Kubernetes Cluster Administration Training Course

Course overview

Kubernetes administrators are expected to keep clusters available, secure and cost-controlled while application teams release continuously. That means diagnosing failing workloads across nodes, networking and storage; managing upgrades without unnecessary disruption; enforcing access boundaries; and recovering services when components fail. This course addresses the operational gap between deploying a containerised application and running a production Kubernetes platform with defined controls, repeatable procedures and evidence for technical decisions.

Participants work through core cluster administration duties using Kubernetes command-line and configuration practices. They configure and troubleshoot cluster components, workloads, Services, ingress paths, persistent storage and scheduling; implement RBAC, NetworkPolicies and Pod Security controls; plan node and control-plane maintenance; and investigate incidents using events, logs and resource metrics. The course also covers deployment safety through rolling updates, probes, disruption budgets and backup-and-restore procedures for cluster state.

Instruction combines focused technical briefings with guided work in a realistic multi-node Kubernetes environment. Participants perform administrator tasks from the command line, analyse deliberately introduced failures and make operational trade-offs in team scenarios. They leave with a completed Kubernetes Cluster Operations Runbook containing maintenance checklists, access-control patterns, troubleshooting steps, upgrade sequencing, backup validation actions and an improvement plan that can be adapted to their own platform.

The course is suited to professionals supporting self-managed Kubernetes clusters or managed services such as Amazon EKS, Azure Kubernetes Service and Google Kubernetes Engine. It is particularly valuable for teams that need stronger operational consistency before scaling Kubernetes use across multiple application squads.

Course objectives

By the end of this course, participants will be able to:

• Configure Kubernetes namespaces, resource quotas, LimitRanges and service accounts for controlled multi-team cluster use
• Troubleshoot failing Pods, Services, DNS resolution, scheduling decisions and node conditions using kubectl diagnostics
• Implement RBAC roles, role bindings, admission controls and NetworkPolicies for least-privilege cluster access
• Manage Deployments, StatefulSets, DaemonSets, probes and PodDisruptionBudgets to maintain workload availability
• Provision and diagnose PersistentVolumes, PersistentVolumeClaims, StorageClasses and StatefulSet storage patterns
• Plan and execute safe node maintenance using cordon, drain, taints, tolerations and workload rescheduling
• Design a Kubernetes upgrade, backup and restore procedure covering control-plane state and application recovery checks
• Produce a Kubernetes Cluster Operations Runbook with incident, maintenance, security and recovery procedures

Benefits of attending

Target competencies

Who should attend

• Kubernetes Administrators — who operate clusters and need repeatable methods for maintenance, troubleshooting and recovery
• DevOps Engineers — who build and support delivery platforms used by multiple application teams
• Site Reliability Engineers — who need to improve service resilience, observability and incident response in Kubernetes environments
• Cloud Platform Engineers — who administer managed Kubernetes services alongside cloud networking, identity and storage
• Systems Administrators — who are moving from virtual machine operations to container orchestration platforms
• Technical Leads — who set operational standards and need to assess the risks of Kubernetes platform changes

Requirements and prerequisites

Participants should be comfortable working at a Linux command line, navigating files, editing YAML and using basic shell commands. Prior exposure to containers is required, including images, registries, containers and the purpose of Docker or a compatible runtime. Participants should also understand basic TCP/IP concepts, DNS, ports, HTTP and storage terminology. Experience deploying simple Kubernetes workloads with kubectl, Pods, Deployments and Services is strongly recommended. This is not a beginner Kubernetes course; it does not require prior control-plane administration, scripting expertise, cloud certifications or production-cluster access.

Training methodology

The instructor uses short architecture briefings followed by guided administration labs in a multi-node Kubernetes environment. Participants work with kubectl, manifests and diagnostic output to resolve failed scheduling, service-discovery, storage and access-control scenarios. Small groups compare recovery and upgrade options against availability requirements, then document the selected procedure. Daily debriefs connect lab actions to production change control. On the final day, each participant consolidates tested commands, decision points and escalation steps into a Kubernetes Cluster Operations Runbook and application plan.

Course outline

Day 1: Cluster architecture and workload operations

• Kubernetes control-plane and worker-node component responsibilities
• kubectl contexts, namespaces, imperative commands and declarative manifests
• Pod lifecycle phases, container states, events and restart behaviour
• Deployments, ReplicaSets and rolling update configuration
• Services, EndpointSlices and internal service discovery
• Scheduler decisions, resource requests, limits and Quality of Service classes
• Namespaces, ResourceQuotas and LimitRanges for multi-team governance

Workshop: Participants deploy a constrained multi-service application and produce a namespace configuration with quotas, limits and verified service connectivity.

Day 2: Diagnostics, networking and storage

• Systematic kubectl troubleshooting using get, describe, logs, exec and events
• CrashLoopBackOff, ImagePullBackOff and Pending Pod diagnosis
• CoreDNS troubleshooting and Kubernetes DNS record patterns
• Service types, kube-proxy behaviour and traffic-routing checks
• Ingress resources, ingress controller routing and TLS secret configuration
• PersistentVolumes, PersistentVolumeClaims, StorageClasses and access modes
• StatefulSets, volume claim templates and persistent workload recovery

Workshop: Participants investigate a deliberately broken application and produce an incident record identifying DNS, Service and persistent-storage faults with corrective actions.

Day 3: Security and policy enforcement

• Authentication methods, kubeconfig credentials and service account tokens
• RBAC Roles, ClusterRoles, RoleBindings and ClusterRoleBindings
• Least-privilege permission design and kubectl auth can-i verification
• NetworkPolicies for ingress and egress traffic segmentation
• Pod Security Admission standards and namespace-level enforcement
• Secrets, ConfigMaps and secure configuration delivery practices
• Admission controllers, audit logging and policy governance considerations

Workshop: Participants secure a shared cluster namespace and produce tested RBAC and NetworkPolicy manifests for separate developer, operator and workload identities.

Day 4: Availability, maintenance and upgrades

• Liveness, readiness and startup probes for reliable traffic management
• PodDisruptionBudgets and voluntary disruption planning
• Node conditions, capacity checks and kubelet operational signals
• Node labels, selectors, affinity, anti-affinity, taints and tolerations
• Cordon, drain, uncordon and DaemonSet-aware maintenance procedures
• Control-plane and worker-node Kubernetes version upgrade sequencing
• etcd backup, restore principles and recovery validation checkpoints

Workshop: Participants plan and execute a simulated node-maintenance and version-upgrade window, producing a change plan with disruption, rollback and validation steps.

Day 5: Operational readiness and runbook design

• Cluster health checks for API server, nodes, workloads and system namespaces
• Resource metrics interpretation using metrics-server and kubectl top
• Capacity management using requests, limits, quotas and node allocatable resources
• Incident triage workflows and evidence capture from events and logs
• Backup schedules, restore tests and application recovery dependencies
• Managed Kubernetes operational boundaries for EKS, AKS and GKE
• Kubernetes Cluster Operations Runbook structure and ownership model

Workshop: Participants complete and peer-review a Kubernetes Cluster Operations Runbook and 90-day improvement plan for a supplied production platform scenario.

Tools & standards covered

Kubernetes, kubectl, etcd, metrics-server

A typical training day

Live online deliveries follow the same structure in the East Africa Time zone, with shorter screen blocks and longer breaks.

What the fee includes

How you can take this course

Certification

Participants who complete the full five days receive the Skillset Development Certificate of Completion, stating the course title, course code, dates and delivery format — suitable for professional-development records and employer reimbursement.

Frequently asked questions