Hey guys.

I have a fileserver at home which was maxed out with 5 harddrives, a dvd writer, some tv cards, asterisk cards and stuff. It was all in a little 'midi' case which was cramped. But it did have some good cooling with a 'rig' made of lollipop sticks, cable ties, and a fan torn off a Pentium 4 heatsink - A-Team style.

Last night I swapped my system over to a Supermicro SC750A huge tower case from eBay. Most of the drives still have adequate cooling, but now a couple of them do not. It seemed ok at the time, it was late and I went to sleep. About 8 hours later, I noticed this in dmesg:

Code:

---

hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }

hda: dma_intr: error=0x40 { UncorrectableError }, LBAsect=150556950, high=8, low=16339222, sector=150556950

ide: failed opcode was: unknown

end_request: I/O error, dev hda, sector 150556950

Buffer I/O error on device hda3, logical block 18550530

---

It is repeated many many times with different block and sector numbers.

Immediately running ide-smart /dev/hda on the drive, a Maxtor 6Y160P0 (my boot drive) showed no 'failures' - is this gospel? hdd-temp shows that at idle the drive is running at 50 degrees C - too hot for my liking. In the middle of the night I expect the drive indexing thing drove it over the 55 degrees operating temperature limit in the specification. It does appear that there is some damage, when I run Cacti some of my graphs are missing - they were definately there last night.

I'm not sure exactly what I should do now. I've shut down the machine until I get home later. Can anyone suggest what would be my next steps to try and recover this mess?

Cheers