Longhorn for Your Edge Deployments
summary
Longhorn runs on a 4-node Raspberry Pi K3s cluster deployed via Rancher K3s and provisions volumes with 3 replicas and anti‑affinity so a single Pi failure does not lose data. The demo uses a stateful data‑aggregator (terminationGracePeriod=10s) and a daemonset data collector, flips one Pi off, shows Longhorn delete stuck parts, rebuild a replica and move the aggregator, and sets up Amazon S3 backup schedules (daily, retain 7) and DR restore that work end to end.
AI-written summary · about