Intermittent Virtual RAID Degraded or Missing Local Disks on Cisco UCS after UCSM 4.3(6d) Upgrade

Products

• VMware vSphere ESXi

Issue/Introduction

Issue/Introduction

This article addresses intermittent storage connectivity issues on Cisco UCS B-Series blade servers following a firmware upgrade to Cisco UCS Manager (UCSM) 4.3(6d).

Symptoms:

• After a host reboot, local storage devices (specifically Virtual RAID M2) are intermittently not detected by the ESXi host. The host reports the Virtual RAID as "degraded" or displays storage devices as missing.
• Issue occurs only after firmware upgrade and reboot.
• Only a subset of upgraded hosts are affected.
• One or both disks are intermittently not discovered during boot.
• The Virtual RAID appears degraded.
• Once the affected disk is physically reseated by the onsite engineer, the disk is detected again and the RAID returns to a healthy state.
• No ESXi configuration changes are required after reseating.
• Other hosts upgraded to the same firmware complete successfully without any storage issues.

Environment

• VMware vSphere ESXi 8.x
• VMware vSphere ESX  9.x
• Cisco UCS B-Series Blade Servers
• UCS Firmware version below 4.3(6d)
• Local M.2 Storage / Virtual RAID M2

Cause

This is known issue for the UCS firmware below 4.3(6d) 

Resolution

Upgrade the UCS firmware to 4.3(6d) for the fix 

Ref: Resolved Caveats in Release 4.3(6.260054)