Disaster Recovery for your Kubernetes Clusters
summary
Describes a practical DR workflow that backs up Kubernetes API objects via the discovery API and snapshots persistent volumes on major clouds using Heptio Ark. Demonstrates restoring an nginx namespace and its ~1KB access log from a Rook PV, shows scheduled backups, namespace/resource/label filtering, hooks for fs freeze, and plugin extensibility for object and block storage. Restores create objects and log conflicts as warnings; large etcd or PV restores can take hours. Targets cloud provider snapshots; roll‑your‑own needed for unsupported volume types.
AI-written summary · about
chapters (15)
- 0:00Intro
- 1:35What happens when things go wrong
- 2:15Traditional disaster recovery
- 3:20Kubernetes Clusters
- 6:07Master and No Disaster Recovery
- 10:52Persistent volumes
- 11:58Heptio ARC
- 19:03Demo
- 24:02Join us
- 25:07What happens if the backup doesnt include the pod
- 26:47How to monitor a backup
- 29:32What causes Heptio to fail
- 30:47Types of resources that dont behave well
- 31:47Performance
- 33:12Resource Management
related talks
30mIs Your Kubernetes Disaster Recovery Actually Ready?
33mNavigating Disaster Recovery in Kubernetes and Crossplane: Strategies and Tools for Managing Stateful Workloads with GitOps for Platform Teams
35mClusters as Cattle: How to Seamlessly Migrate Apps Across Kubernetes Clusters
33mPlan to Fail: A Good Captain Doesn’t Sail Without Life Rafts
25mIntro & Deep Dive: Kubernetes Data Protection WG
31mKubernetes Backup and Migration Strategies with Velero
01h 00mApplication resilience in Kubernetes... certainty even before your DRP?
07mLightning Talk: Optimizing Kubernetes Multicluster Backup & Restore