NetApp, Inc.
Optimistic and failsafe network configuration

Abstract:

An optimistic and failsafe technique validates network configurations of storage and compute nodes deployed as a cluster. An optimistic aspect of the technique, saves an initial network configuration state of each node as a "failsafe" state and an expected network end-state is applied to each node. According to a validation aspect of the technique, each node employs a test to validate connectivity with other nodes in the cluster. In response to every validating node responding to a coordinating node that the validation test succeeded, an "all-clear" message is sent to all of the nodes instructing each node to maintain the applied expected network end-state. If any node is unreachable due to a configuration validation failure, then a failsafe aspect of the technique is invoked wherein the all-clear message is not sent and the remaining nodes of the cluster automatically "roll-back" to the initial failsafe network state after a timeout.