[IP35 Pro] Raid constantly failing (Error occurred)

[adam1v]

Hi,
I cannot find any answers to what im looking for, so I thought id try on here.

I have two samsung 750 hard disks, in raid 0 on my Ip35 pro board.
A few months back, one of the disks was reported in a bad state, when loading through dos, the system would report 'Error Occurred(0)'.

I decided to start fresh, delete the raid, 0 fill the disks and start everything from fresh. I made another raid 0 array, and just a few weeks on, the other disk reported an error.
Theres no degraded slow speed, and the disks appear to be fine.
Around a week after that, the second disk was also marked as an error, and for the last month or so, ive carried on with both disks marked as bad, without any issue or slow speeds.

Has anyone else experienced this?
I thought it might just be me, but I recently set a friends PC up, he has identical setup and one of his disks has been flagged as bad now, and im declinded to think its the disks. If it was the disks im sure I would notice it in the speed.
The samsung HTUIL doesnt see any errors either.

Today I upgraded to the latest Bios, but ofcourse theres no difference.

Anyone have any suggestions?

[Lvcoyote]

Contact the manufacturer of the HDD's in question. Perhaps there is a firmware update.

[adam1v]

Samsung dont/wont release firmware for their drives

[Richh]

Raid arrays built on Intel controllers are normally rock solid. Have been running RAID on ICHxR's since about 2002 (and Highpoint controllers prior to that) and in that time the one array failure I've seen was due to a drive expiring. As such if you're continuously seeing error notifications, the likeliest suspect would be the drives themselves, but there are one or two other causal factors you could try ruling out...

The Intel Matrix Storage Controller will report storage errors that aren't fatal (a "mere" CRC read error is enough to throw a warning), so it may well be that this is much ado about nothing. However, you could, theoretically, provoke this kind of thing by overclocking the system in such a way as the southbridge (or the sb-northbridge link) is running close to the edge of stability.

Are you overclocking? If so, have you tried raising the southbridge-related voltages a click? Also, what about your PCI Express frequency, still on 100MHz/Auto, or raised?

[adam1v]

Right, a slight update.
Im not overclocking and when copy large file (700mb) to/from the PC, it will either flag up a blue screen or reboot itself.

Just now, after it rebooted, the raid reported a bad disk (during startup), reported the volume as non bootable and gave 'missing operating system' message.
I hit the reset button, the system reboots, everything is flagged as healthy and the system starts.
The same process happens over and over.

On a side note, I recently built a brand new setup for my firned, using identical hardware (samsung 1tb, with the same mobo), and he now has exactly the same problem.

I've tried several hard drive tests (including samsungs) and not found any errors on the drives themselves.
Does anyone know any software which could help rule out other hardware issues? i.e the controller?

[adam1v]

apologies for spam, but i need to make 2 posts before posting an image. I think i may have found the cause

[adam1v]

post2

[adam1v]

im not clocking (I havent found a need to clock my QX9650), but I have just spotted this:



Ive tried to re-install, find the solution etc, but its not made any difference.
If im honest, im not really sure what im looking at, and whether or not it could be my cause?