High Availability Linux Architecture (LFS422)

Overview

This High Availability Linux Architecture for Mission Critical Workloads training will provide you with the knowledge and skills needed to successfully engineer, provision, maintain, and administer high-availability clusters to support mission-critical workloads. By leveraging the unique talent and experience from some of the largest Linux deployments in the world, the Linux Foundation built this course as a reference for IT architects and managers looking to engineer, provision, design, deploy and maintain highly available enterprise IT environments. 
 
Heavily focused on mission critical workloads, High Availability Linux Architecture focuses on concepts and best practices used to meet security and high availability requirements using the latest hardware technologies, driven by current open source tools found in the three main Linux distribution families. We will cover why Linux on industry standard hardware is a viable mission-critical platform, what new features have been implemented and tested, and offer updates on vendor support and services.

Audience
 
IT managers, system administrators, DevOps, and technical personnel operating on the Linux platform, as well as Technical IT professionals interested in bridging the gap between system administration and business decision making. Likewise, business analysts and decision makers who require visibility into current best practices in Linux IT architecture will greatly benefit from the topics covered in this class, as well as those professionals considering migrating to Linux from UNIX environments. 
 
Prerequisites
 
Linux Enterprise Architecture for High Availability is designed for IT professionals who are already familiar with Linux and Unix. A working knowledge of fundamental operating system and hardware concepts is required in order to maximize your learning experience, and experience with network addressing and topologies is a definite requirement. A basic foundation in mission critical deployments will also be a benefit.

 

Show details
Course Outline

1. High Availability Basics
 
2. HA Basics: Mission Critical Systems
 
  • Mission Critical Systems
 
3. Intel RAS Features and their Relevance for HA
 
  • Intel RAS Platform
  • CPU Specific Features
 
4. HA Basics: From Unix to Enterprise Linux
 
  • Enterprise Linux Capabilities
  • State of Enterprise Linux and Support Opportunities
  • Comparisons with Conventional Unix Systems
 
5. HA Basics: Linux Distributions and Commercial Support
 
  • Existing Linux Distributions
  • Support Offerings
 
6. HA Basics: An Introduction to High Availability
 
  • What High Availability Is
  • Differences between HA and Fault Tolerance
 
7. Storage Options in High Availability Stacks
 
  • Storage Options
 
8. Advanced Open-Source Storage Solutions
 
  • Highly Distributed and Redundant Storage Solutions
  • Ceph
 
9. Linux Distributions and their HA Stacks
 
  • Anatomy of a HA Stack
  • Heartbeat
  • CMAN and rmanager
  • Corosync and Pacemaker
 
10. High Availability Stacks
 
11. Linux High Availability Stacks
 
  • Anatomy of a HA Stack
  • HA Options for Linux
 
12. Introduction to Corosync
 
  • Where does Corosync fit in the stack
  • Cluster Messaging
  • Totem
  • Creating a Corosync Cluster
 
13. Introduction to Pacemaker
 
  • Where does Pacemaker fit in the stack
  • Cluster Resource Management
  • Pacemaker Resources
  • Starting a Pacemaker Cluster
 
14. Node Fencing in Pacemaker Clusters
 
  • What is Fencing Good for
  • Quorum Considerations
  • 2-Node Clusters
  • 2-Node Cluster Best Practices
  • DRBD-based 2-Node Clusters
  • DRBD Resource Level Fencing
  • Fencing Devices
 
15. Highly Available MySQL With DRBD and Pacemaker
 
  • What are the Parts of the Stack
  • MySQL
  • DRBD
  • Corosync
  • Pacemaker
 
16. Advanced Corosync and Pacemaker Clusters
 
17. Resource Monitoring and Recover with Pacemaker
 
  • Migration Thresholds
  • Failure Timeouts
 
18. HA Virtualization with GlusterFS and Pacemaker
 
  • Key Components: KVM and libvirt
  • DRBD Dual-Primary Setup with GFS2
  • Corosync and CMAN
  • STONITH
  • DRBD Configuration Updates
  • Setup DRBD,DLM,GFS2,VM as Pacemake Resources
  • Live-Migrations
 
19. Site Failover with the Linux HA Cluster Stack
 
  • Booth
  • Inter-site Communication Problems
  • Arbitrators
  • Solving Consensus with Paxos
  • Tickets
  • Pacemaker Integration
  • Arbitrator Configuration
  • Ticket Management Commands
 
20. Integrating Advanced Storage Solutions and Pacemaker
 
  • GlusterFS and Ceph Primer
  • GlusterFS Resource Agents
  • glusterd and Volume Management
  • One-for-All Filesystem Resource Agent
  • Ceph OCF Resource Agent
  • RDB Client Integration
 
21. Cluster Maintenance and Upgrades
 
  • Maintenance Mode
  • Node Standby
  • Updates, Patches and Upgrades
 
22. Automating, Monitoring and Designing HA Clusters
 
23. Monitoring Linux High Availability Clusters
 
  • Corosync Process
  • Pacemaker Processes
  • ClusterMon
  • Nagios Integration
  • Zabbix Integration
 
24. Automating Linux HA Cluster Deployments
 
  • Puppet
  • Puppet by Example
  • Blah
  • Chef
  • Crowbar
  • Cluster vs System Automation
 
25. Designing Systems for High Availability
 
  • Five Nines
  • Uptime Requirements and SLAs
  • Single Points of Failure
  • Special Considerations
 
26. Ask the Trainer Anything
 
  • Questions and Answers