Closed wallrj closed 4 years ago
@JamesLaverack Please review: I've deviated from the backup API in the design document because it seems neater to be able to choose a destination bucket / blob URL for the backup rather than have be generated by the controller. This way I don't need to go looking at the status of the backup to find the URL. And this way backups don't need to be tied to an EtcdCluster generated cluster, they could equally be used for non-eco etcd clsuters. See what you think.
part of #146
Rewrite of the backup controller to launch a
backup-agent
which uploads backup files via aproxy
process, which has access to secrets needed to interact with the chosen bucket API.