CentOS - Jul 2009 - x86_64 EDAC throwing error

Hi All,

We have installed CentOS 5.3 x86_64 in an HP DL585 server with AMD Opteron
64 bit processor and 16 GB RAM. The kernel version is 2.6.18-128.el5 . Now
this has thrown an error message in /var/log/message,

Jul  3 21:41:11 db1 kernel: EDAC k8 MC0: general bus error: participating
processor(local node origin), time-out(no timeout) memory transaction
type(generic read), mem or i/o(mem access), cache level(generic)
Jul  3 21:41:11 db1 kernel: EDAC MC0: CE page 0x65bc7, offset 0x6a0, grain
8, syndrome 0x6e1a, row 0, channel 0, label "": k8_edac
Jul  3 21:41:11 db1 kernel: EDAC k8 MC0: extended error code: ECC chipkill
x4 error
Jul  3 22:00:00 db1 ntpdate[3813]: step time server 120.88.46.10 offset
-4.375417 sec
Jul  3 22:12:57 db1 kernel: EDAC k8 MC0: general bus error: participating
processor(local node origin), time-out(no timeout) memory transaction
type(generic read), mem or i/o(mem access), cache level(generic)
Jul  3 22:12:57 db1 kernel: EDAC MC0: CE page 0x65bc7, offset 0x6a0, grain
8, syndrome 0x6e1a, row 0, channel 0, label "": k8_edac
Jul  3 22:12:57 db1 kernel: EDAC k8 MC0: extended error code: ECC chipkill
x4 error

I understand this is an error message from Error Detection And Control
module and just wanna confirm that this is not a kernel or software related
issue. If Hardware related is it confined only to the Physical memory stick
installed or processor related? Any hint on this??

Regards,

Kurian Thayil.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.centos.org/pipermail/centos/attachments/20090704/ab749883/attachment-0003.html>

Greetings,

On Sat, Jul 4, 2009 at 12:34 PM, Kurian Thayil<kurianmthayil at gmail.com>
wrote:
> Hi All,
>
> We have installed CentOS 5.3 x86_64 in an HP DL585 server with AMD Opteron
> 64 bit processor and 16 GB RAM. The kernel version is 2.6.18-128.el5 . Now
> this has thrown an error message in /var/log/message,
>
> Jul? 3 21:41:11 db1 kernel: EDAC k8 MC0: general bus error: participating
> processor(local node origin), time-out(no timeout) memory transaction
> type(generic read), mem or i/o(mem access), cache level(generic)

While I cant throw much light on this particular problem, I have faced
some peoblem as the opeteron based DL 3x5 g5 servers require RAM to be
populated evenly among the CPU sockets

Just my 2p

Kurian Thayil wrote:
> I understand this is an error message from Error Detection And Control
> module and just wanna confirm that this is not a kernel or software related
> issue. If Hardware related is it confined only to the Physical memory stick
> installed or processor related? Any hint on this??
Login to the iLO/iLO2 interface and look at the system event
log, all DL585s will log memory errors there, and will even
tell you what memory module it is.

nate