Operational Defect Database
BugZero found this defect 418 days ago.
Data sources
All data on this page is proprietary to BugZero® or gathered from public sources
Hewlett Packard Enterprise | a00131053en_us
Advisory: HPE MR Gen10 Plus Controllers - Gen10 Plus Servers May Experience Resets on NVMe Drives if the Controllers are Configured for vSphere VMDirectPath I/O
Last update date:
3/11/2024
Affected products:
HPE MR Gen10 Plus Controllers
HPE ProLiant DL360 Gen10 Plus server
HPE ProLiant DX380 Gen10 Plus server
Affected releases:
No affected releases provided.
Fixed releases:
No fixed releases provided.
Description:
Info
HPE ProLiant DX/DL360 and DX/DL380 Gen10 Plus servers with an HPE MR416i-a Gen10 Plus x16 Lanes 4GB Cache NVMe/SAS 12G Controller or HPE MR216i-a Gen10 Plus x16 Lanes without Cache NVMe/SAS 12G Controller configured with NVMe drives attached and set up for vSphere VMDirectPath I/O may encounter drive resets in the operating system of the virtual machine where the MR Gen10 Plus controller is configured in the pass-through mode. Example: Linux Virtual Machine with MR216a and attached NVMe drives in vSphere VMDirectPath I/O mode under heavy load A log similar to the following may be seen. dmesg log: sd 3:2:1:0: [sdq] tag#6633 BRCM Debug mfi stat 0x2d, data len requested/completed 0x40000/0x0 sd 3:2:1:0: Power-on or device reset occurred sd 3:2:1:0: [sdq] tag#6749 BRCM Debug mfi stat 0x2d, data len requested/completed 0x3000/0x0 sd 3:2:1:0: [sdq] tag#5797 BRCM Debug mfi stat 0x2d, data len requested/completed 0x40000/0x0 sd 3:2:1:0: Power-on or device reset occurred sd 3:2:1:0: [sdq] tag#5798 BRCM Debug mfi stat 0x2d, data len requested/completed 0x3000/0x0 sd 3:2:1:0: [sdq] tag#6393 BRCM Debug mfi stat 0x2d, data len requested/completed 0x40000/0x0 sd 3:2:1:0: Power-on or device reset occurred sd 3:2:1:0: [sdq] tag#6394 BRCM Debug mfi stat 0x2d, data len requested/completed 0x3000/0x0 sd 3:2:1:0: [sdq] tag#6397 BRCM Debug mfi stat 0x2d, data len requested/completed 0x3000/0x0 sd 3:2:1:0: Power-on or device reset occurred sd 3:2:1:0: Power-on or device reset occurred sd 3:2:1:0: [sdq] tag#6401 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE cmd_age=8s sd 3:2:1:0: [sdq] tag#6401 Sense Key : Aborted Command [current] sd 3:2:1:0: [sdq] tag#6401 Add. Sense: No additional sense information sd 3:2:1:0: [sdq] tag#6401 CDB: Write(10) 2a 00 6f c5 ab 20 00 00 18 00 blk_update_request: I/O error, dev sdq, sector 1875225376 Aborting journal on device sdq1-8. sd 3:2:1:0: Power-on or device reset occurred EXT4-fs error (device sdq1) in ext4_init_inode_table:1306: IO failure sd 3:2:1:0: [sdq] tag#6454 BRCM Debug mfi stat 0x2d, data len requested/completed 0x1000/0x0
Scope
HPE MR416i-a Gen10 Plus x16 Lanes 4GB Cache NVMe/SAS 12G Controller HPE MR216i-a Gen10 Plus x16 Lanes without Cache NVMe/SAS 12G Controller HPE ProLiant DX/DL360 Gen10 Plus server HPE ProLiant DX/DL380 Gen10 Plus server NVMe Supported Drive
Resolution
To avoid NVMe drive resets when configured in the specific configuration in the Scope, follow the steps below to modify the virtual machine configuration file. Power off and right-click the Virtual Machine. Edit settings for the Virtual Machine. Select VM options/Advanced settings, and add/modify the setting pciHOLE.start = 2048 to the VM configuration file. NOTES : The resolution is specific to the configuration involving controllers listed in the Scope section of this advisory. This issue is with Intel servers only and not with AMD servers. RECEIVE PROACTIVE UPDATES : Receive support alerts (such as Customer Advisories), as well as updates on drivers, software, firmware, and customer replaceable components, proactively in your e-mail through HPE Support Alerts. Sign up for Support Alerts at the following URL: HPE Email Preference Center NAVIGATION TIP : For hints on navigating HPE.com to locate the latest drivers, patches and other support software downloads, refer to the Navigation Tips document. SEARCH TIP : For hints on locating similar documents on HPE.com, refer to the Search Tips document.
