Making thousands of ingresses available on independent K8s clusters
summary
How to provide high availability for thousands of Kubernetes Ingress resources across independent K8s clusters, using scalable, cluster-independent networking and control-plane patterns. It covers technologies and approaches for making large numbers of Ingresses resilient without coupling clusters, including load-balancing, failover and multi-cluster routing strategies compatible with standard Kubernetes Ingress implementations.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:57Numberly infrastructure
- 3:46Multi-cluster redundancy needs
- 5:10Classic load balancing limits
- 7:40X-Rasile project discovery
- 10:30Missing features and contributions
- 12:39TLS cert sync challenges
- 14:48Timeouts and retries in ingress
- 15:30Traffic weight management
- 17:19Multi-cluster app adaptation
- 18:33Helm charts and GitLab templates
- 19:36Monitoring multi-cluster deployments
- 20:19Cluster failover demo
- 21:26Temporary clusters for spikes
- 22:09Current limits and future work
related talks
42mKubernetes Networking at Scale
28mArchitecting Kubernetes Clusters for High Traffic Sites
20mInside Kubernetes Ingress
38mHow We Manage Traffic for Thousands of Pods in a Multicluster Kubernetes Environment
25mKubernetes Ingress: Do More with Less
49mAdvanced Ingress for Your Kubernetes Apps
37mJourney Through Kubernetes Ingress: Past, Present, and Beyond
06mKeynote: From One to Many, the Road to Multicluster