adds helm information gather

* feat: gather helm information

* feat(gather/hemlchart_info): add resource counter

* docs: adds helmchart_info sample data

* test: adds some unit test for helmchart_gather

* test: expanding gather_helm_info tests

* feat: adding helmchart gatherer

* docs: update gathered-data

* Revert "feat: adding helmchart gatherer"

This reverts commit e57a5c1.

* refactor: revert helmcharts gatherer

* style: fix lint errors

* refactor(helm_info): remove if for label checking

* refactor(gather_info): unexpose labelChartNameKey const

* fix: wrong log message

* test(helm_gather_info): invalid resources

docs/gathered-data.md

@@ -404,6 +404,30 @@ filtered to only include those with a deployment_validation_operator_ prefix.
   - 4.10
 
 
+## HelmInfo
+
+Collects summarized info about the helm usage on a cluster
+in a generic fashion
+
+### API Reference
+None
+
+### Sample data
+- [docs/insights-archive-sample/config/helmchart_info.json](./insights-archive-sample/config/helmchart_info.json)
+
+### Location in archive
+- `config/helmchart_info.json`
+
+### Config ID
+`workloads/helmchart_info`
+
+### Released version
+- 4.15.0
+
+### Backported versions
+None
+
+
 ## HostSubnet
 
 collects HostSubnet information

docs/insights-archive-sample/config/helmchart_info.json

@@ -0,0 +1,19 @@
+{
+    "6c7e03817b27fbe6cc67ae835381df521b8d847dd029fb2df483f1a327b63582": [
+        { "name": "jenkins", "version": "0.0.3", "resources": { "services": 2 } }
+    ],
+    "a085ddf97d8c556fd4f965392f7d3446c4b11df0544848878e06b20c96523064": [
+        {
+            "name": "nodejs",
+            "version": "",
+            "resources": { "deployments": 1, "replicasets": 1, "services": 1 }
+        }
+    ],
+    "e976fcc461dc1b1ad177c083135393e90ded18b3707987d4b6adf78e86e687ed": [
+        {
+            "name": "nodejs",
+            "version": "",
+            "resources": { "deployments": 1, "replicasets": 1, "services": 1 }
+        }
+    ]
+}

pkg/gather/gather.go

@@ -65,7 +65,7 @@ func CreateAllGatherers(
 		gatherKubeConfig, gatherProtoKubeConfig, metricsGatherKubeConfig, alertsGatherKubeConfig,
 		anonymizer, configObserver,
 	)
-	workloadsGatherer := workloads.New(gatherProtoKubeConfig)
+	workloadsGatherer := workloads.New(gatherKubeConfig, gatherProtoKubeConfig)
 	conditionalGatherer := conditional.New(
 		gatherProtoKubeConfig, metricsGatherKubeConfig, gatherKubeConfig, configObserver, insightsClient,
 	)

pkg/gatherers/workloads/gather_helm_info.go

@@ -0,0 +1,167 @@
+package workloads
+
+import (
+	"context"
+	"crypto/sha256"
+	"encoding/hex"
+	"fmt"
+	"strings"
+
+	"k8s.io/apimachinery/pkg/runtime/schema"
+
+	"k8s.io/apimachinery/pkg/api/errors"
+	"k8s.io/client-go/dynamic"
+	"k8s.io/klog/v2"
+
+	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
+
+	"github.com/openshift/insights-operator/pkg/record"
+)
+
+const labelChartNameKey = "helm.sh/chart"
+
+// GatherHelmInfo Collects summarized info about the helm usage on a cluster
+// in a generic fashion
+//
+// ### API Reference
+// None
+//
+// ### Sample data
+// - docs/insights-archive-sample/config/helmchart_info.json
+//
+// ### Location in archive
+// - `config/helmchart_info.json`
+//
+// ### Config ID
+// `workloads/helmchart_info`
+//
+// ### Released version
+// - 4.15.0
+//
+// ### Backported versions
+// None
+func (g *Gatherer) GatherHelmInfo(ctx context.Context) ([]record.Record, []error) {
+	dynamicClient, err := dynamic.NewForConfig(g.gatherKubeConfig)
+	if err != nil {
+		return nil, []error{err}
+	}
+
+	return gatherHelmInfo(ctx, dynamicClient)
+}
+
+func gatherHelmInfo(
+	ctx context.Context,
+	dynamicClient dynamic.Interface,
+) ([]record.Record, []error) {
+	resources := []schema.GroupVersionResource{
+		{Group: "apps", Version: "v1", Resource: "replicasets"},
+		{Group: "apps", Version: "v1", Resource: "daemonsets"},
+		{Group: "apps", Version: "v1", Resource: "statefulsets"},
+		{Group: "", Version: "v1", Resource: "services"},
+		{Group: "apps", Version: "v1", Resource: "deployments"},
+	}
+
+	var errs []error
+	var records []record.Record
+	helmList := newHelmChartInfoList()
+
+	for _, resource := range resources {
+		listOptions := metav1.ListOptions{LabelSelector: "app.kubernetes.io/managed-by=Helm"}
+
+		items, err := dynamicClient.Resource(resource).List(ctx, listOptions)
+		if errors.IsNotFound(err) {
+			return nil, nil
+		}
+		if err != nil {
+			klog.V(2).Infof("Unable to list %s resource due to: %s", resource, err)
+			errs = append(errs, err)
+			continue
+		}
+
+		for _, item := range items.Items {
+			labels := item.GetLabels()
+
+			// Anonymize the namespace to make it unique identifier
+			hash, err := createHash(item.GetNamespace())
+			if err != nil {
+				klog.Errorf("unable to hash the namespace '%s': %v", labels[labelChartNameKey], err)
+				continue
+			}
+
+			name, version := helmChartNameAndVersion(labels[labelChartNameKey])
+			if name == "" && version == "" {
+				// some helm-maneged resource may not have reference to the chart
+				klog.Infof("unable to get HelmChart from %s on %s from %s.", resource.Resource, item.GetNamespace(), item.GetName())
+				continue
+			}
+
+			helmList.addItem(hash, resource.Resource, HelmChartInfo{
+				Name:    name,
+				Version: version,
+			})
+		}
+	}
+
+	if len(helmList.Namespaces) > 0 {
+		records = []record.Record{
+			{
+				Name: "config/helmchart_info",
+				Item: record.JSONMarshaller{Object: &helmList.Namespaces},
+			},
+		}
+	}
+
+	if len(errs) > 0 {
+		return records, errs
+	}
+
+	return records, nil
+}
+
+func createHash(chartName string) (string, error) {
+	h := sha256.New()
+	_, err := h.Write([]byte(chartName))
+	if err != nil {
+		return "", err
+	}
+
+	hashInBytes := h.Sum(nil)
+	hash := hex.EncodeToString(hashInBytes)
+
+	return hash, nil
+}
+
+func helmChartNameAndVersion(chart string) (name, version string) {
+	parts := strings.Split(chart, "-")
+
+	// no version found
+	if len(parts) == 1 {
+		return chart, ""
+	}
+
+	name = strings.Join(parts[:len(parts)-1], "-")
+
+	// best guess to get the version
+	version = parts[len(parts)-1]
+	// check for standard version format
+	if !strings.Contains(version, ".") {
+		// maybe it is a string version
+		if !isStringVersion(version) {
+			// not a valid version, add to name and version should be empty
+			name = fmt.Sprintf("%s-%s", name, version)
+			version = ""
+		}
+	}
+
+	return name, version
+}
+
+func isStringVersion(version string) bool {
+	stringVersions := []string{"latest", "beta", "alpha"}
+	for _, v := range stringVersions {
+		if v == version {
+			return true
+		}
+	}
+	return false
+}