k8s high availability configuration ege cases

Question

we have a pod in production which need to be highly available (100%),so we did the following:

1. We configure 3 instance as HA but then the node died
2. We c