Reading over the documentation for Rancher server disaster recovery it seems like there is not a good method for running Rancher (server) in a multi-site setup. Is the only option to have a standby cluster that backups get sync'd to ( or a remove store ) and then recovered and Rancher reinstalled via Helm on the secondary site ? Is anyone else running Rancher in a multi site setup at all ?

There's a 1:1 mapping of downstream cluster to Rancher instance. Currently, you'd use our backup operator to effectively restore Rancher to a new cluster. Downstream clusters will be unaffected if Rancher goes down.