[CIA]

I finally bumped my machine up to 16GB, 8x2GB sticks. As a precaution I ran Hardware test off the disks that came with my MacPro. When I first installed I ran an extended test during the day and it was fine. A week later I had reinstalled my OS and had some odd issues, so on a whim I ran it in loop mode (extended test) over the weekend, but it failed after 2 passes:
4MEM/46/50000006:0x78ab6e18

Then I ran it again last night when I left work and it was still crunching away just fine today when I came in. 7 passes and no problems.

Was that one failure a glitch in the matrix, or should I be worried? I haven't noticed any odd crashes in daily use, my original issues cleared up after some software tweaks.

Comments?

[olePigeon]

You could try taking the RAM out and reseating it.

[Spheric Harlot]

IME, if AHT finds nothing, it doesn't mean everything is okay.

If it finds an error, there's something broken.

[CIA]

Well, I ran it again over the weekend. Extended test, loop mode. This time it went 42 passes, and finally failed late Sunday night.
4MEM/9/40000006:0x78f7b018

It should be noted there were crazy thunderstorms in the area when it failed. Someone here at work mentioned that if you run a test long enough, something will fail no matter what. So I don't know.

Based on the two error reports, it seems like it's the same stick though. Can anyone decipher it and tell me which actual slot the stick failed in? 4MEM is consistent on both errors.

[Big Mac]

I don't think it's true at all that lengthy testing will always result in an error message. Any failure message would automatically cause me to RMA the module in question. The odd behavior you noted when running the system is a giveaway sign of bad RAM.

[CIA]

I would RMA the stick in question, but can't seem to figure out which one it actually is. 4MEM is consistent with both errors, how do I figure out what actual slot that is?