I have 8GB of Transcend DDR2-800 ECC unbuffered RAM in my home server running WHS 2011. When I first installed it a couple years ago, I ran Memtest 86+ for 24 hours twice to check it out and it passed just fine (tested in both ECC and non-ECC modes). It's been running ever since on the same Asus M4A785-M motherboard with ECC basic mode enabled.
Recently, I started getting unexplained server reboots and freezes. After some troubleshooting, I discovered that one of the DIMMs is bad. What's interesting is that when I run it all by itself, the bad DIMM passes Memtest with no problem. But paired with another DIMM it fails immediately or almost immediately.
The interesting part is that the ram hasn't been touched since I installed it--the box has never been opened. No overclocking and RAM is running at stock settings. And it's not a RAM slot problem, I've tried it in various slots, and did my recent testing in another system with the identical motherboard.
So, what are possible causes of this kind of failure? Just bad luck, or something else?
Recently, I started getting unexplained server reboots and freezes. After some troubleshooting, I discovered that one of the DIMMs is bad. What's interesting is that when I run it all by itself, the bad DIMM passes Memtest with no problem. But paired with another DIMM it fails immediately or almost immediately.
The interesting part is that the ram hasn't been touched since I installed it--the box has never been opened. No overclocking and RAM is running at stock settings. And it's not a RAM slot problem, I've tried it in various slots, and did my recent testing in another system with the identical motherboard.
So, what are possible causes of this kind of failure? Just bad luck, or something else?