PowerScale OneFS

On December 5, 2025, there will be a refresh to the Google certificate authority that we use within OneFS. This refresh previously occurred on December 15, 2021.

This document highlights minimum, recommended, and latest code versions for Dell Technologies Enterprise and Datacenter Infrastructure products. We carefully evaluate and monitor code deployment data to ensure that product releases meet our stringent expectations to provide our customers with reliable products and solutions.

In OneFS 8.0 and later, PowerScale clusters are being discovered as Fraunhofer FOKUS instead of Dell.

Users may notice that if they are part of more than 16 groups they may be denied access if permissions are granted based on a group. This is by design in the NFS protocol and can be worked around with the Map Lookup UID option.

When attempting to install a patch or the Drive Support Package an error may be seen of "failed to stage patch". This knowledge base article covers how to resolve this.

Many OneFS versions have obsolete timeout values configured for the drive stall timer. Increasing these values can help prevent unnecessary drive stall events.

On OneFS 8.0, the firmware update process fails if the BMC/CMC subsystem is unresponsive. Sometimes the warning event to indicate this condition is not generated.

After upgrading to 8.0.1.0, Mac clients are unable to view secondary mounted shares (but can still view first mounted share/volume) and receive a denial of access.

The OneFS event 'Physical Memory low (expected xxxGB, found yyyGB)' indicates a defective DIMM module in the node, or a problem with the node's configuration.

Isilon GEN 6: HW_INFINITY_COMPUTE_FAULT: Compute node fault detected (User Correctable) - Event ID: 900160003

Upon upgrading OneFS, the isi_avscan_d daemon is consuming CPU, affecting node performance.

PowerScale Info Hub for OneFS 9.14.0.0 documentation

Gen6 node fails to boot during the PSI configuration phase.

Configuring automated log-upload on a weekly basis

PowerScale Isilon OneFS Event notification: Cluster-1: The cluster is running low on free space: free space (1.99xxx%) < job-configuration core.free_blocks_pct_threshold_lo (2%). - Event ID: 400100007

This article will help to remove the old entries of a child domain that is already decommissioned on the AD/DNS server.

PowerScale OneFS Event notification: One of the disk pools on your cluster is nearing, or has reached, maximum capacity.- Event ID: 100010015

How to properly delete snapshots from a cluster.

Case-sensitive HDFS platforms such as Cloudera or operating systems such as Linux may experience job failures due to changes in how OneFS reports resolved account names. After upgrading OneFS to 8.2.2.0, OneFS reports a case-preserved response from AD based on the SAMAccountName field. Linux or Cloudera clients reference a separate value from Active Directory that is stored in lowercase. The discrepancy can result in job failures submitted by impacted users. Clusters on OneFS 8.1.2.0 and lower do not display this behavior.

Below are steps to properly configure cascaded replication between Cluster_A, Cluster_B, and Cluster_C.

This article summarizes the commands to modify the user and/or group ownership in OneFS 8+

This KB affects Onefs8.2.2, Onefs 9.0 and Onefs 9.1 When running Smartpools Job, the progress only shows "Started". Status does not change from start to finish.

Documentation on OneFS configuration recommends using the OneFS SmartConnect DNS service to provide Round Robin DNS result via delegation, however in this configuration, if multiple DNS requests are received within 1 second of each other by the client DNS server, they will all receive the same IP address, causing connections to be unbalanced.

Using iperf to test bandwidth from a client to a OneFS cluster.

Nodes may panic with FAILED ASSERTION when fail to decode quota or snapshot attributes from extension blocks after upgrading to OneFS 9.2.0.0 or 9.2.1.0.

This article explains the issue that link status change for igb interface cannot be correctly reflected(found on OneFS 8.2, 9.0, 9.1, 9.2)

Dell has a Customer Replaceable Unit (CRU) program for many of our hardware systems. The CRU program allows customers to replace designated hardware components. For a list of the hardware components that are designated as CRU for a specific hardware system, go to the Dell Warranty and Maintenance Page.

This article explains why the Backup Restartable Extension (BRE) is incompatible with the MultiStreaming feature in OneFS.

SyncIQ jobs can intermittently fail if there are non-networked nodes (NANON).

While configuring Cloudpool Account for the first time or modifying an existing one to use SSL, error is reported.