DatadogSLO CRD

このページは日本語には対応しておりません。随時翻訳に取り組んでいます。
翻訳に関してご質問やご意見ございましたら、お気軽にご連絡ください。
Soft deprecation notice: For new SLO resources, prefer DatadogGenericResource with type: slo. DatadogSLO remains supported for existing users, but DatadogGenericResource is the preferred path for new Datadog API capabilities.

To create a Service Level Objective (SLO), you can use the Datadog Operator and DatadogSLO custom resource definition (CRD).

Prerequisites

Setup

  1. Create a file with the spec of your DatadogSLO deployment configuration.

    Example: Monitor-based SLO

    datadog-slo.yaml

       apiVersion: datadoghq.com/v1alpha1
   kind: DatadogSLO
   metadata:
     name: example-slo-monitor3
     namespace: system
   spec:
     name: example-slo-monitor3
     description: "This is an example monitor SLO from datadog-operator"
     monitorIDs:
       - 1234
     tags:
       - "service:example"
       - "env:prod"
     targetThreshold: "99.9"
     timeframe: "7d"
     type: "monitor"

    Example: Metric-based SLO

    datadog-slo.yaml

       apiVersion: datadoghq.com/v1alpha1
   kind: DatadogSLO
   metadata:
     name: example-slo
     namespace: system
   spec:
     name: example-slo
     description: "This is an example metric SLO from datadog-operator"
     query:
       denominator: "sum:requests.total{service:example,env:prod}.as_count()"
       numerator: "sum:requests.success{service:example,env:prod}.as_count()"
     tags:
       - "service:example"
       - "env:prod"
     targetThreshold: "99.9"
     timeframe: "7d"
     type: "metric"

    For all available configuration options, see the Create an SLO object API reference.

  2. Deploy your DatadogSLO:

    kubectl apply -f /path/to/your/datadog-slo.yaml

Additional examples

Metric-based SLO with Universal Service Monitoring