Skip to main content
MK
Mohan Kumar BN Platform Engineer · Kubestronaut
VIDEOS // MKBN-TECH KNOWLEDGE HUB

Production Kubernetes & OpenShift Video Library

Hands-on technical masterclasses and rapid architecture insights from the MKBN-Tech YouTube Channel (700+ subscribers · 45K+ views). All video players use privacy-enhanced click-to-load embeds.

Subscribe on YouTube (@mkbn-tech) ↗
FEATURED MASTERCLASS2026-10-02

The Zero Downtime Production OpenShift HA Cluster Upgrade

Live, end-to-end production upgrade of a High-Availability Red Hat OpenShift cluster with zero downtime, zero unscheduled pod evictions, and zero observability blind spots—covering the Production Pre-Check Gate, Worker MCP Pausing Pattern, 2-Worker Drain Trap, Control Plane rolling upgrades, 10-minute soak window, and DR insurance.

  • Production Pre-Check Gate & etcd quorum validation
  • Worker MachineConfigPool (MCP) Pausing Pattern & 2-Worker Drain Trap
  • Live Control Plane rolling upgrades & 10-minute telemetry soak window
Watch on YouTubeRead Case Study →
Full Production Masterclass

The Zero Downtime Production OpenShift HA Cluster Upgrade

Live, end-to-end production upgrade of a High-Availability Red Hat OpenShift cluster with zero downtime, zero unscheduled pod evictions, and zero observability blind spots—covering the Production Pre-Check Gate, Worker MCP Pausing Pattern, 2-Worker Drain Trap, Control Plane rolling upgrades, 10-minute soak window, and DR insurance.

  • Production Pre-Check Gate & etcd quorum validation
  • Worker MachineConfigPool (MCP) Pausing Pattern & 2-Worker Drain Trap
  • Live Control Plane rolling upgrades & 10-minute telemetry soak window
Watch on YouTube
Case StudyRepo
Full Architecture Walkthrough

Zero to Production: A Complete GitOps CI/CD Platform on OpenShift

Build a real production-grade GitOps and Serverless CI/CD platform on Red Hat OpenShift—from git push to live multi-cluster production using OpenShift Pipelines (Tekton), OpenShift GitOps (Argo CD), and Knative Serverless.

  • End-to-end Tekton CI build & image promotion pipeline
  • Declarative Argo CD delivery across SNO staging and HA production
  • Knative Serverless workload autoscaling and zero-drift governance
Watch on YouTube
Case StudyRepo
Deep-Dive Tutorial

Stop Using Shared Staging! Build Ephemeral Environments with Argo CD

Eliminate staging bottlenecks by automatically spinning up isolated, full-stack ephemeral Kubernetes preview environments on every Pull Request using Argo CD ApplicationSets and tearing them down on merge.

  • Argo CD Pull Request Generator & ApplicationSet templating
  • Automated per-PR namespace, ingress route & resource quota provisioning
  • Zero-touch environment teardown upon PR merge or closure
Watch on YouTube
Case StudyRepo
Production Architecture Guide

How Modern Platform Teams ACTUALLY Deploy to Kubernetes

Move beyond sanitized localhost demos and explore how enterprise platform engineering teams structure multi-environment repositories, Kustomize overlays, Helm charts, CI gates, and GitOps promotion workflows.

  • Separation of application source repos (`finance-api`) and GitOps config repos (`finance-api-ops`)
  • Environment promotion across Dev, Staging, and HA Production
  • Automated guardrails, health probes, and rollback safety
Watch on YouTube
Case StudyRepo
Module 3 Masterclass

From kubeadmin to Corporate SSO – Real World OpenShift Authentication & RBAC

Harden Red Hat OpenShift cluster access by replacing the bootstrap kubeadmin user with enterprise OAuth identity providers, GroupSync, least-privilege ClusterRoles, and tenant RoleBindings.

  • Configuring OpenShift OAuth identity providers & corporate SSO
  • Designing multi-tenant RBAC hierarchies for dev, SRE, and security teams
  • Safely removing bootstrap kubeadmin credentials in production
Watch on YouTube
Case StudyRepo
Module 2 Masterclass

Make Your OpenShift Apps Production Ready: Storage, Health Checks, Autoscaling & RBAC

Transform basic container deployments into resilient, production-ready OpenShift workloads with persistent storage (PVCs), startup/liveness/readiness probes, HorizontalPodAutoscalers, resource quotas, and security contexts.

  • Configuring Liveness, Readiness, and Startup probes to prevent traffic blackholes
  • PersistentVolumeClaims, ResourceQuotas, LimitRanges, and HPA tuning
  • Working harmoniously with OpenShift Security Context Constraints (SCCs)
Watch on YouTube
Case StudyRepo
Module 1 Masterclass

From Kubernetes to OpenShift: What Changes and Why Enterprises Care

Architectural comparison between upstream vanilla Kubernetes and Red Hat OpenShift—exploring why enterprises adopt OpenShift for integrated CoreOS immutability, OLM operators, Routes, SCC security, and built-in GitOps/Pipelines.

  • Red Hat Enterprise Linux CoreOS (RHCOS) & immutable node lifecycle
  • OpenShift Routes, integrated image registry, and Security Context Constraints
  • Enterprise day-2 operations across hybrid and on-prem data centers
Watch on YouTube
Case StudyRepo
Hands-On Cloud Guide

Run AWS, Azure, and GCP Locally on Your Laptop

Build, test, and validate multi-cloud AWS, Azure, and GCP workflows locally using Floci and local cloud emulators—eliminating cloud billing risk and accelerating Terraform/CLI development loops.

  • Zero-cost local cloud emulation for AWS, Azure, and GCP APIs
  • Rapid local testing of Terraform modules and cloud CLI scripts
  • Shift-left infrastructure validation before cloud deployment
Watch on YouTube
Case Study
Full Tutorial

Supercharge Kubernetes with kubectl-ai + Ollama (Local AIOps Copilot)

Integrate Google’s open-source kubectl-ai with local Ollama Large Language Models (LLMs) to create a privacy-first Kubernetes AI copilot that generates manifests, diagnoses cluster issues, and executes operations.

  • Connecting kubectl-ai to self-hosted Ollama models with zero external data leakage
  • Natural-language Kubernetes manifest synthesis and live cluster inspection
  • Safe human-in-the-loop confirmation workflows for cluster changes
Watch on YouTube
ArticleCase StudyRepo
Full Tutorial

AI-Powered Kubernetes Troubleshooting with K8sGPT & Local LLM

Deploy the K8sGPT CLI and in-cluster Kubernetes Operator backed by local LLMs to continuously scan clusters, analyze CrashLoopBackOff/Service/Ingress misconfigurations, and produce actionable SRE remediation guidance.

  • Installing K8sGPT CLI and in-cluster K8sGPT Operator
  • Automated root-cause analysis for broken Pods, PVCs, and Ingresses
  • Integrating K8sGPT diagnostic metrics with Prometheus & Grafana
Watch on YouTube
ArticleCase StudyRepo
Full Tutorial

Local AI Stack: Ollama + NVIDIA GPU Operator + Open WebUI on Kubernetes

Build a self-hosted, GPU-accelerated local AI platform on Kubernetes using MicroK8s, NVIDIA GPU Operator, Ollama, and Open WebUI to serve private LLMs for engineering and SRE workflows.

  • Configuring NVIDIA GPU Operator passthrough for Kubernetes pods
  • Deploying Ollama and Open WebUI with persistent model storage
  • Serving local LLMs for internal developer and SRE automation
Watch on YouTube
ArticleCase StudyRepo
Full Tutorial

RKE2 High-Availability Cluster Installation With Ansible

Automate the deployment of a security-hardened, FIPS-ready Rancher RKE2 High-Availability Kubernetes cluster across Ubuntu nodes using idempotent Ansible playbooks.

  • RKE2 architecture and CIS security hardening advantages
  • Automated multi-control-plane bootstrap with Ansible roles
  • VIP load balancing and kubeconfig verification
Watch on YouTube
ArticleCase StudyRepo
Full Tutorial

Install an External Ceph Distributed Storage Cluster Using Ansible

Provision a production-grade, self-healing Ceph distributed storage cluster using Ansible to provide block (RBD), file (CephFS), and object storage for Kubernetes and virtualization platforms.

  • Ceph Monitor, Manager, and OSD storage topology design
  • Automated cluster bootstrap and pool provisioning with Ansible
  • Integrating external Ceph storage with Kubernetes CSI drivers
Watch on YouTube
ArticleCase Study
Full Tutorial

Kubernetes Monitoring Stack: Prometheus, Grafana, and Loki

Deploy a complete cloud-native observability stack on Kubernetes combining Prometheus metrics, Alertmanager routing, Grafana Loki centralized logging, and unified operational dashboards.

  • Deploying kube-prometheus-stack and Loki via Helm
  • Correlating Prometheus metrics with Loki pod log streams in Grafana
  • Designing Golden Signal dashboards for cluster and node health
Watch on YouTube
ArticleCase StudyRepo
Full Tutorial

Running Cloud-Native Jenkins on Kubernetes with Dynamic Pod Agents

Deploy Jenkins on Kubernetes backed by persistent storage and dynamic ephemeral pod agents that scale on demand for Maven, Python, and container build pipelines.

  • Stateful Jenkins controller deployment on Kubernetes
  • Configuring Kubernetes Cloud plugin for ephemeral build agent pods
  • Automated CI pipeline execution for Java and Python microservices
Watch on YouTube
ArticleRepo
Full Tutorial

Deploy Self-Hosted Enterprise GitLab on Ubuntu with Ansible

Fully automated installation and configuration of a self-hosted GitLab instance on Ubuntu using Ansible—complete with HTTPS, container registry, and CI runner integration.

  • Idempotent Ansible playbook for GitLab Omnibus provisioning
  • Securing GitLab web access and configuring GitLab CI Runners
  • Backup and maintenance automation for self-hosted source control
Watch on YouTube
ArticleRepo
Full Tutorial

Automate Kubernetes Cluster Deployment with Ansible

Zero-touch provisioning of a multi-node Kubernetes cluster (control plane + worker nodes, containerd runtime, Calico CNI) using reusable Ansible roles and Jinja templates.

  • Automating kernel modules, sysctl networking, and containerd via Ansible
  • Unattended kubeadm init, CNI installation, and worker node join tokens
  • Repeatable cluster bootstrapping in minutes
Watch on YouTube
ArticleCase StudyRepo
Full Tutorial

Install Rancher Multi-Cluster Manager with Let’s Encrypt on Kubernetes

Deploy SUSE Rancher on Kubernetes with cert-manager and automated Let’s Encrypt TLS certificates for centralized multi-cluster operations and RBAC governance.

  • Installing NGINX Ingress Controller and cert-manager ClusterIssuers
  • Helm deployment of HA Rancher server with automated TLS renewal
  • Importing and governing downstream Kubernetes clusters
Watch on YouTube
ArticleCase Study
Full Tutorial

Meet Longhorn: Cloud-Native Distributed Block Storage for Kubernetes

Implement highly available, replicated persistent block storage on Kubernetes using Longhorn—including multi-node volume replication, snapshots, and stateful Gitea workload validation.

  • Longhorn storage engine architecture and node prerequisite checks
  • Dynamic PVC provisioning via the Longhorn CSI driver
  • Volume replication, snapshot recovery, and stateful app resilience
Watch on YouTube
ArticleCase StudyRepo
Full Tutorial

Build a High-Availability Kubernetes Cluster on Bare-Metal with HAProxy & Keepalived

Step-by-step architecture and implementation of a production 3-master stacked etcd HA Kubernetes cluster on Ubuntu 24.04 bare-metal servers fronted by redundant HAProxy and Keepalived VRRP load balancers.

  • Configuring Active/Backup Keepalived VRRP virtual IP failover
  • Layer-4 HAProxy load balancing across 3 Kubernetes API servers
  • Bootstrapping stacked etcd HA control plane nodes with kubeadm
Watch on YouTube
ArticleRepo
60s Rapid Insight

Zero-Config Azure Model Context Protocol (MCP) Server

Connect Claude to Microsoft Azure with zero configuration using the official Azure Model Context Protocol (MCP) Bundle for natural-language cloud architecture and resource inspection.

  • Official Azure MCP Bundle architecture
  • Zero-config authentication and cloud resource querying
Watch on YouTube
60s Rapid Insight

Run AWS Locally on Your Laptop with Floci

How Floci helps platform engineers build, test, and validate AWS cloud workflows locally on a laptop without a cloud account, billing risk, or setup friction.

  • Local AWS service emulation with Floci
  • Risk-free local cloud workflow testing
Watch on YouTube
60s Rapid Insight

Why Kubernetes Pods Get OOMKilled & The Exact Fix

Understand how Linux kernel cgroups enforce Kubernetes memory requests vs. hard limits (Exit Code 137) and how to right-size JVM/container workloads to eliminate OOMKills.

  • Kubernetes scheduler requests vs. Linux OOMKiller cgroup limits
  • Diagnosing Exit Code 137 and right-sizing container memory limits
Watch on YouTube
60s Rapid Insight

Stop Confusing Kubernetes With OpenShift In Under 60 Seconds

The clearest mental model comparing upstream Kubernetes (the engine) with Red Hat OpenShift (the turn-key enterprise platform with integrated OS, networking, security, and CI/CD).

  • Engine vs. Complete Enterprise Platform mental model
  • Built-in OpenShift security, routing, and lifecycle automation
Watch on YouTube
60s Rapid Insight

Why DevOps Engineers Are Replacing Docker with Daemonless Podman

Why enterprise Linux and OpenShift engineers favor Podman’s daemonless, rootless fork/exec architecture and native systemd/Kubernetes pod generation over traditional root Docker daemons.

  • Daemonless & rootless container security architecture
  • Drop-in CLI compatibility and native Kubernetes YAML generation
Watch on YouTube
60s Rapid Insight

Build a Single Node OpenShift (SNO)

Kubernetes gives you the engine to drive those containers but enterprises don’t just need an engine. They need the whole car: the dashboard, the seatbelts, the service manual, and someone to call when it breaks at 3 AM. That, in simple terms, is where Red Hat OpenShift comes in.

  • Red Hat OpenShift architecture & hands-on implementation
  • Production-ready patterns from MKBN-Tech
Watch on YouTube
60s Rapid Insight

The Zero Downtime Production OpenShift HA Cluster Upgrade

In this production masterclass, we execute a live, end-to-end upgrade of a High-Availability OpenShift cluster with zero downtime, zero pod evictions, and zero blind spots.

  • Red Hat OpenShift architecture & hands-on implementation
  • Production-ready patterns from MKBN-Tech
Watch on YouTube
60s Rapid Insight

Build Ephemeral Environments with Argo CD

Imagine opening a PR and instantly getting:

  • GitOps & CI/CD architecture & hands-on implementation
  • Production-ready patterns from MKBN-Tech
Watch on YouTube