{{- if and .Values.metrics.prometheusRule.enabled (.Capabilities.APIVersions.Has "monitoring.coreos.com/v1") -}}
apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: {{ include "karpenter.fullname" . }}
  namespace: {{ .Release.Namespace }}
  labels:
    {{- include "karpenter.labels" . | nindent 4 }}
  {{- with .Values.metrics.prometheusRule.additionalLabels }}
    {{- toYaml . | nindent 4 }}
  {{- end }}
  {{- with .Values.additionalAnnotations }}
  annotations:
    {{- toYaml . | nindent 4 }}
  {{- end }}
spec:
  groups:
    - name: karpenter-ibm.rules
      interval: {{ .Values.metrics.prometheusRule.interval | default "30s" }}
      rules:
        # High-level operational alerts
        - alert: KarpenterIBMControllerDown
          expr: up{job="{{ include "karpenter.fullname" . }}"} == 0
          for: {{ .Values.metrics.prometheusRule.alerts.controllerDown.for | default "5m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.controllerDown.severity | default "critical" }}
            service: karpenter-ibm
          annotations:
            summary: "Karpenter IBM Cloud Provider controller is down"
            description: "The Karpenter IBM Cloud Provider controller has been down for more than {{ .Values.metrics.prometheusRule.alerts.controllerDown.for | default "5m" }}"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        - alert: KarpenterIBMHighErrorRate
          expr: rate(controller_runtime_reconcile_errors_total{job="{{ include "karpenter.fullname" . }}"}[5m]) > {{ .Values.metrics.prometheusRule.alerts.highErrorRate.threshold | default "0.1" }}
          for: {{ .Values.metrics.prometheusRule.alerts.highErrorRate.for | default "10m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.highErrorRate.severity | default "warning" }}
            service: karpenter-ibm
          annotations:
            summary: "Karpenter IBM Cloud Provider has high error rate"
            description: "Karpenter IBM Cloud Provider error rate is {{`{{ $value | humanizePercentage }}`}} over the last 5 minutes"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        - alert: KarpenterIBMSlowReconciliation
          expr: histogram_quantile(0.99, rate(controller_runtime_reconcile_duration_seconds_bucket{job="{{ include "karpenter.fullname" . }}"}[5m])) > {{ .Values.metrics.prometheusRule.alerts.slowReconciliation.threshold | default "60" }}
          for: {{ .Values.metrics.prometheusRule.alerts.slowReconciliation.for | default "15m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.slowReconciliation.severity | default "warning" }}
            service: karpenter-ibm
          annotations:
            summary: "Karpenter IBM Cloud Provider reconciliation is slow"
            description: "99th percentile reconciliation time is {{`{{ $value | humanizeDuration }}`}}"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        # IBM Cloud API specific alerts
        - alert: KarpenterIBMAPIErrors
          expr: rate(karpenter_ibm_api_requests_total{status_code!~"2.."}[5m]) > {{ .Values.metrics.prometheusRule.alerts.apiErrors.threshold | default "0.05" }}
          for: {{ .Values.metrics.prometheusRule.alerts.apiErrors.for | default "5m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.apiErrors.severity | default "warning" }}
            service: karpenter-ibm
          annotations:
            summary: "High IBM Cloud API error rate"
            description: "IBM Cloud API error rate is {{`{{ $value | humanizePercentage }}`}} for {{`{{ $labels.service }}`}} operations"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        - alert: KarpenterIBMAPIRateLimit
          expr: rate(karpenter_ibm_api_requests_total{status_code="429"}[5m]) > {{ .Values.metrics.prometheusRule.alerts.apiRateLimit.threshold | default "0.01" }}
          for: {{ .Values.metrics.prometheusRule.alerts.apiRateLimit.for | default "2m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.apiRateLimit.severity | default "warning" }}
            service: karpenter-ibm
          annotations:
            summary: "IBM Cloud API rate limiting detected"
            description: "IBM Cloud API rate limiting is occurring at {{`{{ $value | humanizePercentage }}`}} for {{`{{ $labels.service }}`}} operations"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        # Node provisioning alerts
        - alert: KarpenterIBMNoNodeProvisioning
          expr: rate(karpenter_nodes_total[30m]) == 0 and on() karpenter_unschedulable_pods > {{ .Values.metrics.prometheusRule.alerts.noNodeProvisioning.podThreshold | default "0" }}
          for: {{ .Values.metrics.prometheusRule.alerts.noNodeProvisioning.for | default "10m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.noNodeProvisioning.severity | default "critical" }}
            service: karpenter-ibm
          annotations:
            summary: "No node provisioning activity with pending pods"
            description: "No nodes have been provisioned in 30 minutes despite {{`{{ $value }}`}} unschedulable pods"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        - alert: KarpenterIBMSlowNodeProvisioning
          expr: histogram_quantile(0.99, rate(karpenter_ibm_instance_provisioning_duration_seconds_bucket[10m])) > {{ .Values.metrics.prometheusRule.alerts.slowNodeProvisioning.threshold | default "600" }}
          for: {{ .Values.metrics.prometheusRule.alerts.slowNodeProvisioning.for | default "15m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.slowNodeProvisioning.severity | default "warning" }}
            service: karpenter-ibm
          annotations:
            summary: "Slow node provisioning detected"
            description: "99th percentile node provisioning time is {{`{{ $value | humanizeDuration }}`}}"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        # Cache performance alerts
        - alert: KarpenterIBMLowCacheHitRate
          expr: karpenter_ibm_cache_hit_rate < {{ .Values.metrics.prometheusRule.alerts.lowCacheHitRate.threshold | default "80" }}
          for: {{ .Values.metrics.prometheusRule.alerts.lowCacheHitRate.for | default "30m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.lowCacheHitRate.severity | default "info" }}
            service: karpenter-ibm
          annotations:
            summary: "Low cache hit rate detected"
            description: "{{`{{ $labels.cache_type }}`}} cache hit rate is {{`{{ $value | humanizePercentage }}`}}"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        # Resource utilization alerts
        - alert: KarpenterIBMHighMemoryUsage
          expr: process_resident_memory_bytes{job="{{ include "karpenter.fullname" . }}"} / 1024 / 1024 > {{ .Values.metrics.prometheusRule.alerts.highMemoryUsage.threshold | default "500" }}
          for: {{ .Values.metrics.prometheusRule.alerts.highMemoryUsage.for | default "30m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.highMemoryUsage.severity | default "warning" }}
            service: karpenter-ibm
          annotations:
            summary: "High memory usage detected"
            description: "Karpenter IBM Cloud Provider memory usage is {{`{{ $value | humanize }}`}}MB"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

        - alert: KarpenterIBMHighCPUUsage
          expr: rate(process_cpu_seconds_total{job="{{ include "karpenter.fullname" . }}"}[5m]) * 100 > {{ .Values.metrics.prometheusRule.alerts.highCPUUsage.threshold | default "80" }}
          for: {{ .Values.metrics.prometheusRule.alerts.highCPUUsage.for | default "30m" }}
          labels:
            severity: {{ .Values.metrics.prometheusRule.alerts.highCPUUsage.severity | default "warning" }}
            service: karpenter-ibm
          annotations:
            summary: "High CPU usage detected"
            description: "Karpenter IBM Cloud Provider CPU usage is {{`{{ $value | humanizePercentage }}`}}"
            runbook_url: "{{ .Values.metrics.prometheusRule.runbookUrl | default "https://github.com/kubernetes-sigs/karpenter-provider-ibm-cloud/blob/main/docs/troubleshooting.md" }}"

    {{- if .Values.metrics.prometheusRule.additionalRuleGroups }}
    {{- toYaml .Values.metrics.prometheusRule.additionalRuleGroups | nindent 4 }}
    {{- end }}
{{- end -}}