💻 Proxmox VE Course IV-A-3. Cluster Join: Joining a Second Node or More to the Cluster
🤝 Two are Stronger Than One: Completing the Cluster
In the previous lecture, we laid the foundation of a massive infrastructure called a cluster. Now, it is time to raise the pillars upon it. The moment individual servers are bound into a single cluster, true resource sharing and non-disruptive service become possible. In this #lecture, we will dive deep into the practical #method of safely and quickly joining a second node, and more, to the cluster we created on the first node.
1. Prerequisites Before Joining a Node
Joining a cluster is not just about connecting servers; it is a sensitive operation that synchronizes configuration files and permissions. Inadequate preparation can lead to communication errors or tangled settings during the joining process.
A. Emptying Virtual Resources
The new node intended for joining must not have any existing virtual machines (VMs) or containers. If there are existing #resources in operation, you must back them up and delete them, then restore them after joining the cluster to prevent conflicts.
B. Checking Network Latency
Network latency between cluster nodes should be less than 2ms. Corosync communication is extremely sensitive; if the #network response speed is slow, the cluster may fail to maintain Quorum, potentially causing the disastrous consequence of frequent node reboots.
C. Matching Software Versions
It is highly recommended to keep the Proxmox VE versions identical across all nodes. Differences in kernel or #software package versions can hinder the #stability of the cluster due to functional compatibility issues.
2. How to Join via GUI Using Join Information
One of the greatest advantages of Proxmox is that it summarizes complex tokens and key exchange processes into a convenient block of text called 'Join Information.'
A. Copying Information from the First Node
On the node where the cluster is already created (Node A), go to 'Datacenter' > 'Cluster' and click the 'Join Information' button. Click the 'Copy Information' button to copy the complex string to your clipboard.
B. Pasting Information on the New Node
Access the web console of the new node (Node B), navigate to the same menu, and click the 'Join Cluster' button. Once you paste the copied information, the #IP address and fingerprint will be populated automatically.
C. Peer Authentication and Completion
Enter the root password of the first node and click 'Join' to start the process. The web interface may become unresponsive for a moment as SSL certificates are renewed and #infrastructure settings are synchronized—simply wait for it to finish.
3. Powerful Node Joining Techniques via CLI
Using the Command Line Interface (CLI) allows you to monitor the joining process more closely and is very advantageous when writing automation scripts.
A. The pvecm add Command
Run the command
pvecm add <First_Node_IP>in the terminal of the new node. This method outputs the #data transfer process in real-time, allowing for immediate debugging if a problem occurs.
B. The Importance of Fingerprint Verification
For security, a process to verify the SSH fingerprint of the server you are connecting to will proceed. This is an #optimized security procedure that prevents man-in-the-middle attacks and ensures you are communicating with the correct node.
C. Observing Quorum Status Changes
After joining is complete, typing
pvecm statuswill show an increase in 'Total votes.' You can observe the process of the cluster becoming more robust and the reliability of decision-making increasing as the number of nodes grows.
4. Essential Post-Join Checks and Troubleshooting
Joining the node is not the end. A verification process is needed to ensure everything is working organically.
A. Verifying Shared Storage and Network Bridges
Ensure that network bridges (such as vmbr0) with the same name are configured on all nodes. These settings must match for seamless movement without #performance degradation during Live Migration, which we will learn later.
B. Synchronizing Firewall Settings
Once joined to the cluster, datacenter-level firewall rules are immediately applied to the new node. Check to ensure that management ports aren't accidentally blocked to increase the efficiency of your #virtualization operations.
C. Strategies to Prevent Node Dropout
If a node frequently shows as 'Offline,' check the bandwidth of the dedicated Corosync network. Allocating a dedicated NIC for cluster communication is the most #optimized alternative from a long-term perspective.
Your Proxmox environment has now evolved beyond the limits of a single server into a true 'Cluster.' As the number of nodes has increased, management efficiency has improved, and your disaster recovery capabilities have skyrocketed. In the next lecture, we will explore how to build 'Shared Storage' and synchronize data using these bound nodes.
Lecture, Method, Resource, Network, Software, Stability, IP, Infrastructure, Data, Optimization, Performance, Virtualization
Optimal performance, best cost efficiency! Experience Proxmox VE-based hosting that perfectly fits your project.
댓글
댓글 쓰기