Infra Backup Operator

The Infra Backup Operator is a Juju charm designed to work seamlessly with the Velero Operator. When related, it automatically applies the necessary configuration to enable backups of Kubernetes resources that are not tied to workloads, but to the cluster’s own configuration and infrastructure. This ensures that critical namespaces, system components, and cluster-level settings are preserved, providing a reliable safeguard for restoring the operational foundation of your Kubernetes environment.

The infra backup operator is designed to work with any kind of Kubernetes cluster and the backup are separated into two groups:

• Cluster Infra Backup
• Namespaced infra backup

Cluster Infra Backup

• All cluster-scoped resources (e.g., CRDs, CSRs, ClusterRoles, StorageClasses, Namespaces)
• By default, all namespaced resources in the following namespaces:
  • kube-system
  • kube-public
  • metallb-system

This behavior is controlled by the charm’s namespaces configuration option, allowing you to customize which namespaces are included in the backup process.

Note: While Persistent Volumes (PVs) are cluster-scoped resources, they are not included in the backup, as they are tied to workloads and should be backed up by the applications themselves. Likewise, Pods are excluded because they are ephemeral by nature and should be managed through higher-level controllers such as Deployments, StatefulSets, or DaemonSets.

Namespaced infra backup

This will be responsible for backup the following kubernetes resources from all namespaces:

• Security and Access Control
  • Role, RoleBinding
  • NetworkPolicy
  • CiliumNetworkPolicies
  • ResourceQuota
  • LimitRange
  • ServiceAccount
  • Gateway and Routes (HttpRoute, TCPRoute, GRPCRoute, TLSRoute)
  • Ingress
• Configuration and Environment
  • ConfigMap
  • Secret
  • HPA / VPA
  • Jobs / CronJobs