Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I've experienced 8 hard disk failures in 3 months and have tried many things to solve the issue permanently but I have failed. I would like to know if you have any advice for me.

System was running Win XP on an Asus P5W-DH Deluxe. I have setup a RAID-1 array.

  1. I started out with 2 x 500 GB 7200RPM Western Digital drives. One died. I took it out to RMA it. On the same day, the router was fried. Assumed a power surge occurred; connected an older UPS to protect the system.

  2. Once I got my hands on an identical disk, I installed it. The RAID array was rebuilt.

  3. A few days later, the other one died. Assumed the rebuild caused it to fail. Took it out for RMA. Before the other one arrived, the remaining one died.

  4. I then discovered I could re-enable them using the Intel Matrix Storage Manager. I re-enabled both and the system seemed fine for a week, until both died again.

  5. I got two new 1.5 TB 7200RPM Seagate drives and re-installed Windows 7. Also replaced the UPS and power supply. They both died again.

The voltage on the plug is stable between 120 and 122V as per the UPS. None of the other devices have had any problems (monitors, etc.).

At this point, I see two options: a) electrical issue in the house that was, for some reason, not blocked by the UPS. b) something else inside the system causing surges? motherboard? onboard raid controller?

Failures happen fairly quickly, between 2 and 14 days after I fix the previous issue.

I just gotten a new computer (Core i7) to replace it. If it is stable, I can determine that b) was the problem. If it fries its hard drive again, I can determine that it is an electrical issue in the house.

Do you have any other thoughts? Any tools I can run on the drives that failed to get more information about the original SMART event history?

share|improve this question
    
Temperature, electro-magnetic fields, power surge, mechanical 'disturbance' (stop kicking your computer f.e. (no seriously, my brother used to do that, ruined a brand new 500GiB drive)). –  Bobby Jan 2 '11 at 20:45
    
maybe a bad HD controller –  hbdgaf Jan 2 '11 at 20:48
    
From your point 4 it seems that the dead disks could be resurrected. What do you mean then exactly by "died"? –  harrymc Jan 2 '11 at 20:54
    
@harrymc Died: SMART error. RAID controller shows array as degraded or failed. System would not boot. Brought back to life by clearing the SMART error. Note: I assume that they can be revived but that they will eventually die again, maybe permanently. –  Jason Kealey Jan 2 '11 at 21:00
add comment

3 Answers

up vote 1 down vote accepted

Degraded/failed does not mean a dead hard drive, you could have a bad raid controller or bad sata data cables, does not sound like the drive are actually bad to me. Did you know there are hard drive made especially for raid controllers and ones that are not commonly fail in this way because of a hard drive error recovery timeout causing the raid controller to fail writing to the disc? You really should use Raid Edition hard drives or use the tler tool to enable this feature in non RE Western Digital hard drives.

https://encrypted.google.com/search?sourceid=chrome&ie=UTF-8&q=western+digital+tler

.

share|improve this answer
    
At this point, I'm leaning towards a defective RAID controller and investigating your solution as next steps for this computer. –  Jason Kealey Jan 2 '11 at 21:40
    
(And will be disabling RAID-1 and keeping the revived disks in the system) –  Jason Kealey Jan 2 '11 at 21:47
    
Ahm...you know that links to search engines are, most of the time, not considered helpful? –  Bobby Jan 3 '11 at 8:27
    
Ahm... they work for me 100% of the time when I don't know the answer. –  Moab Jan 3 '11 at 15:45
    
@Moab: That might be, but sometimes they're just not the answer to everything...at least not on SE. –  Bobby Jan 3 '11 at 16:31
show 1 more comment

You may find several useful tools here : Best Free Hard Drive Health Monitoring & Diagnostic Programs. I personally like SpeedFan.

That said, I don't believe that you had 8 disks go bad on you without a common reason, which might be your hard-disk controller, PSU or motherboard. I would suggest getting a technician to have a look at the computer to verify all voltages and temperatures. Your computer model is not new, so I suppose that it has no warranty.

It might also be that the point of failure is actually the S.M.A.R.T hardware itself. Meaning that there is nothing wrong with the disks. You can verify that by using some "dead" disks in another computer. Some of the tools in the above link will test a disk, so you can stress-test it and see how fast the errors accumulate. Just keep in mind that S.M.A.R.T is usually not very dependable.

If nothing can be found without much expense, maybe it is time to change the computer.

share|improve this answer
    
I am testing a new computer at this point, to see if it also start giving SMART failures. I ran some self-tests in another computer on the disks and they pass. –  Jason Kealey Jan 2 '11 at 21:41
add comment

Sounds like there is an electrical problem in your box. You should probably RMA the motherboard and power supply, and replace the sata cables. If you have one, try to replace the RAID controller. If not, is your computer overheating? Overcooling? Cold temperatures can be just as bad for HDD's than warm temperatures. Try to remain at room temperature or slightly cooler or warmer. Is your computer near a strong magnet? Is it very humid in the room? are you sure that the UPS isn't overvolting or something?

There's a lot of variables in this sitation, try to isolate the problem.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.