Hey everyone! I'm diving into Kubernetes with the aim of managing high availability clusters on-premise. I'd love to get some clarity on a couple of things. I understand that ideally, I should have three data centers close to each other to minimize latency, with each center running a master node and some worker nodes. First, how do these data centers communicate with each other? Is it through a VPN? Second, I've heard I need a load balancer (like MetalLB) that operates across all nodes. Can I use Cloudflare's load balancer to manage traffic to each of the three data centers? Apologies if my explanation is a bit muddled; I'm trying to figure out how to configure high availability implementations on-premise. Thanks for any help!
1 Answer
You definitely need to think about the reliability requirements for your setup. As for communication between data centers, it's typically managed by dedicated connections or VPNs based on your specific configuration. Just remember, mixing up load balancing types can create confusion; you need to clarify which type serves your needs better.
I get that! I just want to find the best way to achieve high availability. For the service load balancer, I’d be using MetalLB, but currently, I have Nginx running to expose the services. My router ports are pointed to my Nginx machine, and I have Cloudflare directing requests to my IP.