Skip to main content

Survive a Node Failure — Live HA Failover

Kill a node mid-traffic. Watch a replica take over. No data loss, no downtime.

4 min

Coming soon

This walkthrough is in production. Subscribe on YouTube to catch it first.

Watch it, or read it

The full walkthrough below — every screen from the video, with the narration transcribed. Skim it, search it, or read it in your own language.

Survive a Node Failure — Live HA Failover — screen 1 — Four nodes, replication factor two: every partition has a primary and a replica on a different node. That's the setup we're about to test by killing a node.

Survive a Node Failure — Live HA Failover — screen 2 — A client writes and reads continuously in the background. We delete a node's pod out from under it — and through the whole failover, reads and writes keep succeeding; the promoted replicas serve them without missing a beat.

Survive a Node Failure — Live HA Failover — screen 3 — The console sees the node go, promotes the replicas on the survivors, then rebuilds the lost replica back to full replication once the pod rejoins — no manual step.

Survive a Node Failure — Live HA Failover — screen 4 — Lose a node, keep serving, lose no data. Four nodes, RF2, tolerant of a failure by design.

More videos

Getting Started — a Distributed Cache for .NET in Minutes

Securing a Distributed .NET Cache — Authentication & RBAC

TLS & Mutual TLS for a Distributed .NET Cache

Distributed In-Memory Cache Architecture, Explained

Linear Scaling Benchmark — 1 → 2 → 3 Nodes, Live

Any Language, One Store — Zaris Speaks Redis

Drop-in Redis — Point StackExchange.Redis at Zaris

Native Data Structures in .NET — Hashes, Lists, Sets, Sorted Sets

Redis Streams, Native — Event Log + Consumer Groups

ASP.NET Core Caching Without Redis — IDistributedCache & HybridCache

Elastic Scaling — Add Nodes Live

Pub/Sub in .NET — Cluster-Wide Messaging

Get started Subscribe on YouTube