SPX86A-800C-WR - ADC Activated (Doc ID 2698328.1)

刘泽宇

DETAILS
ADC Activated

Type

Fault
  alert.memory.intel.dimm.adc.activated

Severity

Minor

Description

Message ID: SPX86A-800C-WR indicates that the ILOM fault manager has applied diagnosis to error reports
received and has determined that Adaptive Device Correction (ADC) has been activated on the first occurrence
of encountering correctable memory errors that could possibly result in an uncorrectable memory error.

This ALERT is used to provide notification that Adaptive Device Correction (ADC) has been activated.

ADC activation will successfully mitigate the bad region(s) found on the DIMM by sparing
those affected memory DRAM devices to other healthy banks on the affected memory channel.

There is no immediate action required to replace the affected DIMM unless
the system is experiencing system performance or system needs to be rebooted.

NOTE:
If you decide to reboot the server, then BIOS will execute Post Package Repair (PPR) on the
affected DIMM in an attempt to permanently repair the affected DRAM region on the DIMM,
thereby potentially reducing or eliminating future memory correctable errors or performance issues.

There is no guarantee that the affected regions on the DIMM will get mapped out, and since ADC is no longer
activated because of the reboot, it's possible that the system could encounter a memory UE event if the affected DIMM was not replaced.



Regarding Post Package Repair (PPR)....

PPR is a feature in the BIOS on Intel-based platforms, that when enabled,
may be able to repair affected DRAM areas on a DIMM.

Upon encountering any memory related fault event during Memory Reference Code (MRC) initialization or
experiencing certain memory correctable events during runtime that can trigger Adaptive Device Correction (ADC) on
first occurrence, then PPR would be activated after the next system initialization and attempt to repair the DIMM.



Automated Response

Adaptive Device Correction (ADC) has been enabled and is currently sparing those affected memory DRAM devices to other healthy banks on the affected memory channel.
Post Package Repair (PPR) may be able to repair the affected DRAM areas on the DIMM after the next system initialization/reboot.
The service-required LED for the chassis and the affected memory DIMM(s) will be illuminated.
This ALERT event will be automatically cleared upon the next system reboot.



Impact

The system will continue to operate in the presence of this alert.
ADC activation will successfully mitigate the bad region on the DIMM by sparing with other banks on the affected memory channel.
The memory DIMM is still in use and is not disabled.
ADC activation may cause overall memory performance to be slightly reduced.



Suggested Action for System Administrator

The indicted DIMM should be scheduled for replacement before the next system reboot.

If a performance issue is observed, then replacement of the indicted DIMM should be scheduled as soon as possible.

This event will be automatically cleared upon reboot.

ADC will not be automatically re-activated if the system is rebooted.
If the memory DIMM is not replaced before the next reboot this can potentially create an increased risk of
running into a memory UE if the affected memory DRAM areas were not successfully repaired by PPR on the reboot.

Due to this situation, if the DIMM is not replaced during the next reboot, then FMA will then report
"SPX86A-800D-9Q - DIMM UE Predicted" against this DIMM location indicating the need for it's replacement.