PowerScale OneFS Upgrade Planning and Process Guide

Parallel upgrades

Dell Technologies recommends using the parallel upgrade option when upgrading any size cluster running OneFS 8.2.2 and later. Parallel upgrades require a smaller maintenance window than rolling upgrades, and do not require the interruption of service like simultaneous upgrades.

A parallel upgrade installs the new operating system on a subset of nodes and restarts that subset of nodes at the same time. Each subset of nodes attempts to make a reservation for their turn to upgrade until all nodes are upgraded. Node subsets and reservations are based on diskpool and node availability.

During a parallel upgrade, node subsets that are not being upgraded remain online and can continue serving clients. However, clients that are connected to a restarting node are disconnected and reconnected. How the client connection behaves when a node is restarted depends on several factors including client type, client configuration (mount type, timeout settings), IP allocation method, and how the client connected to the cluster. In OneFS 9.2.0.0 and later, client connection behavior is managed by the disruption manager settings.