Bob Explores High Availability and Clustering on AlmaLinux

Bob’s next adventure was to implement high availability (HA) and clustering to ensure his services stayed online even during hardware failures or peak loads. He learned to use tools like Pacemaker, Corosync, and HAProxy to build resilient and scalable systems.

“Downtime isn’t an option—let’s make my server unshakable!” Bob declared, diving into HA and clustering.

Chapter Outline: “Bob Explores High Availability and Clustering”

1. Introduction: What Is High Availability?

   • Understanding HA concepts.
   • Key tools: Pacemaker, Corosync, and HAProxy.

2. Setting Up a High-Availability Cluster

   • Installing and configuring Pacemaker and Corosync.
   • Creating and managing a cluster.

3. Implementing Load Balancing with HAProxy

   • Installing and configuring HAProxy.
   • Balancing traffic between multiple backend servers.

4. Testing Failover and Recovery

   • Simulating failures.
   • Monitoring cluster health.

5. Optimizing the HA Setup

   • Fine-tuning resources and fencing.
   • Automating with cluster scripts.

6. Conclusion: Bob Reflects on High Availability Mastery

Part 1: Introduction: What Is High Availability?

Bob learned that high availability ensures continuous access to services, even in the face of hardware or software failures. Clustering combines multiple servers to act as a single system, providing redundancy and scalability.

Key HA Concepts

• Failover: Automatically shifting workloads to healthy nodes during a failure.
• Load Balancing: Distributing traffic across multiple servers to avoid overloading.
• Fencing: Isolating failed nodes to prevent data corruption.

Tools for HA on AlmaLinux

1. Pacemaker: Resource management and failover.
2. Corosync: Cluster communication.
3. HAProxy: Load balancing traffic.

“With HA, my services will always stay online!” Bob said.

Part 2: Setting Up a High-Availability Cluster

Step 1: Installing Pacemaker and Corosync

Bob installed the necessary packages on two nodes (node1 and node2).

• Install HA tools:

  sudo dnf install -y pacemaker corosync pcs
  

• Enable and start the pcsd service:

  sudo systemctl enable pcsd --now
  

• Set a cluster password:

  sudo pcs cluster auth node1 node2
  

Step 2: Configuring the Cluster

• Create the cluster:

  sudo pcs cluster setup --name mycluster node1 node2
  

• Start the cluster:

  sudo pcs cluster start --all
  

• Enable the cluster at boot:

  sudo pcs cluster enable --all
  

• Check the cluster status:

  sudo pcs status
  

“The cluster is live—time to add resources!” Bob said.

Step 3: Adding Resources to the Cluster

Bob added a virtual IP as the primary resource:

• Add a virtual IP resource:

  sudo pcs resource create VirtualIP ocf:heartbeat:IPaddr2 ip=192.168.1.100 cidr_netmask=24
  

• Verify the resource:

  sudo pcs resource show
  

Part 3: Implementing Load Balancing with HAProxy

Step 1: Installing HAProxy

• Install HAProxy:

  sudo dnf install -y haproxy
  

• Enable and start HAProxy:

  sudo systemctl enable haproxy --now
  

Step 2: Configuring HAProxy

Bob configured HAProxy to balance traffic between two web servers.

• Edit the HAProxy configuration file:

  sudo nano /etc/haproxy/haproxy.cfg
  

• Add a load balancing configuration:

  frontend http_front
      bind *:80
      default_backend http_back
  
  backend http_back
      balance roundrobin
      server web1 192.168.1.11:80 check
      server web2 192.168.1.12:80 check
  

• Restart HAProxy:

  sudo systemctl restart haproxy
  

• Verify HAProxy is balancing traffic:

  curl http://<load-balancer-ip>
  

“HAProxy is routing traffic seamlessly!” Bob said.

Part 4: Testing Failover and Recovery

Step 1: Simulating Node Failures

Bob tested failover by stopping services on node1:

sudo pcs cluster stop node1

• Check if the virtual IP moved to node2:

  ping 192.168.1.100
  

Step 2: Monitoring Cluster Health

Bob used the following commands to monitor cluster status:

• View detailed cluster information:

  sudo pcs status
  

• Check resource logs:

  sudo pcs resource debug-start VirtualIP
  

“The cluster handled the failure like a champ!” Bob said.

Part 5: Optimizing the HA Setup

Step 1: Configuring Fencing

Bob configured fencing to isolate failed nodes.

• Enable fencing:

  sudo pcs stonith create fence_ipmilan fence_ipmilan pcmk_host_list="node1 node2" ipaddr="192.168.1.50" login="admin" passwd="password"
  

• Test the fencing configuration:

  sudo pcs stonith fence node1
  

Step 2: Automating with Cluster Scripts

Bob automated resource recovery using custom scripts:

• Add a script as a resource:

  sudo pcs resource create MyScript ocf:heartbeat:Anything params binfile=/path/to/script.sh
  

“With fencing and automation, my cluster is truly resilient!” Bob noted.

Conclusion: Bob Reflects on High Availability Mastery

Bob successfully built a highly available cluster with Pacemaker, Corosync, and HAProxy. By testing failover and optimizing his setup, he ensured his services could withstand hardware failures and peak loads.

Next, Bob plans to explore Linux Virtualization with KVM on AlmaLinux.