Skip to main content

Feature Overview

Vidra is a Kubernetes operator designed to manage custom resources with flexibility and safety in complex environments. Below is an overview of its core features and capabilities, including both current functionalities and potential future developments.

Current Features​

Namespace Handling​

Vidra tracks and manages resource ownership across namespaces, ensuring consistency and preventing conflicts during the resource lifecycle.

Safe Handling of Name Changes​

Vidra includes logic to safely manage resource renames:

  • Detects renamed resources and reconciles to the desired state
  • Cleans up old resources to avoid orphaned objects
  • Ensures safe reconciliation after name updates

Multiple VidraResource Instances per InfrahubSync​

Supports creating multiple VidraResource instances from a single InfrahubSync, enabling:

  • Management of multiple resources from one Infrahub artifact
  • Simplified configuration without duplicating InfrahubSync definitions

Shared Resource Management​

Allows multiple VidraResource instances to manage the same Kubernetes resource:

  • Enables shared ownership (e.g., namespaces) across different VidraResource and InfrahubSync configurations
  • Prevents deletion of resources still in use by other managers
  • Tracks ownership to ensure safe lifecycle operations

Efficient Caching​

Vidra downloads artifacts only if the checksum has changed, reducing unnecessary network calls and improving performance.

Helm Chart Deployment​

Vidra is available as a Helm chart (OCI and standard Helm repository), allowing:

  • Installation via helm repo add and helm install
  • Flexible deployment configurations
  • Easy integration into Helm-based infrastructure

Multicluster Support​

Vidra supports multi-cluster environments:

  • Uses kubeconfig contexts to read/write across clusters (stored in Kubernetes Secrets)
  • Reconciles resources consistently across multiple environments
  • Maintains unique identity and ownership tracking per cluster

Continuous Delivery of Virtual Machines​

Supports continuous delivery of VirtualMachine resources via KubeVirt:

  • Automates creation and updates based on Infrahub artifacts
  • Continuously reconciles VM state with the desired configuration

Event-Driven Reconciliation​

Vidra uses an event-driven model to trigger reconciliation:

  • Responds immediately to resource changes
  • Dynamically adds Informers only for managed resources
  • Minimizes overhead while supporting any Kubernetes resource type
  • Reduces latency in updates and syncs
  • Can be enabled per InfrahubSync or globally

Finalizers for Safe Cleanup​

Finalizers ensure that:

  • Managed resources are cleaned up if the VidraResource is deleted
  • Cleanup logic is safely executed

Experimental Capabilities (Not Yet Fully Validated)​

Vidra’s architecture supports advanced scenarios that are available but not yet fully tested:

Advanced CRs​

Potential for managing complex CRs such as:

  • Network configurations (e.g., Kubenet, SDC) — Infrahub artifacts could define network policies and configurations
  • Cloud-native infrastructure (e.g., Crossplane) — Infrahub artifacts could define Crossplane resources for cloud services
  • Any other Kubernetes resource type (CR or not)

These capabilities are technically feasible but require further testing and validation before being considered production-ready.

Future Improvements​

Webhook-Based Reconciliation (Planned/In Progress)​

Vidra is evolving toward a fully event-based Continuous Integration Operator. Planned webhook support will:

  • Trigger real-time reconciliation on Infrahub changes
  • Reduce update latency
  • Remove reliance on frequent periodic resyncs
  • Provide more immediate feedback and state syncing

Sync to Other Platforms​

The VidraResource abstraction is independent of Infrahub, allowing for future support of:

  • Helm: Manage resources via Helm charts
  • Git: Enable GitOps-style syncing from repositories
  • Other platforms: Any system providing Kubernetes manifests can be integrated

Enhanced Observability​

Upcoming improvements could include:

  • Metrics and logging for operational insights
  • Tracing to follow resource lifecycle events

User Interface​

The managed resource and the status of the InfrahubSync and VidraResource resources could be visualized in a user interface, providing:

  • Visualization of resource state and dependencies
  • Dashboards for real-time monitoring and management