[release-4.16] OCPBUGS-33917: add alert data to upgrade health in oc adm upgrade status

pkg/cli/admin/upgrade/status/alerts.go

@@ -0,0 +1,99 @@
+package status
+
+import (
+	"time"
+
+	configv1 "github.com/openshift/api/config/v1"
+)
+
+// Alerts that will be included in the health upgrade evaluation, even if they were triggered before the upgrade began.
+type AllowedAlerts map[string]struct{}
+
+var allowedAlerts AllowedAlerts = map[string]struct{}{
+	"PodDisruptionBudgetLimit":   {},
+	"PodDisruptionBudgetAtLimit": {},
+}
+
+func (al AllowedAlerts) Contains(alert string) bool {
+	_, exists := al[alert]
+	return exists
+}
+
+type AlertLabels struct {
+	AlertName string `json:"alertname,omitempty"`
+	Name      string `json:"name,omitempty"`
+	Namespace string `json:"namespace,omitempty"`
+	Reason    string `json:"reason,omitempty"`
+	Severity  string `json:"severity,omitempty"`
+}
+
+type AlertAnnotations struct {
+	Description string `json:"description,omitempty"`
+	Summary     string `json:"summary,omitempty"`
+	Runbook     string `json:"runbook_url,omitempty"`
+}
+
+type Alert struct {
+	Labels                  AlertLabels      `json:"labels,omitempty"`
+	Annotations             AlertAnnotations `json:"annotations,omitempty"`
+	State                   string           `json:"state,omitempty"`
+	Value                   string           `json:"value,omitempty"`
+	ActiveAt                time.Time        `json:"activeAt,omitempty"`
+	PartialResponseStrategy string           `json:"partialResponseStrategy,omitempty"`
+}
+
+// Stores alert data returned by thanos
+type AlertData struct {
+	Status string `json:"status"`
+	Data   Data   `json:"data"`
+}
+
+type Data struct {
+	Alerts []Alert `json:"alerts"`
+}
+
+func parseAlertDataToInsights(alertData AlertData, startedAt time.Time) []updateInsight {
+	var alerts []Alert = alertData.Data.Alerts
+	var updateInsights []updateInsight = []updateInsight{}
+
+	for _, alert := range alerts {
+		if startedAt.After(alert.ActiveAt) && !allowedAlerts.Contains(alert.Labels.AlertName) {
+			continue
+		}
+		if alert.State == "pending" {
+			continue
+		}
+		updateInsights = append(updateInsights, updateInsight{
+			startedAt: alert.ActiveAt,
+			impact: updateInsightImpact{
+				level:       alertImpactLevel(alert.Labels.Severity),
+				impactType:  unknownImpactType,
+				summary:     "Alert: " + alert.Annotations.Summary,
+				description: alert.Annotations.Description,
+			},
+			remediation: updateInsightRemediation{reference: alert.Annotations.Runbook},
+			scope: updateInsightScope{
+				scopeType: scopeTypeCluster,
+				resources: []scopeResource{{
+					kind:      scopeGroupKind{group: configv1.GroupName, kind: "Alert"},
+					namespace: alert.Labels.Namespace,
+					name:      alert.Labels.AlertName,
+				}},
+			},
+		})
+	}
+	return updateInsights
+}
+
+func alertImpactLevel(ail string) impactLevel {
+	switch ail {
+	case "warning":
+		return warningImpactLevel
+	case "critical":
+		return criticalInfoLevel
+	case "info":
+		return infoImpactLevel
+	default:
+		return infoImpactLevel
+	}
+}

pkg/cli/admin/upgrade/status/alerts_test.go

@@ -0,0 +1,156 @@
+package status
+
+import (
+	"reflect"
+	"testing"
+	"time"
+
+	configv1 "github.com/openshift/api/config/v1"
+)
+
+func TestParseAlertDataToInsights(t *testing.T) {
+	now := time.Now()
+
+	// Define test cases
+	tests := []struct {
+		name          string
+		alertData     AlertData
+		startedAt     time.Time
+		expectedCount int
+	}{
+		{
+			name: "Empty Alerts",
+			alertData: AlertData{
+				Data: Data{Alerts: []Alert{}},
+			},
+			startedAt:     now,
+			expectedCount: 0,
+		},
+		{
+			name: "Alert Active After Start Time",
+			alertData: AlertData{
+				Data: Data{
+					Alerts: []Alert{
+						{ActiveAt: now.Add(10 * time.Minute), Labels: AlertLabels{Severity: "critical", Namespace: "default", AlertName: "NodeDown"}, Annotations: AlertAnnotations{Summary: "Node is down"}},
+						{ActiveAt: now.Add(-10 * time.Minute), Labels: AlertLabels{Severity: "warning", Namespace: "default", AlertName: "DiskSpaceLow"}, Annotations: AlertAnnotations{Summary: "Disk space low"}},
+					},
+				},
+			},
+			startedAt:     now,
+			expectedCount: 1,
+		},
+		{
+			name: "Alert Active Before Start Time",
+			alertData: AlertData{
+				Data: Data{
+					Alerts: []Alert{
+						{ActiveAt: now.Add(-10 * time.Minute), Labels: AlertLabels{Severity: "warning", Namespace: "default", AlertName: "DiskSpaceLow"}, Annotations: AlertAnnotations{Summary: "Disk space low"}},
+					},
+				},
+			},
+			startedAt:     now,
+			expectedCount: 0,
+		},
+		{
+			name: "Alert Active Before Start Time, Allowed",
+			alertData: AlertData{
+				Data: Data{
+					Alerts: []Alert{
+						{ActiveAt: now.Add(-20 * time.Minute), Labels: AlertLabels{Severity: "info", Namespace: "default", AlertName: "PodDisruptionBudgetAtLimit"}, Annotations: AlertAnnotations{Summary: "PodDisruptionBudgetAtLimit is at limit"}},
+						{ActiveAt: now.Add(-20 * time.Minute), Labels: AlertLabels{Severity: "info", Namespace: "default", AlertName: "AlertmanagerReceiversNotConfigured"}, Annotations: AlertAnnotations{Summary: "Receivers (notification integrations) are not configured on Alertmanager"}},
+					},
+				},
+			},
+			startedAt:     now,
+			expectedCount: 1,
+		},
+		{
+			name: "Alert Active Before Start Time, Not Allowed",
+			alertData: AlertData{
+				Data: Data{
+					Alerts: []Alert{
+						{ActiveAt: now.Add(-20 * time.Minute), Labels: AlertLabels{Severity: "info", Namespace: "default", AlertName: "AlertmanagerReceiversNotConfigured"}, Annotations: AlertAnnotations{Summary: "Receivers (notification integrations) are not configured on Alertmanager"}},
+					},
+				},
+			},
+			startedAt:     now,
+			expectedCount: 0,
+		},
+	}
+
+	// Execute test cases
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			insights := parseAlertDataToInsights(tt.alertData, tt.startedAt)
+			if got := len(insights); got != tt.expectedCount {
+				t.Errorf("parseAlertDataToInsights() = %v, want %v", got, tt.expectedCount)
+			}
+		})
+	}
+}
+
+func TestParseAlertDataToInsightsWithData(t *testing.T) {
+	now := time.Now()
+
+	tests := []struct {
+		name             string
+		alertData        AlertData
+		startedAt        time.Time
+		expectedInsights []updateInsight
+	}{
+		{
+			name: "Alert Active After Start Time",
+			alertData: AlertData{
+				Data: Data{
+					Alerts: []Alert{
+						{ActiveAt: now.Add(10 * time.Minute), Labels: AlertLabels{Severity: "critical", Namespace: "default", AlertName: "NodeDown"}, Annotations: AlertAnnotations{Summary: "Node is down"}},
+					},
+				},
+			},
+			startedAt: now,
+			expectedInsights: []updateInsight{
+				{
+					startedAt: now.Add(10 * time.Minute),
+					impact: updateInsightImpact{
+						level:      alertImpactLevel("critical"),
+						impactType: unknownImpactType,
+						summary:    "Alert: Node is down",
+					},
+					scope: updateInsightScope{
+						scopeType: scopeTypeCluster,
+						resources: []scopeResource{
+							{
+								kind:      scopeGroupKind{group: configv1.GroupName, kind: "Alert"},
+								namespace: "default",
+								name:      "NodeDown",
+							},
+						},
+					},
+				},
+			},
+		},
+		{
+			name: "Alert Active Before Start Time",
+			alertData: AlertData{
+				Data: Data{
+					Alerts: []Alert{
+						{ActiveAt: now.Add(-10 * time.Minute), Labels: AlertLabels{Severity: "warning", Namespace: "default", AlertName: "DiskSpaceLow"}, Annotations: AlertAnnotations{Summary: "Disk space low"}},
+					},
+				},
+			},
+			startedAt:        now,
+			expectedInsights: []updateInsight{},
+		},
+	}
+
+	// Execute test cases
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			insights := parseAlertDataToInsights(tt.alertData, tt.startedAt)
+			if !reflect.DeepEqual(insights, tt.expectedInsights) {
+				t.Errorf("parseAlertDataToInsights() got %#v, want %#v", insights, tt.expectedInsights)
+			}
+		})
+	}
+
+}