Platform Common CRDs

Component Category

Platform internals / custom resource definitions

Component Description

Platform Common CRDs is an internal Helm chart (dp-common-crds v0.0.3) that installs shared custom resource definitions required by multiple platform components. It deploys into kube-system to ensure CRDs are available cluster-wide before any component that depends on them starts.

The chart currently deploys the CoreOS monitoring CRDs:

• ServiceMonitor — defines Prometheus scrape targets for Kubernetes services
• PodMonitor — defines Prometheus scrape targets for individual pods

These CRDs are consumed by components across the platform (Keycloak, OpenFGA, Temporal, Argo Workflows, KubeAI, ToolHive, and others) to expose metrics to the Grafana/Prometheus observability stack.

Why It Is Used

In BullSequana AI, Platform Common CRDs provides a single point of ownership for CRDs that would otherwise be duplicated or installed inconsistently across components. Deploying them at sync wave 0 guarantees they exist before any other platform component attempts to create custom resources.

Learn More

This is an internal platform chart. The CRDs originate from the prometheus-operator project.

Deployment notes

This component deploys at sync wave 0, the earliest in the platform. It has no user-facing configuration beyond the default chart version.

Interacts With

• Grafana and Prometheus, which consume the ServiceMonitor and PodMonitor resources to discover scrape targets.
• ArgoCD, which reconciles the CRDs as part of the common parent application.