Constant blue screens with larger amounts of RAM

We have a machine that was running with 2 gigs of of RAM, we used an image master to copy the drive to a larger one with more space. Upon booting up the new larger drive that is an exact duplicate the system would stay up for about 20 minutes before it blue screens. Both 1 gig sticks are fine as they alone in the machine keep it up for days at a time. Both slots are good because either 1 gig stick in either slot (all four configs) stay up for hours at a time. Also a 1 gig stick coupled with a 512 stick have no problems. Ran memtest86 on all sticks, passed all tests. Obviously its not a problem with the slots or the sticks themselves. Whats odd is if the system sits idle with no users (terminal server) the machine will not crash, but when disk activity starts, its usually a matter of minutes before it blue screens with 2 gigs in. The system is fully updated. And there were NO hardware changes aside from the actual drive before this problem started. All i could imagine is some sort of corruption from the copying, but what would cause this only to happen at above 1.5 gig? For good measure i did 2 chkdsk /r s on the box and updated all possible device drivers, as it looks from this log that it may be a driver, here is the text after running through debugwiz. Note I have not tested since chkdsks and device updates as its a strenious task getting everyone out of this box, im hoping I only have to do it one more time so i wanna be sure i cover all the bases before i do.


Debuglog.txt

 

This looks like pool corruption to me.

you've got a couple of drivers that are pretty long in the tooth.

eb607000 eb607680 giveio giveio.sys Wed Apr 03 19:33:25 1996 <-- why is this running on a server??
eb2f8000 eb2fc800 RTL8139 RTL8139.SYS Wed Aug 18 11:33:29 1999 <-- nic


Heres my guess, the crashes are a result of a bug in a driver, but its not normally exposed, since the 2G configuration is changing how the virtual space is being allocated. This kind of bug is pretty rare, but not unheard of.

My recommendation at this point would be to update these two drivers, (noting i dont think theres an update for GIVEIO.SYS) and if the problem persists, enable special pool, monitor for a new crash, and then run it through the wiz again to identify the malfunctioning component.

 

Thanks for the reply. I guess I don't know what giveio.sys is for? Might it have something to do with an external USB drive we had connected at one point? Also I do recognize that RTL8139.SYS as a Realtek driver, which is actually not in the machine anymore, I wonder why its still loading. The only active NIC in the system is a SMC (SMC1211.SYS, although oddly not loaded at the time of the crash?)

The good news is we solved the problem. Underclocking the memory. Coincidentally this motherboards max amount of memory is 2gig, and rated at 266mhz. The memory we have in is rated at 333 so it must be a failure on the part of the memory bus on the motherboard? I'm thinking running max memory at max speed just caused it to choke. We currently have the 2 gigs running stably at 200mhz. Thinking positively, these problems did help my arguement for full rehaul of the rig.

Thanks again for the reply