On 29/11/12 02:38 AM, Marc Shapiro wrote:
On 11/28/2012 09:24 AM, Gary Dale wrote:
I'm afraid I can't really answer that. All I can do is repeat that
when MemTest86+ reports an error, it is a good indication that you
have a problem. I've seen this on systems where MemTest86+ reported
only a few problems but the computer locked up intermittently in use.
Replacing the memory with ones that MemTest86+ passed cured the lockups.
I also have a motherboard that was running reliably for 2 years then
started locking up. MemTest86+ reported that the memory was OK but
Klaus Knopper's suggestion that it was a chipset problem seemed
reasonable since the problem occurred during operations where both
memory and disk access were high. The manufacturer meanwhile replaced
the board three times with repaired boards all of which displayed the
same problem. Their repair testing tests components in isolation,
which usually is OK but in this case failed to trigger the real problem.
The problem seems to be more common on modern hardware than on older
systems. I hadn't seen it before but now have seen it on a couple
different motherboards. Slowing down the memory access cures it.
Well, it's not just Memtest86+ that is reporting errors. I decided to
get a second opinion, as it were, and installed memtester from the
Debian repository. I let it run through 5 cycles of its tests. I have
included the results of the first loop below. The other loops were
similar. None of the five loops found any errors in the first 9
tests. The 'Checkerboard' test found errors in 2 of the 5 tests and
only the first test found errors in the 'Waking zeros' test. As you
can see, I was testing 7GB on an 8GB system. This was running from
inside an xterm while I was cruising the web. As you can see from the
results of running 'free' while the test was going, I had under 80GB
(less than 1% of my total memory) free, so the full memory was getting
a workout. I am just having difficulty with the idea that there are
this may memory errors throughout the range of the RAM and I keep
right on working with no apparent difficulties? Others that I have
talked to, who have actually had memory go bad on them (I never have,
before) say that it is extremely obvious and normal operations are not
possible. I believe the term he used was that "the OS completely
wigged out!"
Do I really have bad memory? Or is this some other kind of
aberration. Sunday is the end of my in-store return period. After
that I have to ship things back to manufacturers which is a problem if
I don't really know were the trouble lies.
Of course bad memory doesn't always make the computer lock up
completely. MemTest86+ and memtester clearly are able to trigger errors
without the system screeching to a halt. Conversely, an entire stick can
fail leading to your system running with less memory than you installed.
In this case, you may not notice any problems except for slower performance.
When it's only certain patterns that fail, a system failure would depend
on that particular byte experiencing that particular pattern shift in a
manner that causes the program to noticeably fail. A bit being flipped
in a data area, for example, may lead to a subtle data error, such as a
glitch in a video, or it may not even show up - such as a bit in an
unused portion of a buffer.
My advice is that if MemTest86+ shows you have a memory problem, believe
it. Even if it doesn't show errors, that doesn't mean that there aren't
memory access issues.
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/50b7572b.2060...@rogers.com
