PowerFlex Software

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.

PowerFlex’s Fine Granularity Storage Pools use Persistent Memory (PMEM) also known as Non-Volatile DIMM (NVDIMM) to accelerate and compress data. These DIMMs may have bad blocks just like regular RDIMMs and must be monitored periodically for faults like any other type of disk. In an earlier release, the SDS device will only detect the bad block upon accessing the relevant section and failing the read by failing the SDS. For the earlier release, the following steps can be followed proactively to detect NVDIMMs with bad blocks and replace them. Note: since we have inherent redundancy in the product the data of these blocks are backed up on another node and in case of failure the SDS will fail and the other SDSs will service the IO.

PowerFlex Manager (PFxM) 3.7.0 has introduced file system sizing changes on the appliance. For all users to get these changes, we recommend using the backup and restore method to upgrade instead of the standard PowerFlex Manager upgrade process. This article provides the steps to upgrade PowerFlex Manager using the backup and restore method.

Continual Meta Data Manager (MDM) switchover causes cluster instability.

Deploying PowerFlex may fails when using SSH authentication on the PowerFlex Gateway due to the method by which the ssh keys are created.

This article provides the procedure for redeploying a PowerFlex Gateway (GW).

When a process or kernel module initiates communication over the network, a local port is allocated from the ephemeral port range as part of the socket.

iDRAC logs RED097 because Non‑Volatile Memory express (NVMe) firmware isn’t in inventory, aborting Storage‑Only upgrade; cause is system selecting wrong NVMe package (5CM3G) instead of correct 5V3P7.

This Article is an alert reference guide for the PowerFlex Block subsystem SIO03.XX.XXXXXXX

Using PowerFlex Management Portal (PFMP), capture a group’s network mappings, service details, credentials and node IPs, remove the group (non‑delete), then re‑add it with the recorded information.

14‑gen Ready Nodes with ≥12 DIMMs can reboot, recorded in iDRAC as a BP1 power‑good voltage out‑of‑range error; signal noise on the backplane power cable creates loss alerts, risking data loss.

PowerFlex Security Configuration Guide documentation may not accurately list all requires ports and protocols for PowerFlex Manager functionality. An updated table has been provided in this article to highlight all requirements.

This guide catalogs PowerFlex Block alerts for subsystem PFX000XX.XXXXXXX, detailing each alert code, description, severity level and potential system impact without providing remediation steps.

This article shows how to use PuTTY to enable session logging and perform the appropriate show tech‑support commands on Cisco Nexus or PowerSwitch switches, capturing logs and CPU data for analysis.

Rebalance in Fine Granularity pools may be erratic after an SDS disconnects, rebuilds and reconnects; the SDS must clear its devices before data can be rebalanced, and the time depends on data volume.

The "Software Serial ID" or "SWID" (usually starts wit ELMSIO....) is used to open Service Request for PowerFlex, and to start software renewal process. If the Software Serial ID is not available, you can also create cases using the "System Serial number" (usually starts with "ELMVX....)

The following procedure explains how to manually install an external SDC on a Debian-based server. The steps in the procedure below are relevant for Debian/Ubuntu operating systems, including Proxmox Virtual Environment.

PowerFlex Manager back up to external NetApp NFS share will not work.

During the execution of the procedure described in "Deploy VxFlex OS v3.x" for "Prepare the DAX devices" section 9 - "Convert the namespace device to the acceleration device name of type /dev/daxX.X:", it reports "invalid argument"

The user is unable to prepare an NVDIMM as a DAX device and add it to the PowerFlex acceleration pool.

During the PowerFlex upgrade process, a Slave (or Slaves in a 5_node configuration) MDM process might keep crashing due to low free space.

The Presentation Server alerts show the secondary Main Data Management (MDM) certificates are not approved.

Use this procedure to verify and upgrade the CPLD firmware version on PowerFlex nodes. PowerFlex Manager does not verify or upgrade CPLD.

PowerFlex Manager 4.x may show false alerts—CONFIGURED_SPARE_CAPACITY_SMALLER_THAN_LARGEST_FAULT_UNIT, STORAGE_POOL_UNBALANCED, NOT_ENOUGH_FAULT_UNITS_IN_SP due to a known software issue pending fix.

OS conversion for a Resource Group (RG) completes and causes DU when the system has SDS IP roles that are not configured as expected.

This article explains how to understand the purple screen (PS) root cause.

The PowerFlex Management Platform (PFMP) runs an SVM OS conversion (CentOS to SLES) and fails to complete.

This article explains step by step how To Convert Untagged DATA Networks On Access Mode To VLAN tagged Interfaces After Upgrade And OS Conversion.

In PFMP 4.8, when using the UI to perform a "Replace Node OS" operation, it fails due to missing MDM packages.

OS conversion for a Resource Group (RG) completes and causes DU when the system has fewer Standby Managers or Tie-Breaker MDMs than expected.