Backing up your Kubernetes cluster is critical for disaster recovery and ensuring continuity. Besides KubeBackup, other tools offer complementary and robust solutions for backing up both cluster configurations and data. In this post, we’ll explore Velero, Longhorn, and other tools that are designed to handle backups in Kubernetes environments.

1. Velero

Velero

Velero is one of the most popular tools for backing up Kubernetes clusters and persistent volumes.

Key Features:

  • Cluster and Namespace Backups: Supports both full-cluster and per-namespace backups.
  • Restore on Demand: Easily restore resources and volumes to a cluster.
  • Migration Support: Move applications between clusters.
  • S3-Compatible Storage: Store backups in AWS S3 or any compatible storage.

Installation:

velero install --provider aws --bucket my-bucket --backup-location-config region=us-east-1

Usage:

# Create a backup
velero backup create cluster-backup --include-namespaces '*' --wait

# Restore from backup
velero restore create --from-backup cluster-backup --wait

2. Longhorn

Longhorn

Longhorn is a lightweight, reliable distributed block storage system for Kubernetes. It is an excellent tool for persistent volume backups.

Key Features:

  • Snapshots and Backup to S3: Create PVC snapshots and back them up to external storage.
  • Disaster Recovery: Quickly restore backed-up volumes to recover applications.
  • UI for Backup Management: A simple UI for managing snapshots and backups.

Installation:

helm repo add longhorn https://charts.longhorn.io
helm repo update
helm install longhorn longhorn/longhorn --namespace longhorn-system

Example Usage:

  1. Take a Snapshot:

    kubectl annotate pvc my-pvc -n my-namespace backup=true

  2. Restore a Volume from Snapshot: Use the Longhorn UI or CLI tools to restore the snapshot to a new PVC.

3. etcd Backups

Since etcd stores the state of your Kubernetes cluster, backing up etcd is essential for cluster recovery.

Taking a Manual etcd Snapshot:

rke2 etcd-snapshot save --name snapshot-$(date +%Y%m%d-%H%M%S)

Restoring from Snapshot:

rke2 etcd-snapshot restore --name snapshot-YYYYMMDD-HHMMSS

Using etcd snapshots along with YAML exports ensures a full recovery path for cluster state and object configurations.

4. Restic

Restic is an open-source backup tool that supports various storage backends, including S3, Azure, and local disks. While it’s not specific to Kubernetes, it can be used to back up persistent data.

Key Features:

  • End-to-End Encryption: Ensures that backups are encrypted.
  • Deduplication: Efficiently stores data by eliminating duplicates.

Example Restic Backup Command:

export RESTIC_PASSWORD="my-password"
export AWS_ACCESS_KEY_ID="your-access-key"
export AWS_SECRET_ACCESS_KEY="your-secret-key"

restic -r s3:s3.amazonaws.com/my-bucket backup /data

Restore Command:

restic -r s3:s3.amazonaws.com/my-bucket restore latest --target /restore-path

5. TrilioVault

TrilioVault is an enterprise-grade backup solution for Kubernetes applications. It is designed for organizations that require robust backup policies and compliance.

Key Features:

  • Application-Aware Backups: Supports stateful applications.
  • Incremental Backups: Reduces backup size by only storing changes.
  • Role-Based Access Control (RBAC): Manage access to backup resources.

Best Practices for Kubernetes Backups

  1. Automate Backups: Use tools like Velero or Longhorn with cron jobs or CI/CD pipelines.
  2. Offsite Storage: Store backups in S3 or cloud storage to protect against local failures.
  3. Test Restores: Regularly verify that your backups can be restored successfully.
  4. Monitor Backup Health: Use Prometheus and AlertManager to track backup success and detect issues.

Conclusion

Backups are a critical part of Kubernetes operations. Tools like Velero, Longhorn, KubeBackup, etcd snapshots, and Restic ensure you have a full backup strategy, covering both configuration and persistent data. Incorporate these tools into your workflow and automate the process to safeguard your clusters against unexpected failures.