CSCws49799

On Cisco UCS C-Series servers, drives may be incorrectly reported as Degraded in the Out-of-Band (OOB) interface after hot-plugging during Virtual Machine (VM) operations on ESXi 9. This issue occurs because the system fails to properly process specific NVMe-MI command responses, leading to inaccurate performance metrics and persistent, false health alerts that do not clear even after physical drive reseating.

There is no direct workaround for this issue. The problem is being addressed through code improvements that enhance how the system handles device response.

C-Series: 6.0(2.260068)

CSCwt18924

On Cisco UCS C220 M6 servers with VIC 1467 adapters, a fatal crash due to out-of-memory errors occurs when upstream ACI switches are upgraded and rebooted. Physical ports show as up, but vNICs remain stuck in initializing and do not recover automatically. Adapter logs indicate resets from fatal errors.

Reboot the server to restore vNIC connectivity.

4.3(5.250030)

CSCwt34436

On Cisco Unified Edge servers (UCSXE-130C-M8-20, UCSXE-130C-M8-32), the server discovery workflow fails at the Wait for Server Inventory Completion task. This issue occurs after upgrading Unified Edge and causes the Server Discovery workflow to halt at the Inventory Collection task, affecting 3 out of 4 servers.

Manually initiate a 'Rediscover' on individual servers to complete server discovery successfully.

6.0(2.260042)

CSCws19521

The downgrade of the server firmware to any release earlier than 6.0(2.260040) for Cisco UCS x210C M8 server equipped with UCSX-X10C-PT4F controller fails due to a Complex Programmable Logic Device (CPLD) firmware version mismatch.

Exclude the CPLD component from the HFP.

6.0(2.260040)

CSCwt42347

During the Host Firmware Package (HFP) upgrade from release 4.3(2) to 6.0(2), Cisco UCS X-Series servers with UCSX-X10C-RAIDF controller experience failure at the stage Preparing to check hardware configuration causing the Unified BIOS Management (UBM) upgrade to hang. The firmware update process does not complete, and BMC logs show the update process running indefinitely.

There is no known workaround.

This issue occurs when a UBM firmware update is needed. The UBM component affects only the UCSX-X10C-RAIDF controller and does not impact other mezzanine cards that include UBM parts.

6.0(2.260040)

CSCwt41941

On Cisco UCS C220 and C240 M8 servers, firmware versions using Intel Secondary Service Processor (SSP) 2.0 are not backward compatible with earlier releases using SSP 1.6.

The firmware versions that support SSP 2.0 are: 4.3(6.260003), 6.0(1.250192), and 6.0(2.260044) or later.

Downgrading to an earlier version causes the Baseboard Management Controller (BMC) to fail BIOS attestation, leading to critical system instability.

Symptoms include systems getting stuck in the KVM console, boot failures, POST hangs, Catastrophic Error (CATERR) events, inoperable NVMe devices, and temperature sensor failures along with critical thermal threshold errors.

If you downgrade to an earlier firmware version and BMC fails, use the following recovery steps:

1. Upgrade BMC Firmware: Use the Host Software Upgrade (HSU) or the Cisco IMC interface to upgrade only the BMC firmware component to the latest version (SSP 2.0 compatible).

2. Perform a Host Power Reset: A full host power reset may help recover the system state, though it is not guaranteed to be 100% effective in all scenarios.

6.0(2.260044)

6.0(1.250192)

4.3(6.260003)

CSCwt35920

Firmware updates may fail on certain RTX Pro 6000 GPUs shipped as spares, leaving them at an older firmware version. This issue is limited to specific spare units and does not affect the overall update process.

There is no known workaround.

X-series: 6.0(2.260040)

XE-series: 6.0(2.260042)

C-series: 6.0(2.260044)

B-series: 6.0(2.260040)

CSCwr82017

On Cisco UCS X210c M8, C220 M8, and C240 M8 servers, the system continuously reboots if you enable Intel SGX and set the PRMRR size to an unsupported value in the system memory settings. This issue occurs only when an unsupported PRMRR size is configured.

When enabling Intel SGX, make sure to configure the PRMRR size to a supported value.

X-Series M8: 6.0(1.252114)

C-series M8: 6.0(1.250192)

CSCwr05911

On Cisco Unified Edge servers, when the host is powered off, the IMM Inventory page displays the LOM-NIC-1 interfaces as up on eCMC:A, but as down on eCMC:B. This inconsistency occurs only when the host is powered off. There is no functional impact due to this issue.

There is no known workaround.

6.0(1.251030)

CSCwr63339

On Cisco Unified Edge servers, if the chassis enters an insufficient power state during runtime (after the node’s BMC has already booted), you can still power on the CPU using the physical power button on the front panel. This may cause the BMC to become unstable for that node, as indicated by specific logs/events. The issue does not occur if the chassis is already in a power-deprived state before the server boots.

Remove the node that is causing high power usage, or upgrade to a power supply unit (PSU) with higher wattage to provide enough power.

6.0(1.251030)

CSCwr88145

On Cisco UCS XE M8 servers, the 2x10G front port (FP) LEDs remain ON even after the network cable is unplugged, if the port speed is set to 1 Gbps.

The LED only turns off after a full server power cycle.

Restart the server to turn off the LED after unplugging the cable from the 1 Gbps port.

6.0(1.251030)

CSCwr04128

On Cisco Unified Edge servers, server discovery intermittently fails in the Intersight workflow with the error message: "Adapter model CPUIntegratedNIC is not supported." This issue occurs occasionally and does not affect every discovery attempt.

To resolve, re-initiate the discovery process when this error appears.

6.0(1.251030)

CSCwq72323

The Compute Sled may fail to power off due to a very rare internal CPU issue. During the power off process, the CPU does not set SLP4_N and SLP3_N signals low, preventing the system from powering down. Manual power operations via IMM may return errors such as Failed to set host power or Timer expired. The issue persists until the next AC cycle.

To restore normal operation, perform an AC cycle of the Compute Sled:

1. Log in to Intersight.

2. Navigate to Unified Edge > Unified Edge Chassis

3. Open the Actions drop-down menu on the right

4. Click Power Cycle Unified Edge Slot, enter the server slot number, and press the Power Cycle button

The system will become operational after a few minutes.

6.0(1.251030)

CSCws30219

Cisco UCS M7 servers are susceptible to host lockups triggered by ECC (Error Correcting Code) memory events.

This issue is specifically related to how the server’s BIOS interacts with the EFI (Extensible Firmware Interface) during error conditions. During these events, the BIOS does not correctly retrieve a required variable, causing EFI to return an invalid parameter. As a result, the operating system is unable to continue normal operation and freezes without logging any errors. When this occurs, the host becomes unresponsive and KVM (keyboard, video, mouse) input is not accepted.

Single-bit ECC errors—while not logged due to DDR5 specifications—can trigger this condition, resulting in complete loss of host responsiveness.

Downgrade server bundle:

• For X-Series servers, downgrade the server bundle to version 5.4(0.250040) or an earlier release.

• For C-Series servers, downgrade the server bundle to version 4.3(6.250044) or an earlier release.

X-Series M7: 5.4(0.250048)

C-series M7: 4.3(6.250053)

CSCwq61388

On Cisco UCS C-Series M6 servers running CIMC 4.3.5 or later, enabling VMD in BIOS with multiple NVME drives can cause storage subsystem initialization to exceed 15 minutes, leading to Intersight Server Profile deployment failures. Disabling VMD or reducing the number of drives can resolve the issue. Earlier firmware versions may not be affected.

• If VMD is not needed, disable it.

• If disabling VMD is not possible, use the minimum number of NVME drives for setup, as more drives slow down system startup.

4.3(5.250001)

CSCwm33093

On Microchip-based storage cards, the Virtual Drive (VD) shows as "Inoperable" even when all disks are online because the firmware automatically changes drive assignments to optimize performance, ignoring user settings from management tools.

Use arcconf output to check drive assignments in spans. Firmware may override management tool settings for better performance.

4.3(5.240134)

CSCwq17020

After installing U3 Micron drives with capacities of 3.8TB or larger in JBOD mode behind the UCSX-X10C-RAIDF controller, Linux OS fails to boot due to BIOS errors related to loading the EFI boot image.

This issue occurs specifically on Cisco UCS X-Series M8 servers equipped with Intel® processors and affects multiple Linux distributions. The problem does notoccur when the drives are configured in RAID 0.Microsoft Windows® and Linux OS boot successfully on smaller capacity drives or when using RAID 0.

Install the OS on drive configured in RAID.

X210c: 5.4(0.250037)

X215c: 5.4(0.250035)