Managing Kubernetes: What You Need to Know About Day 2 - Craig Tracey, Heptio
summary
After deploy the focus is on predictable Day‑2: keep etcd external (3–5 nodes), tune kubelet reserved resources (dynamic config coming), and choose AZ strategy deliberately (single multi‑AZ cluster vs cluster‑per‑AZ affects MTTR). Use OIDC via Dex + Gangway for onboarding, enforce RBAC, resource quotas and PodSecurityPolicies, and run sonobuoy and Ark for conformance and backup. Prefer controllers over ad hoc containers — a custom controller ≈20 lines solved service‑discovery churn for a customer.
AI-written summary · about
chapters (11)
- 0:00Intro
- 0:18Meet Craig Tracey
- 2:34Who is this talk for
- 4:34Kubernetes is cultural
- 5:32Think simplicity
- 6:28Dont boil the ocean
- 8:44Three areas to focus on
- 13:03Availability
- 16:09User Management
- 19:59Admission Control
- 27:43Wrap Up
related talks
06mDay 2 with Kubernetes - This Is Fine!
32mLaying the Foundation: Real World Kubernetes Deployment Patterns
30mManaged Kubernetes Service Day Zero Survival Pack
36mKubernetes Day 2: Cluster Operations [I]
47mKubernetes Day 2 to 5
33mManaged Kubernetes Service: Day Zero Survival Pack
53mFlying Above the Clouds: Securing Kubernetes
46mWhat I wish I knew about Kubernetes bef…