memtest86 hangs, but Linux and Windows run okay

[anybody]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD to
test some memory sticks. I have tested four sticks, and the
memtest86 test hangs no matter which I use. It barely gets started in
the test before the computer goes dead and I have to power cycle.
However it seems to run Fedora and Windows XP and SuSE fine. Also it
passes the BIOS memory diagnostics. Should I be concerned?

Dell 4500 (Intel BIOS, Intel 845 chipset, P4, 533 MHz)
Spectek PC2100 256M
Micron PC2100 128M
2 x Kingston Value RAM PC2100 256M

 

[steven67@]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD to
test some memory sticks. I have tested four sticks, and the
memtest86 test hangs no matter which I use. It barely gets started in
the test before the computer goes dead and I have to power cycle.
However it seems to run Fedora and Windows XP and SuSE fine. Also it
passes the BIOS memory diagnostics. Should I be concerned?

Dell 4500 (Intel BIOS, Intel 845 chipset, P4, 533 MHz)
Spectek PC2100 256M
Micron PC2100 128M
2 x Kingston Value RAM PC2100 256M

..


Is that an older version of Memtest86? If so, try the latest versions of
Memtest86 and Memtest86+.

Memtest86: http://www.memtest86.com

Memtest86+: http://www.memtest.org

 

[anybody]

anybody wrote:




.


Is that an older version of Memtest86? If so, try the latest versions of
Memtest86 and Memtest86+.

Memtest86: http://www.memtest86.com

Memtest86+: http://www.memtest.org

Thanks. It displays "Memtest -86 v3.1".
Apparently the latest Memtest-86 is v3.1a.

 

[Jose Maria Lopez Hernandez]

anybody wrote:




..


Is that an older version of Memtest86? If so, try the latest versions of
Memtest86 and Memtest86+.

Memtest86: http://www.memtest86.com

Memtest86+: http://www.memtest.org

And have in mind that Linux is quite resistant to hangs,
something can fail and Linux will still be working ok.
Maybe you can have strange application fails, and
problems compiling programs or the kernel, that are the
things that use to happen when you have defective memory.

--

Jose Maria Lopez Hernandez
Director Tecnico de bgSEC
(e-mail address removed)
bgSEC Seguridad y Consultoria de Sistemas Informaticos
http://www.bgsec.com
ESPAÑA

The only people for me are the mad ones -- the ones who are mad to live,
mad to talk, mad to be saved, desirous of everything at the same time,
the ones who never yawn or say a commonplace thing, but burn, burn, burn
like fabulous yellow Roman candles.
-- Jack Kerouac, "On the Road"

 

[kony]

Thanks. It displays "Memtest -86 v3.1".
Apparently the latest Memtest-86 is v3.1a.

Was there even a non-A 3.1 version?

3.1a also displays (only) "Memtest-86 v3.1"

 

[S.Heenan]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD
to test some memory sticks. I have tested four sticks, and the
memtest86 test hangs no matter which I use. It barely gets started in
the test before the computer goes dead and I have to power cycle.
However it seems to run Fedora and Windows XP and SuSE fine. Also it
passes the BIOS memory diagnostics. Should I be concerned?

Dell 4500 (Intel BIOS, Intel 845 chipset, P4, 533 MHz)
Spectek PC2100 256M
Micron PC2100 128M
2 x Kingston Value RAM PC2100 256M

Try running Prime95 under Windows or Linux.
http://www.mersenne.org/freesoft.htm

The "Torture Test" will usually root out RAM or CPU problems rather quickly.

 

[Charles Sullivan]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD to
test some memory sticks. I have tested four sticks, and the
memtest86 test hangs no matter which I use. It barely gets started in
the test before the computer goes dead and I have to power cycle.
However it seems to run Fedora and Windows XP and SuSE fine. Also it
passes the BIOS memory diagnostics. Should I be concerned?

Dell 4500 (Intel BIOS, Intel 845 chipset, P4, 533 MHz)
Spectek PC2100 256M
Micron PC2100 128M
2 x Kingston Value RAM PC2100 256M

I'm unfamiliar with your BIOS, but most BIOS memory tests just
check for the presence of blocks of memory and are by no stretch
a true diagnostic. As suggested by other responses, make sure
you get memtest86+, which tweaked the program to work with some
newer memory management circuitry. (I suspect your particular
memtest86 version is the problem rather than actual bad memory.)

 

[Michael Buchenrieder]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD to
test some memory sticks. I have tested four sticks, and the
memtest86 test hangs no matter which I use. It barely gets started in
the test before the computer goes dead and I have to power cycle.
Urgh.

However it seems to run Fedora and Windows XP and SuSE fine. Also it
passes the BIOS memory diagnostics. Should I be concerned?

Yes. The BIOS test is nothing. It just counts how much memory
is available, nothing else. Its value for testing purposes
is below zero; if the RAM test fails, your memory might
already have been faulty for ages.

Dell 4500 (Intel BIOS, Intel 845 chipset, P4, 533 MHz)
Spectek PC2100 256M
Micron PC2100 128M
2 x Kingston Value RAM PC2100 256M

Do the specs for the board match the specs of the modules?

Hint:
Try compiling a kernel. That's a good stress test as well,
though I fear that most likely you have a HW problem of
some kind, given the symptoms.


Michael

 

[Frank Fussenegger]

Make sure you have the latest BIOS for your motherboard. memtest86
would hang on my system too until I installed a BIOS update that fixed
the problem.

 

[spodosaurus]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD to
test some memory sticks. I have tested four sticks, and the
memtest86 test hangs no matter which I use. It barely gets started in
the test before the computer goes dead and I have to power cycle.
However it seems to run Fedora and Windows XP and SuSE fine. Also it
passes the BIOS memory diagnostics. Should I be concerned?

Dell 4500 (Intel BIOS, Intel 845 chipset, P4, 533 MHz)
Spectek PC2100 256M
Micron PC2100 128M
2 x Kingston Value RAM PC2100 256M

Isn't memtest86 supposed to be put on its own bootable floppy or cdrom
and run separately from other things?

 

[Peter D.]

Frank Fussenegger wrote in comp.os.linux.hardware:

Make sure you have the latest BIOS for your motherboard. memtest86
would hang on my system too until I installed a BIOS update that fixed
the problem.

[snip]

Yes. The README file does say that some BIOSes cause it trouble.

 

[Douglas Clinton]

Try running Prime95 under Windows or Linux.
http://www.mersenne.org/freesoft.htm

The "Torture Test" will usually root out RAM or CPU problems rather quickly.

There is a nice GUI version of Prime95 from:

www.dblsys.ca/sp2004/index.htm

 

[anybody]

Yes. The BIOS test is nothing. It just counts how much memory
is available, nothing else. Its value for testing purposes
is below zero; if the RAM test fails, your memory might
already have been faulty for ages.

I'm not talking about the POST memory test. The Intel BIOS has a bunch
of diagnostics (eg hard drive, CDROM, keyboard, mouse) built in. It
doesn't seem as thorough as Memtest-86 though.

 

[anybody]

Isn't memtest86 supposed to be put on its own bootable floppy or cdrom
and run separately from other things?

That is about how the SuSE install CD works. It presents an option to
run Memtest-86 without booting into the OS.

 

[David Wang]

Which version of Memtest86 were you using? 3.0?
Memtest86 3.0 should support Intel 845 based chipsets.
http://www.memtest86.com/

 

[otruyo]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD to
test some memory sticks. I have tested four sticks, and the
memtest86 test hangs no matter which I use. It barely gets started in
the test before the computer goes dead and I have to power cycle.
However it seems to run Fedora and Windows XP and SuSE fine. Also it
passes the BIOS memory diagnostics. Should I be concerned?

Dell 4500 (Intel BIOS, Intel 845 chipset, P4, 533 MHz)
Spectek PC2100 256M
Micron PC2100 128M
2 x Kingston Value RAM PC2100 256M

Besides BIOS upgrade, and the kernel compilation test, you could also
have a look at your chipset setting in the BIOS setup. Maybe you are
using some specs that your chipset - RAM combination don't like. Try
with some changes, increasing delay times, and so on. I am almost no
confident with this, but maybe it should help.

 

[Michael Heiming]

I am using the memtest86 program from the SuSE 9.0 install/rescue CD to

[..]

Besides BIOS upgrade, and the kernel compilation test, you could also
have a look at your chipset setting in the BIOS setup. Maybe you are
using some specs that your chipset - RAM combination don't like. Try
with some changes, increasing delay times, and so on. I am almost no
confident with this, but maybe it should help.

Hi!

You might like to check your nntp server settings, you are aware
that the post you are replying is 2 month old?

 

[Joe Beanfish]

Besides BIOS upgrade, and the kernel compilation test, you could also
have a look at your chipset setting in the BIOS setup. Maybe you are
using some specs that your chipset - RAM combination don't like. Try
with some changes, increasing delay times, and so on. I am almost no
confident with this, but maybe it should help.

Maybe try the "other" version of memtest86

http://www.memtest86.com/
http://www.memtest.org/

BTW, BIOS memory diagnostics aren't. All the bios does is detect
presence. It's "test" is fairly worthless.

 

[clifto]

BTW, BIOS memory diagnostics aren't. All the bios does is detect
presence. It's "test" is fairly worthless.

Years ago I had the top off my case, and when some coffee hit the wrong
spot in my throat I spewed a mouthful all over the inside. I killed the
power and tilted the unit to let it drip dry, and hoped. When I started
it back up, it easily passed the BIOS memory test, but it barfed on
the *very first* memory write it tried (at x86 address 0000:0000, the
interrupt list).

I managed to rinse the boards off with distilled water and recover the
computer, but I had learned all I needed to know about BIOS memory
tests.