Skip to content

Hibernator

Hibernator Logo

Declarative Kubernetes operator for suspending and restoring cloud infrastructure during off-hours.

What is Hibernator?

Hibernator is a Kubernetes operator that provides centralized, declarative management for suspending and restoring cloud resources during user-defined off-hours. It extends beyond Kubernetes to manage heterogeneous cloud infrastructure (EKS, RDS, EC2, and more) with dependency-aware orchestration and auditable execution.

Key Capabilities

  • Timezone-aware scheduling with start/end times and day-of-week patterns
  • Schedule exceptions with an on-demand schedule override (extend, suspend, replace)
  • Dependency orchestration using DAG, Staged, Parallel, or Sequential strategies
  • Pluggable executor model for AWS (EKS, RDS, EC2, Karpenter)
  • Isolated runner jobs with scoped RBAC, IRSA, and projected ServiceAccount tokens
  • Real-time progress streaming via gRPC or HTTP webhooks between runners and control plane
  • Event-driven notifications via Slack, Telegram, or generic webhooks on lifecycle events
  • Durable restore metadata persisted in ConfigMaps for safe recovery

Why Hibernator?

Problem: Teams running non-production environments (DEV/STG) waste resources during off-hours. Ad-hoc scripts lack coordination, auditability, and safe restore semantics when dealing with dependencies across Kubernetes clusters, databases, and compute instances.

Solution: Hibernator provides intent-driven infrastructure suspension with:

  • Declarative HibernatePlan CRDs defining what to suspend, not how
  • Controller-managed dependency resolution preventing race conditions
  • Central status ledger with execution history, logs, and restore artifact references
  • GitOps-friendly configuration with validation webhooks