09/28/2022, 9:46 AM
Hello, I’ve having some strange issues with my setup: • 3 microk8s nodes in ha • rancher-master deployed from helm on microk8s cluster • ingress-nginx for ingress with 3 pods • one vsphere cluster with vcenter • added vsphere cloud provider in rancher ui • created node template to deploy k8s in vsphere from rancher • The strange thing that i’m seeing is that when deploying a new k8s cluster, rancher keeps creating multiple vms on vcenter, even when deploying single node k8s it tries to bring up 2-4 nodes, then deletes 2-3 of then, keeping only one, eventually creating the one node k8s, but if i try to scale up(+1) it again tries to bring up 2-6 vms, and repeat, without finishing the scale up process • microk8s version: 1.24(ubuntu) • rancher-master helm chart version: 2.6.8(tried also 2.6.7) same stuff happens Did anyone happened to come across this strange issue when deploying to vsphere ? Is there something that i’m missing in the setup: helm install command: helm install rancher-master rancher-stable/rancher --set addLocal=“true” --set=ingress.enabled=“false” --namespace cattle-system --set --set replicas=3 --set tls=“external” --version 2.6.8