Service resiliency within on-premise products

Described is a technology by which a remote network (or local) service is monitored for failure, e.g., by determining whether data corresponding to operation of the service achieves a threshold. When a threshold is achieved that indicates that the service has failed, the service is automatically deactivated for use with a local workload, and the workload switched to use a second service or system having functionality similar to that of the failed service. The second service may be monitored for failure, as may the failed service to determine when it is again operational. In this manner, a remote primary service may have another local or remote secondary service/system automatically substituted for it when it fails, and be restored for use when it returns to acceptable operation.