TODO

TODO

- Test nginx loadbalancing
- Test PV,PVC
- Prometheus + Grafana
  - add nginx
  - add x2go

- https://www.cloudops.com/fr/blog/guide-de-survie-rook-et-ceph/

- kube-proxy : test mode ipvs
- upgrade to kubernetes 19.1
- network:
  - kube-router instead of manual routes
  - flannel
- runtimes : https://kubernetes.io/docs/setup/production-environment/container-runtimes/#container-runtimes
  - low-level: kata
- install addons : https://kubernetes.io/docs/concepts/overview/components/
  - autoscaler : https://github.com/kubernetes/autoscaler
  - ingress managers
    - traefik
    - ingress-nginx : https://github.com/kubernetes/ingress-nginx
    - envoy : https://www.envoyproxy.io/

FIX
- edit crio socket

ISSUES
- facing this when reboot 3 controllers -> https://github.com/etcd-io/etcd/issues/10655
- kubectl timeout... exiting shell or ping
- health check coredns ?
- health check with haproxy ? https://github.com/kelseyhightower/kubernetes-the-hard-way/blob/master/docs/08-bootstrapping-kubernetes-controllers.md#provision-a-network-load-balancer
- kubectl proxy print errors
  - E1025 18:11:15.522577 3195026 proxy_server.go:144] Error while proxying request: context canceled
- metrics : not working
- long time after deleting a pod