Home > Storage > PowerScale (Isilon) > Product Documentation > Storage (general) > PowerScale OneFS Technical Overview > Software upgrade
Upgrading to the latest version of OneFS allows you to take advantage of any new features, fixes, and functionality. Clusters can be upgraded using two methods: Simultaneous or rolling upgrade.
A simultaneous upgrade installs the new operating system and reboots all nodes in the cluster simultaneously. A simultaneous upgrade requires a temporary, sub-two-minute, interruption of service during the upgrade process while the nodes are restarted.
A rolling upgrade individually upgrades and restarts each node in the cluster sequentially. During a rolling upgrade, the cluster remains online and continues serving data to clients with no interruption in service. Prior to OneFS 8.0, a rolling upgrade can only be performed within a OneFS code version family and not between OneFS major code version revisions. From OneFS 8.0 onwards, every new release will be rolling-upgradable from the prior version.
Non-disruptive upgrades (NDUs) allow a cluster administrator to upgrade the storage operating system while their end users continue to access data without error or interruption. Updating the operating system on a cluster is a simple matter of a rolling upgrade. During this process, one node at a time is upgraded to the new code, and the active NFS and SMB3 clients attached to it are automatically migrated to other nodes in the cluster. Partial upgrade is also permitted, whereby a subset of cluster nodes can be upgraded. The subset of nodes may also be grown during the upgrade. An upgrade can be paused and resumed allowing customers to span upgrades over multiple smaller Maintenance Windows. Also, OneFS 8.2.2 and later offer parallel upgrades, whereby clusters can upgrade an entire neighborhood, or fault domain, at a time, substantially reducing the duration of large cluster upgrades. OneFS 9.2 and later combine OS and firmware upgrades, substantially reducing the impact and duration of upgrades by allowing them to occur in tandem. 9.2 and later also include drain-based upgrades, whereby nodes are prevented from rebooting or restarting protocol services until all SMB clients have disconnected from the node.
OneFS supports upgrade rollback, providing the ability to return a cluster with an uncommitted upgrade to its previous version of OneFS.
OneFS powered clusters support automatic drive firmware updates for new and replacement drives, as part of the non-disruptive firmware update process. Firmware updates are delivered using drive support packages, which both simplify and streamline the management of existing and new drives across the cluster. This ensures that drive firmware is up to date and mitigates the likelihood of failures due to known drive issues. As such, automatic drive firmware updates are an important component of OneFS’ high availability and non-disruptive operations strategy. Drive and node firmware can be applied as either a rolling upgrade or through a full cluster reboot.
Prior to OneFS 8.2, node firmware updates had to be installed one node at a time, which was a time-consuming operation especially in large clusters. Node firmware updates can now be choreographed across a cluster by providing a list of nodes to be simultaneously updated. The upgrade helper tool can be used to select a wanted combination of nodes that can be updated simultaneously and an explicit list of nodes that should not be updated together (for example, nodes in a node-pair).
As part of an upgrade, OneFS automatically runs a pre-install verification check. This verifies that the configuration in your current installation of OneFS is compatible with the version of OneFS that is intended for the upgrade. When an unsupported configuration is found, the upgrade is stopped and instructions on troubleshooting the issue are displayed. Proactively running the pre-installation upgrade check before starting an upgrade helps to avoid any interruption due to incompatible configuration.