##Preface
This document describes the actions for catastrophe recovery. It is still work in progress.
To restore cluster from backups, you will need the latest etcd-data and etc backup tarballs, located on the bastion host. If you have lost bastion too, restore the tarballs to bastion first from offsite backup.
Launch deployment container and run
cd poc/playbooks
restore/full_restore.bash
This will provision a fresh cluster and restore etcd data (openshift objects) and generated certificates on it. The provisioning will be done in pieces, take a look at restore/full_restore.bash for details, comments and troubleshooting.
Note: The current version of the script works with openshift-ansible#release-3.7. For release-3.6, use
restore/full_restore_36.bash