What is causing my HDDs to fail?

D

Dark Light

Weaksauce
Joined
Jul 19, 2009
Messages
77
I am currently using 2x 750GB HDD in Raid0, with the onboard Raid controller from my P5Q Deluxe. 3 weeks ago I started having problems with my current setup, it would temporarily loss one of the HDDs, that are in Raid, when it would do this my PC would freeze. The freezing and the loss of the HDDs got worse and worse over the day, until I could not pick one of my HDDs at all during boot. Shortly after that the 2nd HDD came up with an error on the Raid check. I then used Ubuntu for a day, and after that everything worked fine, no HDD problems at all for three weeks.

Yesterday the problems returned, in the same order as they did the previous time, I'm writing this thread from Ubuntu live CD. Only this time there are more problems then before, some of them not related to the HDDs.

When I get a freeze, from the loss of a HDD, the PC locks up. I was playing Left 4 Dead last night and when that happened the game and PC would freeze for about half a second and I would get a sound loop of about 0.1 seconds. I then ran HDD tune, needless to say that at some points the transfer rate went down to 0, after a while when I got another freeze, the PC did not could out of that one. I got a blue screen of death, lucky the blue screen of death froze so the PC did not restart, so I could read the message. The blue screen of death stated something about the raid driver.

Today my PC worked fine for about an hour, then the problems started again, only this time a lot worse. Firstly when I initially booted my PC I got a boot message stating that the overclock has failed, I have got this message on other occasions when things have gone wrong, that do not involve overclocking, when I had the FSB set the manual but all the settings at stock so my overclock had not failed. Today on occasions my PC would boot but nothing would come up on the screen. Also not sure if its relevant but when my PC is overclock and when I press the power button my PC turns on, and then immediately turns off again, before turning back on, this time booting. My PC did not do that today.

This has happened to me before, this is the 3rd Raid set that has given me problems. The last time this happened it happened in the exact same way as this time, I thought that it was a faulty HDD. So I returned it and got a replacement, that was about a year ago, and everything worked until now. I also had a similar problem with a previous set off HDDs, I think that it happened the same way as the 2nd and 3rd set. It occurred after I got a PSU, motherboard and CPU upgrade.

I have had 3 HDD fail, and possibly 5 give errors on the same system, I did get a new motherboard after the 1st failure and sometime between the 1st and the 2nd I got new memory I also got a new graphics card a while ago.

It seems very unlikely that all the HDD failures are not related, and I can rule out all my components apart from my PSU, CPU and DVD drive.

It seems unlucky that my CPU would cause these problems, without indicated elsewhere and I doubt that it is my DVD drive. SO that just leaves my PSU, which would explain the PC turning off then on again.

So what is causing my HDDs to fail.

Complete list of components.

P5Q Deluxe
Q6600 @ 3.0 (temps are below what the stock heatsink gets on stock)
4GB DDR2 1066
4870 1GB 1000\775
2 x 750GB 32MB cache Raid0 (3 HDD in total, I returned one)
HDT-S1283
Antec 900
750W Tough Power ThermalTake

List of components that were in the system with the Q6600 and PSU
P5K-E
Q6600 @ 3.6 (Same Q6600)
DDR2 667 5-5-5-15 @ DDR2 800 6-6-6-18
8800 GTS 640
8800 GTS 320
An old PCI-e card I used as a temp card
2 x 250GB HDD Raid0 (3 HDD in total, I returned one)

There has been a total of 6 HDD in the system, 3 of which has completely failed, one that gave an error, one that might have an error not sure, and one that has given no problems at all (however I did that have that set of HDDs in the system very long).

I'd like to add that it is possible that the failed HDDs were connected with the same power connector and\or sata cable, and that the HDDs in the 3rd set are cooled with a after market heatsink.
 
First off I'd run memtest just to make sure your memory isn't acting up and please set the timing and voltage manually do not use "auto".
Have you looked at you voltage levels to see if anything is off there?
You should be looking at the 12v and 5v readings with a DMM, and you can take the readings off any 4 pin molex.
 
I ran memtest a while ago, all was OK. Last I checked using PC Probe the 12v and 5v readings were just around 1% off.
 
Dark Light said:
I ran memtest a while ago, all was OK. Last I checked using PC Probe the 12v and 5v readings were just around 1% off.
Click to expand...

Run Memtest again. And this time do the following:
Let Memtest+ run for at least three hours on each stick of RAM separately as well as test the RAM all together. Go for a full 24 hours if you want to be completely sure that the RAM is not a problem. If you start seeing errors, than your RAM is defective or you have incorrect settings for the RAM.

Also, PC Probe or any software that reads PSu voltages are completely and utterly inaccurate. Might as well consult a witch doctor if you want more accuracy.

If you can, buy and use a digital multi-meter to monitor the voltages coming from the PSU. Follow the instructions for "Using a multi-meter to check voltages":
http://www.bfgpower.com/troubleshooting.html#DMM

Also, do you have brownouts or blackouts in your area/home?
 
I start memtest around 10 mins ago, I had it in my floppy drive and when I reset everything to default it booted in, I never intentionel booted into memtest. I'm in the UK so I have not had a black out since I got my PSU, motherboard and CPU upgrade.
 
That is what I am going to find out.

Also I ran memtest for 2:51 hours and I got 0 errors.
 
Last edited:
update: I ran Ubuntu all day today, and now the HDDs seem to be working, I presume that I will start losing the drive once the system has been on for a while.
 
Who's got the cliffs notes? :D

Leading causes of HDD failures are power fluctuations (typically caused by an air conditioner or similar; some areas have bad power) and heat.
 
My HDDs worked fine again, since using Ubuntu for a day. Now the whole process has sped up, before it took a month before I had problems again, this time only 6 days. Also this time it went straight from working to blue screen of death to not being picked up all in one go.
 
Alot of reading... need cliffnotes for lazy ppl like me.

What hard-drives are these again?
 
Drive 1 - Seagate (returned)
Drive 2 - Seagate
Drive 3 - Maxtor (apparently the two drives, the Seagate and the Maxtor, are identical) this is the drive that has completely failed.
 
Well there's your problem!
Click to expand...

I have > 100 of these at work and they do not fail at a higher rate than any other manufacturer. This highest occurrence of failure is DOA or infant mortality and that I can attribute that mostly to poor packaging from newegg and others. Also since drives average 1/3 of the price they were 10 years ago testing at the factory is not as rigorous as it used to be.
 
^
You forgot that Seagate had the flaky firemware issue that cause worldwide panic when their hard-drives where bricking all over the place.

Perhaps that is what OP is on and needs to update the firmware.
 
When you have these drives installed in your case, how close in proximity are they from each other? Do you have active cooling to cool these drives when they are running?

Try a DIFFERENT PSU. Not that Thermaltake. If I were you, I'd try an Antec or a Corsair HX620. Make absolutely sure that you are using each drive on it's own power source from the PSU; do NOT use any 3rd party splitters or extensions. Also, try different SATA cables.

If none of these work, you might want to start looking at your controller, but I doubt that's what it is, since you had a different board. This is most likely, 99.8% going to be either a heat or a power issue. My bet is on heat.
 
I have my HDDs installed in the 5.25" slots, cooled by aftermarket heatsinks. Also I am not going to buy another PSU just in the hope that it is that, is there any other way I can test the PSU?
 
I'd second ajm's concern on power connectors. However, seeing so many of your HDs fail, any idea about the quality of power going into your PSU from the socket? I.e. are you behind a UPS or some form of AVR?
 
Dark Light said:
I have my HDDs installed in the 5.25" slots, cooled by aftermarket heatsinks. Also I am not going to buy another PSU just in the hope that it is that, is there any other way I can test the PSU?
Click to expand...

What are your "aftermarket heatsinks"?

Most if not all hard drives (newer ones at least) are going to require some sort of active cooling. Just because you install it into a 5.25" slot doesn't mean squat if there is no cooling going on to whisk away the heat. Your aftermarket heatsink can only do so much to transfer heat from the sides of the drives to the case. You will still need active cooling.

I find it hard to believe that you don't have another PSU to test. Right now you are in troubleshooting mode; you should be willing to do whatever it takes to find out what the problem is and how to solve it. I can guarantee you that whatever little you spend trying to figure out what's the problem is far cheaper then losing your data at the wrong time; take it from someone who's experienced it.

So yes, you are going to have to get some active cooling, and then you are going to have to try another PSU.

Last but not least. Have you tried running drive diagnostics on the drive (i.e. disk manufacturer diagnostics)? If so, what were the error codes that you received?

If the system is stable enough, try installing Speedfan. This will give you access to the SMART status of your drives. See if your drives are healthy enough. Also, please list the EXACT hard drive/drive controller errors you receive in your Event Viewer. This will help us to determine what's going on.
 
I am cooling my HDDs with a Akasa Hard Drive cooling kit, AK-HD03-BK, along with a Antec 900. Also speed fan does not give me any information on SMART, I have run drive diagnostics and they gave no errors.

Also while I was writing this Pc Probe came up with a warning saying that 3.3v was at 4.08v, not sure if this is just a misread or something more.

I ran Ubuntu for about an hour, and the drives seem fine for now. I forgot to add whenever I can't pick up the drives I can't boot into Ubuntu unless I remove the drives from the system.

I did get this error when the drive failed momentarily about 1 hour before I lost them today.

The device, \Device\Ide\iaStor0, did not respond within the timeout period.
Click to expand...

- System

- Provider

[ Name] iaStor

- EventID 9

[ Qualifiers] 49156

Level 2

Task 0

Keywords 0x80000000000000

- TimeCreated

[ SystemTime] 2009-10-13T13:42:56.332160300Z

EventRecordID 14053

Channel System

Computer ----------------

Security


- EventData

\Device\Ide\iaStor0
0F0005000100000000000000090004C00100000000000000000000000000000000000000000000000000000000


--------------------------------------------------------------------------------

Binary data:


In Words

0000: 0005000F 00000001 00000000 C0040009
0008: 00000001 00000000 00000000 00000000
0010: 00000000 00000000 00000000 00


In Bytes

0000: 0F 00 05 00 01 00 00 00 ........
0008: 00 00 00 00 09 00 04 C0 .......À
0010: 01 00 00 00 00 00 00 00 ........
0018: 00 00 00 00 00 00 00 00 ........
0020: 00 00 00 00 00 00 00 00 ........
0028: 00 00 00 00 00 .....
Click to expand...
 
If the hard drives are DEAD (i.e. they don't work in another computer), it's most likely your PSU. This isn't really something you can test in the short term, but I'd try replacing it and see if it helps.
 
The strange thing is, they do work, after disconnecting them and running Ubuntu, they work fine for a stort while.
 
Ubuntu works differently than Windows.

If they work fine for only a short while... then they aren't working right at all.

What exact hdds do you have? Like model #?
 
Dark Light said:
I am cooling my HDDs with a Akasa Hard Drive cooling kit, AK-HD03-BK, along with a Antec 900. Also speed fan does not give me any information on SMART, I have run drive diagnostics and they gave no errors.
Click to expand...

Please provide the name of the diagnostics you ran.

Dark Light said:
Also while I was writing this Pc Probe came up with a warning saying that 3.3v was at 4.08v, not sure if this is just a misread or something more.
Click to expand...

PC Probe or whatever else motherboard manufacturer utility are not renowned for their stability or reporting. Get rid of it. Use Speedfan, Everest, or another program to monitor your temps and voltages.

Dark Light said:
I ran Ubuntu for about an hour, and the drives seem fine for now. I forgot to add whenever I can't pick up the drives I can't boot into Ubuntu unless I remove the drives from the system.

I did get this error when the drive failed momentarily about 1 hour before I lost them today.
Click to expand...

I can't really respond to this until you tell us the name of the diagnostics you ran, and what type of diagnostic. If you run the manufacturer's diagnostic 24-48 hours (NOT a quick pass test) and the drives pass, it's suffice to say it's not a hardware problem and that you should start looking at other things like RAM, motherboard, CPU and whatnot.
 
My first thoughts would be a bad PSU or mainboard. Power issues can be really hard to pin down other than by testing the PSU properly (DMM and oscilloscope) or swapping it with a known good one.
 
You must log in or register to reply here.
Back
Top