7

How can I monitor if there are any errors in RAM that get corrected by ECC? The processor is an Intel Xeon (Ivy Bridge) processor, the operating system is Scientific Linux 6.3. On a previous system I had an AMD CPU, and on that system I could use edac-util to get this info, and it would also issue alerts to the kernel log.

3
  • Here is what happens when I run edac commands on the Intel system. Sorry for the poor formatting. # edac-util / "edac-util: Fatal: Unable to get EDAC data: Unable to find EDAC data in sysfs" | # edac-ctl --status / "edac-ctl: drivers not loaded." | # edac-ctl --mainboard / "edac-ctl: mainboard: ASUSTeK COMPUTER INC. P8C WS" Commented Mar 14, 2013 at 22:39
  • 1
    What's your chip set? Also check the output of (z)grep CONFIG_EDAC_ /proc/config(.gz)
    – daisy
    Commented Mar 15, 2013 at 0:30
  • It's C216 (an LGA 1155 workstation board). I don't have /proc/config*, but maybe /boot/config-2.6.32-358.2.1.el6.x86_64 does the same, in that case I get: # CONFIG_EDAC_DEBUG is not set CONFIG_EDAC_DECODE_MCE=m # CONFIG_EDAC_MCE_INJ is not set CONFIG_EDAC_MM_EDAC=m CONFIG_EDAC_MCE=y CONFIG_EDAC_AMD64=m # CONFIG_EDAC_AMD64_ERROR_INJECTION is not set CONFIG_EDAC_E752X=m CONFIG_EDAC_I82975X=m CONFIG_EDAC_I3000=m CONFIG_EDAC_I3200=m CONFIG_EDAC_X38=m CONFIG_EDAC_I5400=m CONFIG_EDAC_I7CORE=m CONFIG_EDAC_I5000=m CONFIG_EDAC_I5100=m CONFIG_EDAC_I7300=m CONFIG_EDAC_SBRIDGE=m Commented Mar 16, 2013 at 14:16

2 Answers 2

3

Since version 3.17 of the Linux kernel, ECC errors on E3 Xeons can be monitored using the ie31200_edac driver, introduced by this commit. This uses the standard EDAC interface so errors can be listed using edac-util.

In a little more detail:

sudo modprobe ie31200-edac

loads the driver, which will result in lines like

[   14.635299] EDAC MC: Ver: 3.0.0
[   14.637898] EDAC MC0: Giving out device to module ie31200_edac controller IE31200: DEV 0000:00:00.0 (POLLED)

appearing in the kernel log (that’s on a C216 Haswell system); then

edac-util

will report any errors.

Xeon D, E5 and E7 memory controllers are supported using the sb_edac or skx_edac modules.

1

As for as I can find only E5 Xeons are supported with the sb_edac module

http://www.spinics.net/lists/linux-edac/msg00846.html

0

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .