Multi-bit memory errors detected on a memory device at location(s) DIMM_A10

Products

• Security Analytics

Issue/Introduction

This message appears on the console during system boot and will not proceed to boot Security Analytics.  It will get stuck in a booting loop.

Environment

Dell R640xl

Dell 14G servers

Resolution

This type of POST loop after memory errors was patched in BIOS 2.10.2. These steps will resolve the POST loop and likely the memory error/s after the last restart.

BIOS 2.10.2 fixed an Intel errata reported on IPU 2020.1 and 2020.2 where the system could potentially hang or reboot during POST when "Configuring Memory..." is displayed on the screen. This scenario may be encountered during system boot, where self-heal is attempted. The root cause is a race condition where during the self-heal process, a refresh cycle may not have completed when the CPU issues self-heal related commands to the DRAM.

Steps to resolve POST hang or loop on 14G servers:

1. Pull all memory dimms from the server.
2. Then replace only A1 and B1 dimms (please make sure to actually reseat A1 and B1 not just pull all but A1 and B1).
3. Then Flash the BIOS and iDRAC to latest release with only A1 and B1 dimms.
4. Latest BIOS: https://dl.dell.com/FOLDER08418169M/1/BIOS_8JY6W_WN64_2.14.2.EXE
5. Latest iDRAC: https://dl.dell.com/FOLDER08299705M/1/iDRAC-with-Lifecycle-Controller_Firmware_FPTF1_WN64_5.10.10.00_A00.EXE
6. Then shutdown again and replace all the other dimms.
7. Restart and monitor POST you may see memory configuration has changed hit F1 to continue.
8. Restart the server 1 more time to confirm issue resolved and no dimm errors.
9. We recommend updating all 14G server BIOS and iDRAC to avoid this behavior going forward.

Here is the online link to the system memory map of dimm slot locations: Dell EMC PowerEdge R640 Installation and Service Manual