Kubespray Experiment

[haminhcong]

Plan

Setup KubeSpray on Bare Metal Environment:

• 3 Master
• 3 Worker
• 3 Etcd
• Storage: Rook/Ceph GlusterFS - not enough resources
• Load Balancer: Node Port/Metal LB ?

Targets

• [x] Using Kubespray setup K8S Cluster on HA topology scenario

• [ ] Integrate Kubespray with Other Playbooks like: pre-install, install post-install https://youtu.be/da8jDsf7BjM?t=991

• [ ] Replace Etcd/Master/Worker scenario

• [ ] Add Etcd/Master/Worker scenario

• [ ] Remove Etcd/Master/Worker scenario

• [ ] Add monitoring/logging

• [ ] Monitor Master/worker by external Prometheus

• [ ] Monitoring cluster Prometheus by external Prometheus

• [ ] OpenStack Cloud Teraform provisioning Cluster VMs and Cluster LB

• [ ] OpenStack Cloud External Cloud Provider: LB, CSI Cinder

• [ ] Add monitoring/logging to existing ansible playbook on kubespray - or create new playbook

• [ ] Add Metal LB

• [ ] Add Container Volume Storage solution: Ceph/GlusterFS ?

• [ ] Restore Cluster from Etcd Snaphot - Disaster Recovery ? https://github.com/kubernetes-sigs/kubespray/blob/master/docs/recover-control-plane.md

  References

• https://kubespray.io/#/

• https://computingforgeeks.com/deploy-kubernetes-cluster-centos-kubespray/

• https://github.com/kubernetes-sigs/kubespray/tree/master/contrib/terraform/openstack

• https://github.com/kubernetes-sigs/kubespray/tree/master/contrib/network-storage/glusterfs

• https://youtu.be/da8jDsf7BjM?t=991

• https://documentation.suse.com/sbp/all/pdf/SBP-rook-ceph-kubernetes_color_en.pdf

• https://github.com/patrickstjohn/running-rook-in-production

• https://www.benevolent.com/engineering-blog/deploying-metallb-in-production

• https://kubesail.com/blog/storage-on-kubernetes

• https://devopstales.github.io/cloud/k8s-metallb/