Loading...
Loading...
With HPE Compute Scale-up Server 3200 firmware, immediately after an AC power-cycle, the Rack Management Controller (RMC) will always show the total memory size of the nPar and of each chassis as zero. When the nPar successfully boots once after an AC power-cycle, the RMC then shows the total configured memory size of the nPar(s) and all of the nPar's chassis. This is the expected behavior.However, an issue occurs where the RMC of a multi-chassis HPE Compute Scale-up Server 3200 running firmware 1.50.120 will incorrectly show zero (0) total memory size for all non-Monarch chassis after an AC power-cycle and after the nPar has successfully booted once. Since the memory size of the non-Monarch chassis is zero, the RMC shows the nPar memory size to be the same as the base chassis memory size. If the nPar has multiple chassis, the nPar total memory size shown by the RMC is incorrect because the non-Monarch chassis memory size is not included in the calculation.This is a reporting/cosmetic issue and does not imply that a non-Monarch chassis has no memory, or that the nPartition has reduced memory. All installed and configured memory is available and can be confirmed when booted to UEFI or to the OS. Use the UEFI shell hinv command or use individual OS commands to query the total amount of available physical memory.The following CLI commands are affected:show nParandshow chassisExample of a 16S nPar:Notice that all installed DIMMs are shown under OK/In in all the chassis but the Memory (GiB) RAM in the non-Monarch chassis is 0. This does not affect the Monarch chassis of an nPartition.Example of a Multi-partition displaying the problem:Notice that all installed DIMMs are shown under OK/In in all the chassis but the Memory (GiB) RAM in the non-Monarch chassis is 0. This does not affect the Monarch chassis of an nPartition.Affected Redfish property:The TotalSystemMemoryGiB property of MemorySummary in the Redfish URIs redfish/v1/Systems/Partition# and /redfish/v1/CompositionService/ResourceBlocks/<CHASSISGEOID>/Systems/1 where # is then Par number and <CHASSIS GEOID> is the id of the non-Monarch chassis.Example with Redfish:The problem is also seen if using the GUI to access and manage the HPE Compute Scale-up Server 3200. The amount of total configured memory displayed is not correct. To reiterate, this does not mean that the memory is not available.In the example below there are a total of 8 x 128GB DIMMs installed, 4 in the base chassis and 4 in the non-Monarch chassis for a total of 1024GB.Due to this reporting issue, only 512GB total of volatile memory is displayed.Select the Hardware tab and check the status of the memory in the non-Monarch chassis; it can be noted there that the 4 DIMMs are seen and healthy:
All multi-chassis HPE Compute Scale-up Server 3200 systems with firmware 1.50.120 installed and an AC power-cycle has been performed on the complex or an individual non-Monarch chassis has been AC power-cycled due to service.
This cosmetic issue has been addressed in firmware version 1.55.98:https://support.hpe.com/connect/s/softwaredetails?language=en_US&collectionId=MTX-9ad292ca931d4944&tab=releaseNotesThis is a cosmetic issue.There is NO action required.Confirm that ALL installed DIMMs are OK/In using the show npar or show chassis commands or via the GUI. Confirm the amount of available volatile memory using the UEFI hinv command or individual OS commands to query the total amount of available physical memory matches the confirmed OK/In.The base chassis refers to the lowest UPOS chassis in the complex and its Product Serial Number is used for warranty and support purposes. The monarch chassis refers to the chassis of an nPar that has PCH enabled. As a result, there can be multiple nPars each with its own monarch chassis and non-monarch chassis, but there is only one base chassis. This cosmetic issue only affects non-monarch chassis.Document VersionRelease DateDetails2April 10, 2025Updated to indicate the permanent fix, System ROM v 1.55.981January 14, 2025Original document release
Operating Systems Affected:Not Applicable
Click on a version to see all relevant bugs
Hewlett Packard Enterprise Integration
Learn more about where this data comes from
Bug Scrub Advisor
Streamline upgrades with automated vendor bug scrubs
BugZero Enterprise
Wish you caught this bug sooner? Get proactive today.