Intel-Based HPE Gen11 Servers - 使用第五代（Sapphire Rapids）LCC 处理器可能会观察到多个 UMCE错误

Intel-Based HPE Gen11 Servers - Multiple UMCEs May Be Observed with 5th Generation (Sapphire Rapids) LCC Processor

文档ID：a00151188en_us    Last Updated: 2025-09-02

影响范围

任何HPE Intel Gen11 服务器搭载如下Gen 5 (Sapphire Rapids) LCC Die 处理器：

- Bronze 3508U
- Silver 4510
- Silver 4509Y

HPE Intel Gen11 服务器搭载第五代 (Sapphire Rapids) LCC处理器，服务器处于闲置状态时，可能会检测到如下一个或多个UMCE或PCIe错误。

3 strike watchdog timeout (MSCOD shows E1Cx)

Uncorrectable Machine Check Exception (Processor 1, APIC ID 0x00000000, Bank 0x00000003, Status 0xB2000000'E1CD0400, Address 0x00000000'00000000, Misc 0x00000000'00000000) Uncorrectable Machine Check Exception (Processor 1, APIC ID 0x00000010, Bank 0x00000003, Status 0xBE000000'E1C40400, Address 0xFFFFF806'526B3842, Misc 0xFFFFF806'526B3842)

IO MCA 指向 PCH root port (Bus/Dev/Func: 0x00/0x0D/0x00)

Uncorrectable Machine Check Exception (Processor 1, APIC ID 0x00000000, Bank 0x00000006, Status 0xBA000000'00000E0B, Address 0x00000000'00000000, Misc 0x00000000'00680000)

TOR timeout指向物理内存空间 0x500

Uncorrectable Machine Check Exception (Processor 1, APIC ID 0x00000000, Bank 0x00000009, Status 0xFE200000'000C110A, Address 0x00000000'00000500, Misc 0x200773C0'00240086)

iLO 上检测Uncorrectable PCIe error (Bus/Dev/Func: 0x01/0x00/0x02)

Uncorrectable PCI Express Error Detected. [UNKNOWN] 0 (Segment 0x0, Bus 0x1, Device 0x0, Function 0x2)

N/A

此问题HPE仍在调查中，HPE将在未来BIOS版本中解决此问题；

临时方案，可以禁用Core C6和Package C6来规避：

重启按F9进入BIOS：

1. 进入 System Utilities，导航到System Configuration > BIOS/Platform Configuration (RBSU) > Workload Profile"，选择 "Custom"

2. 导航Power and Performance Options ->Minimum Processor Idle Power Core C-State，选择No C-State

3. 导航Power and Performance Options -> Minimum Processor Idle Power Package C-State，选择No Package State

4. 保存重启退出。