Referenced Memory Could Not Be Read Error!

[nemaNIN]

Hello All ~

I just built a new mahcine:

MSI K8t Master2-FAR
(2) Opteron 244s
(2) 512Mb OCZ PC3200 Registered/ECC RAM
(2) 160GB WD 7200rpm SATA HDs in Raid 0 on the mobo
etc

Since setting up this box, i've been gettting the following message about once a day in a number of different applications:

Application popup: devenv.exe - Application Error : The instruction at "0x4a619262" referenced memory at "0x00000000". The memory could not be "read".


this is my first experiene w/ registered RAM, AMD procs, and RAID setups.... I'd guess that this is my RAM's fault, since the memory can't be read..... but i really dont know. is this hardware? or driver errors?

any ideas? should i RMA my RAM before the chance expires? i should have until the 23rd....

thanks for the ideas... is there a better place to ask this question?

nema.

 

[StormRider]

It could be a bug in the application. I've seen this message in my code and then I traced it to a bug where I try to access memory beyond an array boundary. Or I try to access an object through a NULL pointer -- which is what your case sounds like.

 

[nemaNIN]

im pretty sure, though, that it's not always the same program.

could it be a hardware thing? it's the whole error checking on the RAM that im wondering about.... bad ram corrupts data and b/c it's ECC wont send it to the application.... is that a possibility?

 

[nemaNIN]

Got the same error last night, but this time in explorer.exe..... so i dont think it's an application bug...

anyone have suggestions?

 

[charloscarlies]

I would run memtest and see if they're causing the problems. Then go from there.

 

[nemaNIN]

i've run the memtest3.1a CD boot disk twice.

both times it appeared to stop responding after 41 seconds (the time stopped counting, at least, and i let it sit for a couple minutes).

the first time: 41 seconds, 48% pass 2
sencond time: 41 seconds, 50% pass 2

so... does that mean bad ram?

 

[Mark R]

It's unusual for memtest to crash - this may mean a bad motherboard or CPU.

Try a different memory testing program - I've found Windows Memory Diagnostic to be harsher than memtes86 and faster too.

Alternatively, if you have spare parts, try swapping the parts out.


Windows Memory Diagnostic Download Page

 

[nemaNIN]

i actually have tried 2 new CPUs (i started with 242s, but was able to do a free upgrade when the 244s dropped in price last week.)

so i dont think it's the procs.

and i'd try replacing the RMA'ing the RAM before the mobo... swapping sticks is hella easier than replacing the mainboard

is OCZ an ok brand?

am i missing other possible causes?

what temp should Opteron's run at? according to CoreCenter Pro CPU1 = 46 and CPU2 = 55.... seems alright to me... right?

 

[Chu]

Originally posted by: nemaNIN
Hello All ~

I just built a new mahcine:

MSI K8t Master2-FAR
(2) Opteron 244s
(2) 512Mb OCZ PC3200 Registered/ECC RAM
(2) 160GB WD 7200rpm SATA HDs in Raid 0 on the mobo
etc

Since setting up this box, i've been gettting the following message about once a day in a number of different applications:

Application popup: devenv.exe - Application Error : The instruction at "0x4a619262" referenced memory at "0x00000000". The memory could not be "read".


this is my first experiene w/ registered RAM, AMD procs, and RAID setups.... I'd guess that this is my RAM's fault, since the memory can't be read..... but i really dont know. is this hardware? or driver errors?

any ideas? should i RMA my RAM before the chance expires? i should have until the 23rd....

thanks for the ideas... is there a better place to ask this question?

nema.

Make sure you don't have sasser. Null memory references in random programs (accessing address 0x0000000) is a classic sympton.

-Chu

 

[MplsBob]

Here's one that I actually had the presence of mind to write down:
The instruction at "0x7C575ac6" referenced memory at "0x077bc000". The memory could not be "read ".

I have gotten several messages along this line and it always seems to be when the room is getting warmer. I had been thinking, at least in my case, that I was having some thermal problems with my ram. I do have one side cover constantly removed from my machine.

You might keep an eye open and see if you see any heat related pattern to their occurrence.

To the best of my recollection I was getting these long before Sasser showed up.

I ran memtest several times and had no problems.

I would dearly love to hear from someone who has the cure for it.

My motherboard is an Asus A7N8X DELUXE using a matched pair of corsair ram to form 1gb total. My machine is not overclocked. I do have two 34gb wd 10k rpm hard drives in mb sata RAID 0.

 

[MplsBob]

I have started doing some research on this subject. It appears that a lot of people have experienced this and similar error messages.

As a starter is this:


"Memory Could Not Be Read" Error Message While Doing File Operation
Microsoft Knowledge Base Article - 267858

It also has references to a couple of other knowledge base articles, and all of this has different levels of applicability depending upon the which Windows operating system you are using.

Good luck.
At the moment I don't have the time to pursue things much farther today. I will try looking farther on Monday.