Medik8s - The Medical Toolkit For Your Kubernetes Cluster
summary
medicaids provides automated node failure detection, safe fencing and fast remediation for Kubernetes by combining a node-healthcheck operator that creates remediation CRs and a self-node-remediation daemonset that arms a watchdog, powers off or reboots failing nodes, force-deletes pods and volume attachments, and asks peer agents if API server is unreachable. demo uses a 60s timeout (configurable from default 5 minutes), reboots the node and reschedules workloads successfully, and components are available via OperatorHub and GitHub.
AI-written summary · about
chapters (13)
- 0:00Introduction to medik8s
- 1:06Kubernetes basics recap
- 6:22Custom resource definitions and operators
- 7:04Problems medik8s solves
- 8:52Failure detection in medik8s
- 10:38Self node remediation operator
- 13:29Standalone remediation benefits
- 13:51Installing medik8s operators
- 15:40Demo setup overview
- 17:55Node health check configuration
- 18:59Simulating node failure
- 20:06Remediation process in action
- 21:35Demo conclusion and resources
related talks
18mK8s Troubleshooting Demystified: Five Best Practices
46mBuilding a zero downtime Kubernetes cluster
28mThe Magic of Kubernetes Self-Healing Capabilities
23mDon't Let Users Find Your Outages: Synthetic Monitoring for Kubernetes Platforms
40mOperating a Self-Healing Bare-Metal Kubernetes Platform at Global Scale
30mEdge Kubernetes and Healthcare: A Next-Gen Architecture that Scales Across Thousands of Locations
11mK8s Resources Ensuring Performance and Optimization
45mKubernetes Operability Tooling