Inside Kubernetes
Run resilient workloads across scheduling, networking, storage, and failures
Kubernetes runs your workloads through control loops you cannot see: a scheduler places pods, controllers roll out replicas, kube-proxy rewrites traffic, and CSI drivers attach volumes. This Deep Dive traces those mechanics from the API server to the node, so you can predict how the cluster reacts to a deploy, a node failure, or a resource squeeze, and diagnose Pending pods, unreachable Services, and stuck mounts from the evidence the cluster gives you.
See the Invisible
Interactive simulators visualise what's hidden from view.
Hands-On Labs
Step through executions tick by tick. Manipulate state.
Why, Not Just What
Understand the reasoning behind every design decision.
Quizzes & Cheatsheets
Verify your understanding and keep a quick reference handy.
Get Certified
Earn a shareable certificate to prove your deep expertise.
Become the Engineer Who Supervises AI
As AI generates more code, understanding what that code does becomes more valuable, not less. Someone must verify AI output, debug failures, and make architectural decisions.
Build Your Architectural EdgeStop treating the cluster as a black box that sometimes says no
You apply a manifest and a pod sits in Pending. A rollout stalls at half the replicas. A Service resolves in DNS but traffic never arrives. Each of these has a precise cause inside the reconciliation loop, the scheduler's filtering and scoring, or the Service data plane, and guessing at it costs hours. This Deep Dive walks the full path a workload takes, from desired state in etcd to a running container on a node, so that when the cluster misbehaves you can name which component made which decision and read the events, conditions, and probe results that prove it.
What You Will Be Able to Do
Predict placement and disruption
Reason about where the scheduler will put a pod given requests, affinity, taints, and topology spread constraints, and use PodDisruptionBudgets and priority to control what gets evicted under pressure or during a node drain.
Trace traffic from client to container
Follow a request through Ingress or Gateway API routing, Service virtual addresses, EndpointSlices, and the kube-proxy data plane, and identify which hop drops it when NetworkPolicy or DNS misbehaves.
Diagnose from cluster evidence
Read events, conditions, logs, and probe results to explain Pending pods, CrashLoopBackOff, unready workloads, and storage attachment failures, instead of restarting things and hoping.
What's Covered
How the API server, etcd, controllers, scheduler, and kubelet turn desired state into running pods, and how Deployments, StatefulSets, DaemonSets, and Jobs manage replicas, rollouts, and rollbacks on top of that loop.
Pod identity, init and sidecar containers, ConfigMap and Secret delivery, startup, readiness, and liveness probes, restart policy, and graceful termination with preStop hooks and grace periods.
Requests and limits as scheduling inputs and runtime constraints, filtering and scoring, affinity and anti-affinity, taints and tolerations, QoS classes, preemption, pressure eviction, and PodDisruptionBudgets.
CNI connectivity, Service types and EndpointSlices, cluster DNS, NetworkPolicy enforcement, and layer 7 routing through Ingress and the Gateway API, including how the two coexist during migration.
PersistentVolume and claim binding, StorageClass dynamic provisioning, StatefulSet per-replica claims, snapshots and expansion, HorizontalPodAutoscaler behavior, node drains, and troubleshooting across the whole path.
The Curriculum
Comprehensive Lessons! Each with theory, interactive simulation, and quiz.
Reconciliation and the Control Plane
Pod Runtime, Configuration, and Health
Workload Controllers and Rollouts
Scheduling, Capacity, and Disruption
Service Networking and Policy
Ingress and Gateway API Traffic
Persistent Storage and Stateful Workloads
Scaling and Operational Troubleshooting
Ready to see what's really happening?
All deep dives included with your subscription. Cancel anytime.