backup script is not working for migrated etcd clusters

[bsctl]

Release v0.12.0 introduced a kamaji-etcd-client service for backup in addition to headless service kamaji-etcd. See #109

For migrated etcd clusters from previous versions < v0.12.0, the certificates for kamaji-etcd-client service are not created and the backup scripts fails. For fresh etcd cluster installed with v0.12.0, the certificated are correctly created.

As workaround, for < v0.12.0 or migrated to v0.12.0, use the old backup script found in tag v0.11.0.

git clone https://github.com/clastix/kamaji-etcd
cd ~/kamaji-etcd/scripts
git fetch --tags
git tag -l
git checkout tags/v0.11.0

# To run the script, use the following command:
# ./backup.sh [-e etcd_name] [-s etcd_service] [-n etcd_namespace] [-j schedule]
# Parameters
# -e etcd_name: Name of the etcd StatefulSet
# -s etcd_service: Name of the etcd service 
# -n etcd_namespace: Namespace of the etcd StatefulSet
# -j schedule: Cron schedule for the backup job (default: "0 0 * * *", which means daily at midnight)

./backup.sh -e customer01-etcd -s customer01-etcd -n customer01 -j "30 21 * * *"