Skip to main content

Datadog Integration

Overview

OpsWorker integrates with Datadog as an alert source. Datadog monitors send alerts to OpsWorker via webhook, and OpsWorker investigates the underlying Kubernetes issue using its standard agent toolchain.

OpsWorker does not currently query Datadog metrics or APM data directly during investigations. For metric and log queries during investigation, use the Grafana MCP integration — it covers PromQL, LogQL, and Grafana dashboards, including dashboards backed by Datadog datasources if you've configured them in Grafana.

Webhook Setup

1. Create a Webhook in Datadog

  1. In Datadog, go to Integrations → Webhooks
  2. Click New Webhook
  3. Name: opsworker
  4. URL: Your OpsWorker webhook URL (from the portal → Integrations → select the cluster → Datadog)
  5. Leave payload as default
  6. Click Save

2. Add to Monitors

Add @webhook-opsworker to the notification text of monitors you want OpsWorker to investigate. Example:

@webhook-opsworker

The {{check_type.name}} check has been triggered for {{host.name}}.

You can route only specific monitors to OpsWorker (recommended) or route everything (high alert volume — combine with alert rules to filter what triggers investigations).

3. Verify

Trigger a test alert from a configured monitor. Check the OpsWorker portal under Alerts for the incoming signal. If the alert matches an alert rule with auto-investigation enabled, an investigation starts within seconds.

Supported Monitor Types

Monitor TypeOpsWorker ReceivesInvestigation Quality
MetricYesGood — alert name and tags map cleanly to Kubernetes resources
APMYesGood when traces include kube_* tags
LogYesGood for error-rate and pattern detection alerts
ProcessYesLimited — process monitors often lack namespace/pod context
CompositeYesQuality depends on the underlying monitors

Investigation works best when Datadog tags include the Kubernetes context — kube_namespace, pod_name, deployment — so OpsWorker can map the alert directly to a resource topology.

Next Steps