Can I use ServiceLB outside of k3s?

0
10
Asked By TechyTurtle99 On

I'm curious if anyone has managed to run ServiceLB, which is the built-in load balancer for k3s, on something like k0s instead. It's a bit tricky since I couldn't find any clear guidance on making it work independently. Has anyone had any experience with this or know if it's even feasible?

5 Answers

Answered By HomeLabHero77 On

I've personally switched from ServiceLB to Kube-vip for my K3s clusters, and it's been great! I don't have to use NodePort since I have full IPs. I also rely on Traefik for my ingress, which works well for me.

Answered By LoadBalancerLover24 On

Have you considered alternatives like HAProxy or Traefik? They can be deployed as NodePort services, but that means you'd use higher port numbers since you won't have port 80 open. You might also opt for a HostPort deployment, but that limits you to one replica per node. With ServiceLB, you can have multiple replicas on a single node for better rolling updates. Just wondering if you've thought about these options for your ingress needs!

Answered By KubernetesKeen17 On

Just a heads up, I ran into challenges with Kube-vip and MetalLB using ARP or BGP protocols that are often blocked by cloud providers. If you're stuck on cloud VMs, that might complicate things for you!

Answered By FeatureFinder34 On

I've given up on ServiceLB due to limitations and switched to hostports instead. Sure, it’s not ideal, but if you want to dive deeper into ServiceLB alternatives, check out these GitHub links: One is for a standalone HAProxy load balancer, and another might offer a similar function. It's worth exploring!

Answered By CloudyCoder88 On

Actually, ServiceLB can run standalone like the RKE2 cloud controller manager. You would just need to disable the node features and focus on the load balancer controller. There's a GitHub repo with more info on it. Have you checked it out?

Related Questions

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.