# Hand-written PrometheusRule: Sloth-compatible metric names and labels (no Sloth binary / Go generator).
#
# Copy into a chart's templates/ as e.g. prometheus-rules-slo.yaml and merge values from
# slo-handwritten.values.example.yaml.
#
# - Seven short windows: `range` over `.Values.slo.burnRateWindows` (default Sloth set).
#   Record names: `slo:sli_error:ratio_rate<window>` (e.g. ratio_rate5m).
# - 30-day rolling average: `slo:sli_error:ratio_rate30d` (uses sum_over_time/count_over_time on the first window's 5m series).
# - Meta: objective, error budget, burn rates, budget remaining, `sloth_slo_info`.
# - Multi-window burn-rate alerts (14.4 / 6 / 3 / 1 factors) when `slo.alerting.enabled`.
#
# Alert expressions reference windows by index into the default list [5m, 30m, 1h, 2h, 6h, 1d, 3d].
# If you override `burnRateWindows`, keep the same seven durations in the same order, or adjust the alert blocks accordingly.
#

{{- $s := .Values.slo | default dict }}
{{- if $s.enabled }}
{{- $jobLine := default (printf "job=\"%s\"" .Chart.Name) $s.jobFilter }}
{{- $availObj := $s.availabilityObjective | default 0.999 | float64 }}
{{- $latObj := $s.latencyObjective | default 0.80 | float64 }}
{{- $le := $s.latencyHistogramLe | default "0.5" }}
{{- $partOf := $s.partOf | default "subDomain" }}
{{- $domain := $s.domain | default "domain" }}
{{- $ebAvail := subf 1.0 $availObj }}
{{- $ebLat := subf 1.0 $latObj }}
{{- $windows := $s.burnRateWindows | default (list "5m" "30m" "1h" "2h" "6h" "1d" "3d") }}
{{- $w0 := index $windows 0 }}
{{- $w1 := index $windows 1 }}
{{- $w2 := index $windows 2 }}
{{- $w3 := index $windows 3 }}
{{- $w4 := index $windows 4 }}
{{- $w5 := index $windows 5 }}
{{- $w6 := index $windows 6 }}
{{- $rb := $s.runbookUrl | default "" }}
{{- $alerting := ($s.alerting | default dict).enabled | default false }}
{{- $availPct := printf "%.10g" (mulf $availObj 100) }}
{{- $latPct := printf "%.10g" (mulf $latObj 100) }}

---
apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  annotations:
    saas.company.com/team: {{ .Values.global.team }}
    saas.company.com/environment: {{ .Values.global.environment }}
    saas.company.com/pop: {{ .Values.global.pop }}
  labels:
    app.kubernetes.io/component: SLO
    app.kubernetes.io/managed-by: helm-slo-handwritten
    app.kubernetes.io/instance: {{ .Release.Name }}
    app.kubernetes.io/name: {{ .Chart.Name }}
    app.kubernetes.io/part-of: {{ $partOf }}
    prometheus: infra
  name: {{ .Chart.Name }}-slo
spec:
  groups:
  - name: sloth-slo-sli-recordings-{{ .Chart.Name }}-requests-availability
    rules:
{{- range $w := $windows }}
    - record: slo:sli_error:ratio_rate{{ $w }}
      expr: |
        (
          (sum(rate(handler_http_response_seconds_count{namespace="{{ $.Release.Namespace }}",{{ $jobLine }},status_code=~"5.."}[{{ $w }}])))
          or
          (sum(rate(handler_http_response_seconds_count{namespace="{{ $.Release.Namespace }}",{{ $jobLine }}}[{{ $w }}])) * 0)
        )
        /
        clamp_min((sum(rate(handler_http_response_seconds_count{namespace="{{ $.Release.Namespace }}",{{ $jobLine }}}[{{ $w }}]))), 1e-9)
      labels:
        category: availability
        clusterType: {{ $.Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ $.Values.global.environment }}
        namespace: {{ $.Release.Namespace }}
        pop: {{ $.Values.global.pop }}
        severity: warning
        sloth_id: {{ $.Chart.Name }}-requests-availability
        sloth_service: {{ $.Chart.Name }}
        sloth_slo: requests-availability
        sloth_window: {{ $w }}
{{- end }}
    - record: slo:sli_error:ratio_rate30d
      expr: |
        sum_over_time(slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"}[30d])
        / ignoring (sloth_window)
        count_over_time(slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"}[30d])
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
        sloth_window: 30d
  - name: sloth-slo-meta-recordings-{{ .Chart.Name }}-requests-availability
    rules:
    - record: slo:objective:ratio
      expr: vector({{ $availObj }})
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
    - record: slo:error_budget:ratio
      expr: vector({{ $ebAvail }})
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
    - record: slo:time_period:days
      expr: vector(30)
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
    - record: slo:current_burn_rate:ratio
      expr: |
        slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"}
        / on(sloth_id, sloth_slo, sloth_service) group_left
        slo:error_budget:ratio{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"}
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
    - record: slo:period_burn_rate:ratio
      expr: |
        slo:sli_error:ratio_rate30d{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"}
        / on(sloth_id, sloth_slo, sloth_service) group_left
        slo:error_budget:ratio{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"}
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
    - record: slo:period_error_budget_remaining:ratio
      expr: 1 - slo:period_burn_rate:ratio{sloth_id="{{ .Chart.Name }}-requests-availability",
        sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"}
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
    - record: sloth_slo_info
      expr: vector(1)
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-availability
        sloth_mode: api-gen-prom
        sloth_objective: "{{ $availPct }}"
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-availability
        sloth_spec: prometheus/v1
        sloth_version: v0.16.0
{{- if $alerting }}
  - name: sloth-slo-alerts-{{ .Chart.Name }}-requests-availability
    rules:
    - alert: {{ .Chart.Name }}-HighErrorRateAvailability
      annotations:
        contact: {{ .Values.global.team }}
        {{- if $rb }}
        runbook_url: {{ $rb }}
        {{- end }}
        summary: High error rate (5xx) on '{{ .Chart.Name }}' requests
        title: (page) {{`{{ $labels.sloth_service }}`}} {{`{{ $labels.sloth_slo }}`}} SLO error budget burn rate is too fast.
      expr: |
        (
            max(slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (14.4 * {{ $ebAvail }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w2 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (14.4 * {{ $ebAvail }})) without (sloth_window)
        )
        or
        (
            max(slo:sli_error:ratio_rate{{ $w1 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (6 * {{ $ebAvail }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w4 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (6 * {{ $ebAvail }})) without (sloth_window)
        )
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_severity: page
    - alert: {{ .Chart.Name }}-HighErrorRateAvailability
      annotations:
        contact: {{ .Values.global.team }}
        {{- if $rb }}
        runbook_url: {{ $rb }}
        {{- end }}
        summary: High error rate (5xx) on '{{ .Chart.Name }}' requests
        title: (ticket) {{`{{ $labels.sloth_service }}`}} {{`{{ $labels.sloth_slo }}`}} SLO error budget burn rate is too fast.
      expr: |
        (
            max(slo:sli_error:ratio_rate{{ $w3 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (3 * {{ $ebAvail }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w5 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (3 * {{ $ebAvail }})) without (sloth_window)
        )
        or
        (
            max(slo:sli_error:ratio_rate{{ $w4 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (1 * {{ $ebAvail }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w6 }}{sloth_id="{{ .Chart.Name }}-requests-availability", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-availability"} > (1 * {{ $ebAvail }})) without (sloth_window)
        )
      labels:
        category: availability
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_severity: ticket
{{- end }}

  - name: sloth-slo-sli-recordings-{{ .Chart.Name }}-requests-latency
    rules:
{{- range $w := $windows }}
    - record: slo:sli_error:ratio_rate{{ $w }}
      expr: |
        ((
          sum(rate(handler_http_response_seconds_count{namespace="{{ $.Release.Namespace }}",{{ $jobLine }},status_code=~"2.."}[{{ $w }}]))
          -
          sum(rate(handler_http_response_seconds_bucket{namespace="{{ $.Release.Namespace }}",{{ $jobLine }},status_code=~"2..",le="{{ $le }}"}[{{ $w }}]))
        )
        )
        /
        clamp_min((sum(rate(handler_http_response_seconds_count{namespace="{{ $.Release.Namespace }}",{{ $jobLine }},status_code=~"2.."}[{{ $w }}]))), 1e-9)
      labels:
        category: latency
        clusterType: {{ $.Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ $.Values.global.environment }}
        namespace: {{ $.Release.Namespace }}
        pop: {{ $.Values.global.pop }}
        severity: warning
        sloth_id: {{ $.Chart.Name }}-requests-latency
        sloth_service: {{ $.Chart.Name }}
        sloth_slo: requests-latency
        sloth_window: {{ $w }}
{{- end }}
    - record: slo:sli_error:ratio_rate30d
      expr: |
        sum_over_time(slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"}[30d])
        / ignoring (sloth_window)
        count_over_time(slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"}[30d])
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
        sloth_window: 30d
  - name: sloth-slo-meta-recordings-{{ .Chart.Name }}-requests-latency
    rules:
    - record: slo:objective:ratio
      expr: vector({{ $latObj }})
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
    - record: slo:error_budget:ratio
      expr: vector({{ $ebLat }})
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
    - record: slo:time_period:days
      expr: vector(30)
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
    - record: slo:current_burn_rate:ratio
      expr: |
        slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"}
        / on(sloth_id, sloth_slo, sloth_service) group_left
        slo:error_budget:ratio{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"}
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
    - record: slo:period_burn_rate:ratio
      expr: |
        slo:sli_error:ratio_rate30d{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"}
        / on(sloth_id, sloth_slo, sloth_service) group_left
        slo:error_budget:ratio{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"}
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
    - record: slo:period_error_budget_remaining:ratio
      expr: 1 - slo:period_burn_rate:ratio{sloth_id="{{ .Chart.Name }}-requests-latency",
        sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"}
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
    - record: sloth_slo_info
      expr: vector(1)
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_id: {{ .Chart.Name }}-requests-latency
        sloth_mode: api-gen-prom
        sloth_objective: "{{ $latPct }}"
        sloth_service: {{ .Chart.Name }}
        sloth_slo: requests-latency
        sloth_spec: prometheus/v1
        sloth_version: v0.16.0
{{- if $alerting }}
  - name: sloth-slo-alerts-{{ .Chart.Name }}-requests-latency
    rules:
    - alert: {{ .Chart.Name }}-HighErrorRateLatency
      annotations:
        contact: {{ .Values.global.team }}
        {{- if $rb }}
        runbook_url: {{ $rb }}
        {{- end }}
        summary: High latency error rate on '{{ .Chart.Name }}' requests
        title: (page) {{`{{ $labels.sloth_service }}`}} {{`{{ $labels.sloth_slo }}`}} SLO error budget burn rate is too fast.
      expr: |
        (
            max(slo:sli_error:ratio_rate{{ $w0 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (14.4 * {{ $ebLat }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w2 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (14.4 * {{ $ebLat }})) without (sloth_window)
        )
        or
        (
            max(slo:sli_error:ratio_rate{{ $w1 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (6 * {{ $ebLat }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w4 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (6 * {{ $ebLat }})) without (sloth_window)
        )
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_severity: page
    - alert: {{ .Chart.Name }}-HighErrorRateLatency
      annotations:
        contact: {{ .Values.global.team }}
        {{- if $rb }}
        runbook_url: {{ $rb }}
        {{- end }}
        summary: High latency error rate on '{{ .Chart.Name }}' requests
        title: (ticket) {{`{{ $labels.sloth_service }}`}} {{`{{ $labels.sloth_slo }}`}} SLO error budget burn rate is too fast.
      expr: |
        (
            max(slo:sli_error:ratio_rate{{ $w3 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (3 * {{ $ebLat }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w5 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (3 * {{ $ebLat }})) without (sloth_window)
        )
        or
        (
            max(slo:sli_error:ratio_rate{{ $w4 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (1 * {{ $ebLat }})) without (sloth_window)
            and
            max(slo:sli_error:ratio_rate{{ $w6 }}{sloth_id="{{ .Chart.Name }}-requests-latency", sloth_service="{{ .Chart.Name }}", sloth_slo="requests-latency"} > (1 * {{ $ebLat }})) without (sloth_window)
        )
      labels:
        category: latency
        clusterType: {{ .Values.global.clusterType }}
        domain: {{ $domain }}
        environment: {{ .Values.global.environment }}
        namespace: {{ .Release.Namespace }}
        pop: {{ .Values.global.pop }}
        severity: warning
        sloth_severity: ticket
{{- end }}
{{- end }}