dmesg entries and memtest

Johan Booysen johan at matrix-data.co.uk
Thu Sep 6 12:11:14 UTC 2007


Can anyone maybe help me understand what the following means?  
 
I know it's something to do with memory, but can't find good information
on what is happening here.
 
This is part of the output of dmesg on a RHEL5 server that's being used
for testing Oracle 10r2 (if that helps):
 
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
EDAC MC0: CE page 0x27ca8, offset 0x0, grain 4096, syndrome 0xf3e0, row
4, channel 1, label "": e7xxx CE
EDAC MC0: CE - no information available: e7xxx CE log register overflow
 
I've tried memtest but it seems to hang after running for about 45
minutes.  It tells me that it's done 10% and is on test 6 at that stage,
but to be honest I'm not sure exactly how to interpret what memtest
shows.  I've taken a picture if that helps:
www.supernaut.plus.com/06092007313.jpg
 
Will really appreciate it if anyone can help me with advice on this.
The only thing I can think of is to remove memory chips and run memtest
on each one individually (if I know what to look out for in memtest, of
course - I find the memtest info on the Internet quite confusing at the
moment and need to figure this out as soon as possible).
 
Thanks very much.



More information about the redhat-list mailing list