Score:0

Seeking advice regarding ingress versus load balancing for app in microk8s cluster

at flag

I have a seven node microk8s cluster set up on an Ubuntu server in my cellar. multipass shows me I have three IP address per VM since I created these hosts using the bridged network option. I thought having a routable IP might make it easier to expose the VMs and their services to my local network but probably I just made my platform more complex :P

I have succeeded in deploying https://ubuntu.com/tutorials/install-a-local-kubernetes-with-microk8s "microbot" web app but have not been able to access it using a browser even at the local console. I can connect to the app using curl 10.152.183.40 if I'm ssh'd into a control plane VM.

ubuntu@m8s7:~$ microk8s kubectl get all --all-namespaces |grep microbot-service
default        service/microbot-service       NodePort       10.152.183.40    <none>        80:31031/TCP                 34m

I'd like to make this (or any web app) run on a 192.168.10.0/24 addresses or at a node's hostname:port .

Name                    State             IPv4             Image
mk8s0                   Running           10.74.44.144     Ubuntu 21.10
                                          192.168.10.14
                                          10.1.123.64
mk8s1                   Running           10.74.44.172     Ubuntu 21.10
                                          192.168.10.13
                                          10.1.238.128

This page https://www.ibm.com/docs/en/fcifi/3.0.1?topic=administering-changing-kubernetes-master-node-ip-address suggests I might be able to make the entire cluster available behind a single IP address but there is no instance of externalIPs: when I run kubectl -n kube-system edit svc .

Should I be implementing ingress or a load balancer or can I tell k8s to run my app natively on a 192.168.10.0/24 address?

Thanks for any advice.

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.