Hard drive failure??

I just bought two of these hard drives (LN15743) and about a day after using my PC without problems (just been doing some downloading), I've received the following error message:

http://img364.imageshack.us/img364/7771/erroryr7.th.jpg

They were both wrapped in proper stuff, should I rma it or is it just falsely reporting SMART failure? Is there anyway I can check? Using Intel ICH7R Raid controller, in Raid0. I'm at a loss as to how it could fail so quickly, cos ebuyer have sent me DOA hard drives before that other people have returned as faulty (woopee), but new ones pretty much always work. Don't want to end up killing loads of hard drives, but I've got a 'safe' set up, theres a big deep 120mm fan sucking air through them, and I haven't been absolutely churning the hard drives.

:( How should I be sure it's dying?

Sytem Report from Intel Matrix Storage Manager
System Information

Kit Installed: 6.2.1.1002
Kit Install History: 6.2.1.1002
Shell Version: 6.2.1.1002

OS Name: Microsoft® Windows Vista™ Home Premium
OS Version: 6.0.6000 Build 6000
System Name: RICHARD-PC
System Manufacturer: ASUSTeK Computer INC.
System Model: P5W DH Deluxe
Processor: Intel(R) Core(TM)2 CPU 6300 @ 1.86GHz
BIOS Version/Date: American Megatrends Inc. 1901 , 01/24/2007

Language: ENU



Intel(R) RAID Technology

Intel RAID Controller: Intel(R) 82801GR/GH SATA RAID Controller
Number of Serial ATA ports: 4

RAID Option ROM Version: 5.1.2.1002
Driver Version: 6.2.1.1002
RAID Plug-In Version: 6.2.1.1002
Language Resource Version of the RAID Plug-In: 6.2.1.1002
Create Volume Wizard Version: 6.2.1.1002
Language Resource Version of the Create Volume Wizard: 6.2.1.1002
Create Volume from Existing Hard Drive Wizard Version: 6.2.1.1002
Language Resource Version of the Create Volume from Existing Hard Drive Wizard: 6.2.1.1002
Modify Volume Wizard Version: 6.2.1.1002
Language Resource Version of the Modify Volume Wizard: 6.2.1.1002
Delete Volume Wizard Version: 6.2.1.1002
Language Resource Version of the Delete Volume Wizard: 6.2.1.1002
ISDI Library Version: 6.2.1.1002
Event Monitor User Notification Tool Version: 6.2.1.1005
Language Resource Version of the Event Monitor User Notification Tool: 6.2.1.1005
Event Monitor Version: 6.2.1.1005

Array_0000
Status: No active migration(s)
Hard Drive Write Cache Enabled: Yes
Size: 306.7 GB
Free Space: 0 GB
Number of Hard Drives: 2
Hard Drive Member 1: WDC WD1600YS-01SHB0
Hard Drive Member 2: WDC WD1600YS-01SHB0
Number of Volumes: 1
Volume Member 1: Operating

Array_0001
Status: No active migration(s)
Hard Drive Write Cache Enabled: Yes
Size: 465.7 GB
Free Space: 0 GB
Number of Hard Drives: 2
Hard Drive Member 1: Hitachi HDT725025VLA380
Hard Drive Member 2: Hitachi HDT725025VLA380
Number of Volumes: 1
Volume Member 1: Data

Operating
Status: Normal
System Volume: Yes
Volume Write-Back Cache Enabled: No
RAID Level: RAID 0 (striping)
Strip Size: 128 KB
Size: 306.7 GB
Number of Hard Drives: 2
Hard Drive Member 1: WDC WD1600YS-01SHB0
Hard Drive Member 2: WDC WD1600YS-01SHB0
Parent Array: Array_0000

Data
Status: Normal
System Volume: No
Volume Write-Back Cache Enabled: No
RAID Level: RAID 0 (striping)
Strip Size: 128 KB
Size: 465.7 GB
Number of Hard Drives: 2
Hard Drive Member 1: Hitachi HDT725025VLA380
Hard Drive Member 2: Hitachi HDT725025VLA380
Parent Array: Array_0001

Hard Drive 0
Usage: Array member
Status: SMART event
Device Port: 0
Device Port Location: Internal
Current Serial ATA Transfer Mode: Generation 2
Model: WDC WD1600YS-01SHB0
Serial Number: WD-WCAP01396971
Firmware: 20.06C03
Native Command Queuing Support: Yes
Hard Drive Write Cache Enabled: Yes
Size: 153.3 GB
Number of Volumes: 1
Volume Member 1: Operating
Parent Array: Array_0000

Hard Drive 1
Usage: Array member
Status: Normal
Device Port: 1
Device Port Location: Internal
Current Serial ATA Transfer Mode: Generation 2
Model: Hitachi HDT725025VLA380
Serial Number: VFA100R10066ZB
Firmware: V5DOA52A
Native Command Queuing Support: Yes
Hard Drive Write Cache Enabled: Yes
Size: 232.8 GB
Number of Volumes: 1
Volume Member 1: Data
Parent Array: Array_0001

Hard Drive 2
Usage: Array member
Status: Normal
Device Port: 2
Device Port Location: Internal
Current Serial ATA Transfer Mode: Generation 2
Model: Hitachi HDT725025VLA380
Serial Number: VFA100R1CHNWXB
Firmware: V5DOA52A
Native Command Queuing Support: Yes
Hard Drive Write Cache Enabled: Yes
Size: 232.8 GB
Number of Volumes: 1
Volume Member 1: Data
Parent Array: Array_0001

Hard Drive 3
Usage: Array member
Status: Normal
Device Port: 3
Device Port Location: Internal
Current Serial ATA Transfer Mode: Generation 2
Model: WDC WD1600YS-01SHB0
Serial Number: WD-WCAP01608721
Firmware: 20.06C03
Native Command Queuing Support: Yes
Hard Drive Write Cache Enabled: Yes
Size: 153.3 GB
Number of Volumes: 1
Volume Member 1: Operating
Parent Array: Array_0000

Pop to the Western Digital website and download their hard-drive analysis and testing tool.

This would then show if the drives faulty or has errors which we can then advise the best course of action to take.

Best Regards,

Hmm, I couldn't run the diagnostics tool, because WD diagnosis tool saw the array as just one single drive (and then said the cable wasnt plugged in). Not much use there. I'll see what else they've got on their website.

eta: there's a firmware update on WD website for my HDD. It's for RAID arrays, too.

Quote:

---

This firmware resolves an issue where a WD1600YS, WD2500YS, WD4000YS, or WD5000YS hard drive is dropped from a RAID set without reporting errors after a period of normal usage. This utility is designed to upgrade the firmware of the following hard drives:

---

My current version: 20.06C03
Updated version: 20.06C06

*ponders* I'm scared of powering down my PC incase it doesn't start up again. I've got no important things I can't afford to lose on this array, just the OS, and I don't have any live CDs. Guess if it failed, I could dissassemble the array and use the working hdd. So much faff though.

Quote:

---

Originally Posted by Dreaming

Hmm, I couldn't run the diagnostics tool, because WD diagnosis tool saw the array as just one single drive (and then said the cable wasnt plugged in). Not much use there. I'll see what else they've got on their website.

---

No problem, most of these hard disk diagnostic tools have difficulty seeing through a RAID array.

All you have to do is temporarily disable RAID in the BIOS of your motherboard. Scan the drive with the diagnostic tool, and then re-enable RAID before restarting your OS. You won't lose the array as a result of temporarily disabling it like this.

However, as far as the risk of data loss is concerned, if the drive concerned is on a RAID 0 array, you're gonna lose the data if the drive does need replacement anyway. As such, well worth thinking ahead to your method of backup now.

Ok, so put it on a floppy ms-dos bootable disk?

Just tried active SMART which is supposed to the be the dogs, but that two just read the drives as japanese symbols :p.

Quote:

---

Originally Posted by Dreaming

Just tried active SMART which is supposed to the be the dogs, but that two just read the drives as japanese symbols :p.

---

Don't bother with any third party tools, there's more chance for them to confuse the issue than there is for them to help. The WD diagnostic is quite comprehensive and, alongside the Matrix Storage Manager inside the OS, will tell you everything you need to know.

It could well be that the SMART event is not a critical one, in which case there's no desperate need to worry. However, it is equally possible that the drive is defective. You need the WD tool to tell you that, and the way to get it working is to temporarily disable RAID.

I'm just checking, I've got my UBCD which has 'a' version of the wd tools, might be worth downloading the latest version tho. It normally comes with this kind of stuff.

Yep definitely worth adding the latest version of the diagnostic.

As far as reading SMART data is concerned, about the best reader I've seen is the one included in SpeedFan - although again this won't see through RAID arrays so it doesn't help your current situation that much. WD tool is the thing to use.

Ok, on doing a 'quick' test, it interrupts about a second in and says that the hard is not good call cs etc one or more things is below threshold or worst case, I look down the list and it appears to be:

5 Re-allocated sector count

value: 107
threshold: 140
worst case: 107

Whereas from what I can see, most of them are way above threshold

(raw read error rate is 196, worst is 196, threshold is 51)

Not sure how to interpret the results, just wrote them down on a notepad lol.

Going to get a chinese now and turn my computer off and unplug it for the night. Last I saw of it, it was trying to boot into Windows and was taking well over a minute on the black / green microsoft loading screen.

Doesn't look good though :(

get a new hdd, stop faffing about :p

Yea, think I will, it stopped working altogether pretty much. Did a little google, and it appears the error I'm having, is when there are lots of bad sectors. For every bad sector, it reallocates a reserve sector. When the amount of reserve sectors gets low (below 140 in this case) it indicates the drive is close to failing. Seeing as there is a maximum value of 253, to have a two day old drive down to 107 probably means the actual drive is faulty in some way.

Just keeping you all informed, will raise an RMA now. :)

Ok guys. Just put the replacement one in (two weeks after receiving it - been busy here!) and it worked, started installing windows, went to have a bath, got out and the new hard drive has a smart event too. Check it out, exactly the same - raw read error rate value was 128 this time.

So another RMA or is something else to blame here? Just can't believe two hard drives from the same people both fail on the same thing in the same circumstances in the same computer in the same set up after the same amount of time on the same sata channel... whereas two others I have work fine.

I'll just raise another RMA then..

Are you sure you're RAID controller is working properly? To corrupt that many sectors that quickly there's gotta be something wrong.

It certainly won't help that you've got a PSU of Q-tec standards... that's got to be switching the HDDs for fun....

Yea, thats the thing. I made a thread a while ago asking people whether I should upgrade my psu but nobody responded.

Bear in mind I have 3 hard drives in my system and they're all completely fine. My raid controller is an integrated ich7r, both failed hard drives were on port 0. *shrug*

Nother thing about the power - the broken hard drive and working hard drive were both running off the same molex wire. *shrug*

I bought the PSU off scan a few months ago, I wouldn't expect it to kill hard drives as soon as I put one in.

eta: also bear in mind the two other drives are also in a raid0 array without so much as a hint of a problem.

So just wondering, is it most likely the PSU at fault or is that speculation?

I'm speculating, but it's certainly not helping matters. I'll put it another way - if I were to have a system with that PSU, I wouldn't be loading it up with 3 hard drives.

Ah.

Could you advise me as to the best upgrade path. It seems one of my RAM things may be faulty now too, argh! I'm a grand in debt, between jobs and can't afford this right now :p lol (fortunately 0% interest).

What's a good 'solid' reliable no thrills PSU? Prefer quieter ones as my system is pretty quiet all round. Have a P180 chassis as well so a long cable to reach the 24ppin on the motherboard is a plus. Thanks gav

First of all, check the memory separately - give Memtest86 a go and let it run a couple of passes.

For that system, a the 430w Seasonic should be adequate. I've no idea if the cable will stretch (I've a 'standard' layout - might be best ordering the extension (1) (2)
from Scan too while you're at it. Certainly give it a Google - the chances are someone else somewhere will have the combination, as they're both pretty popular among enthusiasts.

Please remember though that I'm not saying the hard drive failures are being caused by the PSU. I'm just saying I wouldn't even be tempted to run it. Secondly, if you're starting to get more issues (looking at the RAM), then perhaps the PSU really is struggling and is putting out some dirty rippling power, unable to sustain what's being asked of it (cheap PSUs are almost always measured by peak power rather than what they can sustain) - that can cause issues, especially hard drives, them being mechanical, if they're constantly switching. Thirdly, a new PSU might not bring back the hard drive to the way it should be. If it's been damaged, it's going to stay damaged. If it is just constantly switching on and off though (which will certainly be audible), then chances are it'll still work. Fourthly, I really wouldn't be overclocking with that PSU ;)

If at all possible try to borrow some parts from a mate - a chunkier (brand rather than wattage!) PSU would be perfect to check out everything - failing that, an external hard drive enclosure for the 3rd hard drive (they have their own power supply, so take the load off one of yours).

Think I'm done now. :P

Well I use 4 drives in my optimal set up, which is going to go up to 5! So really do want a fair bit of watts to play with.

There is no mechanical failure with the hard drive - I've got one of the same model that works fine. When I put it into a raid array with the other one scan sent me though, it threw up a smart error after a day, saying that it had used up it's reserve sectors when i checked it (basically, every time it ****s up and write a bad sector, it uses a reserve sector. when these get low you get a s.m.a.r.t event warning saying it's running out). I got a replacement, but the exact same thing happened with that. Weird thing being, the first hard drive works fine.

Sorted an RMA for the second one, but certainly perplexed. It could be the power - if it's not giving it a steady stream of power I can imagine it could mess up writing a sector meaning it's no longer usable, and if it's a continuous issue it could keep happening.

At the moment I'm taking a big risk as I've two t7k500s in raid0 to store stuff on. This is going to change into a raid5 in the future, but as my onboard controller has 4 ports, the hitachi drives will also need a seperate raid controller.

Going to give memtest86+ a go, just need to work out how it works :p. Burn .iso and boot to it I think? *shrug*

Thanks for your advice. I'd never considered a PSU an integral part of the system before but am beginning to see why most enthusiasts see it as the foundation stone of every good system.

And I may have found a job. Good day for me today!

Quote:

---

Originally Posted by Dreaming

Going to give memtest86+ a go, just need to work out how it works :p. Burn .iso and boot to it I think? *shrug*

---

Exactly - when it starts up, it will do everything without any input. :)

It's not the memory :p. Is related to higher fsbs, but who knows... got another thread going in hexus.help for that anyway.

So sent the second faulty one off yesterday, and now the first one I got is reporting a smart event. Ive ordered a new psu, and a UPS as well. When I get them I'll check the motherboard though, because setting the fsb to 400mhz causes blue screens, so it's not very happy.

Really confused as to what counts as RMA-able though. Becuase I bought the motherboard, power supply and hard drives from scan (the RAM is ok). So do I do another RMA and send the motherboard back? But I can get it to hobble along, it just seems to be killing hard drives :/. So maybe it's the PSU that's faulty (well, I'm replacing that anyway to rule it out) - do I RMA that?

It's £20 a time if it's not faulty on it's own, which I don't think any of them are. Frustration!

Or if there's a good simple c2d board with wifi that I can buy it might be easier than chasing ghosts. Really disappointing though, can't imagine what caused it.

RMA everything or what?

Quote:

---

Originally Posted by Dreaming

At the moment I'm taking a big risk as I've two t7k500s in raid0 to store stuff on.

---

Why are you taking a big risk?

Just for the novelty. There's nothing *essential* on there, just like game patches. Only using a tiny bit of it. But I realise if one goes then both of them do. Was more of a trial run :p.

I had a brainwave. The three hard drives I've had were all on 4 in molex connections. The two working ones are on sata style connections. Not sure why that makes a difference, but maybe I can rule out the motherboard.

The reason it can't set the fsb higher is because it doesn't have the clean voltage perhaps. The reason it's writing bad sectors to hard drives, is because it doesn't have clean power... My new psu (newpower, would have got it from scan but they didn't have one) should be healthy and eco friendly too :p

Not sure if the PSU counts for an RMA because it's still *working* in that my system is still using it at the moment. Later on today will take it out this system to revent it damaging anything though. According to Pc wizard there are 15 volts on the +12 rail (probably wrong mind), and the others are out of specification by up to .5 of a volt too.

Far be it from me to preach, but if it does this after a few months then should scan be selling this kind of psu? I know you get what you pay for and all... but I thought the fact that scan's buyers had gone for it would be some small form of garuntee that the product wouldn't be utter cack (as appossed to ebuyer for example, where it's just rebranded refurbished melted stuff).