High Availability under Linux - Haute disponibilité sous Linux
From infrastructure to orchestration
Charles Sabourdin, Sébastien Rohaut
Published: 2024
ChatGPT's summary
📘 Genre
Infrastructure & DevOps / High Availability (HA) / Linux Systems
💡 Core Idea
Ensuring availability, redundancy, and reliability of Linux-based services—from hardware to orchestration—by applying DevOps principles. This book guides system engineers to build highly available environments across all layers: infrastructure, services, automation, orchestration, and resilience.
🧠 Key Concepts & Lessons
1. Fundamentals of High Availability
The book begins with designing a Java application running on Tomcat, then builds a fault-tolerant architecture using RAID, network bonding, DNS, etc.—focusing on HA best practices at the infrastructure level.
2. Automation with Docker and Ansible
The authors introduce a demo app as a guiding thread to demonstrate how to automate image creation and deployment using Docker and Ansible—laying the foundation of a DevOps workflow.
3. Network & Service Availability
You’ll learn how to make your network and services resilient using load balancers and tools like NGINX, HAProxy, VRRP with Keepalived, and Quagga—covering IP failover and routing redundancy.
4. Orchestration with Kubernetes
The book then shifts to orchestration, introducing Kubernetes step by step using Minikube, kubectl, and the API—covering how to abstract and schedule resources for highly available deployments.
5. Advanced Kubernetes Deployment
Deep dive into core Kubernetes objects:
- Pods, ReplicaSets, Deployments
- Services, Endpoints
- ConfigMaps, Secrets
- StatefulSets, PersistentVolumes, StorageClasses
- Ingress controllers, Jobs, CronJobs, DaemonSets
6. High Availability Storage
The book explores challenges in storage HA, like single points of failure, and how to mitigate them using redundant NFS clusters, XFS, and discusses the pros/cons of NAS/SAN architectures.
7. Beyond HA: Disaster Recovery, Security, Monitoring
It goes further than pure availability, tackling real-world production needs:
- Disaster Recovery Planning (DRP): backup strategies, failover protocols, externalization, testing
- Security: least privilege principle in containers, Kubernetes network policies, RBAC, firewalls, update cycles
- Monitoring: metrics (Nagios, Centreon, SNMP), tracing, autoscaling using Kubernetes metrics APIs
- Cloud-native storage: Ceph, MinIO, GlusterFS, Rook, OpenEBS
📦 Structure of the Book
- Infrastructure & essential services
- Containers (Docker)
- Load balancing & service exposure
- Orchestration
- Kubernetes deployments
- HA Storage systems
- Kubernetes clustering & final integration
- Advanced topics: DRP, security, monitoring, cloud storage
✅ What Makes It Stand Out
- Step-by-step approach, illustrated with a running demo app (Java + Tomcat)
- Comprehensive coverage: from physical servers to automated orchestration
- Hands-on examples and code available via GitHub
- Focus on real-world reliability: including backup, disaster recovery, and proactive monitoring
👥 Who Should Read This
- System administrators and DevOps engineers looking to design resilient Linux systems
- Infrastructure engineers working with containers, Kubernetes, load balancers, or storage
- Tech leads or sysadmins who want to implement high availability, disaster recovery, or continuous monitoring practices
🧾 Final Thoughts
This book is a comprehensive toolbox for building highly available Linux infrastructures, from bare metal to fully orchestrated Kubernetes clusters. Structured, practical, and deeply rooted in DevOps principles, it equips engineers with everything they need to build fault-tolerant, automated, secure, and monitored systems that stand the test of time.