counters:
+  - ^status_details      => status_details
+  - ^update_details      => update_details
+  - ^validation_results  => validation_results
+
+plugins:
+  - ClusterSoftware
+
+

conf/rest/default.yaml

@@ -16,6 +16,7 @@ objects:
 #  CIFSShare:                    cifs_share.yaml
   CloudTarget:                 cloud_target.yaml
   ClusterPeer:                 clusterpeer.yaml
+  ClusterSoftware:             clustersoftware.yaml
   Disk:                        disk.yaml
   EmsDestination:              ems_destination.yaml
 #  ExportRule:                  exports.yaml