We need to test different scenarios for how disaster recovery would work for the newly Kubernetes Cluster.
Possible scenarios that we should check:
- Lose of Connectivity
- Lose of Node (Potential kill kubelet service only to spare Elastic if they reside on the server)
- Restoration of backup
- DDOS simulation ?
- Rolling back faulty release
We need to test different scenarios for how disaster recovery would work for the newly Kubernetes Cluster.
Possible scenarios that we should check: