example-slo-spec.yaml

apiVersion: sloth.slok.dev/v1
kind: PrometheusServiceLevel
metadata:
  name: example
  namespace: monitoring
  labels:
    release: "prometheus-stack"
spec:
  service: "example"
  labels:
    owner: "platform"
  slos:
    # Metric sample:
    #   http_request_duration_seconds_count{container="app", endpoint="metrics", instance="10.49.61.69:9080", job="autopilot-backend-workspace-service", method="GET", namespace="autopilot", pod="autopilot-backend-workspace-rollout-d976d4977-fxwsc", route="/v1/workspaces/:workspaceId", service="autopilot-backend-workspace-service", status_code="200"}
    - name: http-error-rate
      objective: 99.99
      description: "99.99% of the requests should be successful"
      sli:
        plugin:
          id: lokalise/http-error-rate
          options:
            metricName: http_request_duration_seconds_count
            serviceLabelName: service
            serviceLabelValue: autopilot-backend-workspace-service
            errorLabelName: status_code
            errorLabelValue: (5..|429)
            additionalLabels: route=~"/v1/workspaces/.*"
            minimumRequestsPerSecond: "10"
      alerting:
        name: HighErrorRate
        pageAlert:
          annotations:
            name: "High error rate in '{{ $labels.sloth_service }}'"
            runbookurl: todo
        ticketAlert:
          disable: true

    # Metric sample:
    #   http_request_duration_seconds_bucket{container="app", endpoint="metrics", instance="10.49.63.203:9080", job="autopilot-backend-import-service", le="5", method="POST", namespace="autopilot", pod="autopilot-backend-import-rollout-68d6f9bd9-xsfg7", route="/v1/projects/:projectId/actions/upload-content-item-collection", service="autopilot-backend-import-service", status_code="202"}
    - name: http-latency
      objective: 99.99
      description: "99.99% of the requests should be below 500ms"
      sli:
        plugin:
          id: lokalise/http-latency
          options:
            metricName: http_request_duration_seconds_bucket
            serviceLabelName: service
            serviceLabelValue: autopilot-backend-import-service
            upperLimitBucket: "5"
            additionalLabels: route=~"/v1/projects/.*"
            minimumRequestsPerSecond: "10"
      alerting:
        name: HighLatency
        pageAlert:
          annotations:
            name: "High latency in '{{ $labels.sloth_service }}'"
            runbookurl: todo
        ticketAlert:
          disable: true