Dying Hard Drive?

[Garfield3d]

Hi everyone. I appreciate any help that you guys might have, even outlandish theories.

System Specs:
P4 2.8Ghz "Northwood"
Asus P4P800 "Springdale"
Creative Audigy 2
eVGA Geforce 7800 GS 256MB
1 GB Corsair DDR1-PC3200
Antec 380W TruePower (Sonata I case)
Windows Vista
300 GB 7200.8 refurbished HD

Symptoms:
-- Periodically, Windows Vista will take nearly 2-3 minutes to start up (normally only takes about 30 seconds).
-- The computer will hang (mouse and keyboard are unresponsive, hard freeze); have to reboot; everytime this happens, the HD light is on.
-- Windows ScanDisc has detected several bad/corrupt sectors in one recent scan
-- In Event Viewer, after one crash, I was given a long list of errors (32 in total) where the Source was "volsnap" and the Event ID was "57", with the explanation: "The system failed to flush data to the transaction log. Corruption may occur."
-- In Event Viewer, every crash is listed ambiguously as "The previous system shutdown at 5:36:50 PM on 6/12/2007 was unexpected." with the Source as "EventLog" and the EventID is "6008."

Basically, I wanted to ask you guys to confirm (or contend) if this is a faulty hard drive. Replacing a hard drive is a big hassle, so I don't want to do something that I don't have to. What is particularly unusual, though, is that I received this hard drive as a replacement drive from Seagate. About 2-3 weeks ago, my original hard drive (that lasted about 8 months) experienced many of the symptoms that I just listed. I sent that HD in for replacement, and received my current drive now. It worked fine for about one day, but then it started experiencing the symptoms I listed above. I've only had the hard drive for about 5 days now.

It seems odd that the HD died again. I'm wondering if maybe it could be my motherboard, or some other problem. However, if the motherboard or southbridge was damaged, I would have expected my computer to be bad for every second it was turned on. Instead, these system crashes come and go every few hours, and the slow HD access comes in bursts (slow for about 2 minutes, like it's reading a bad area, then fine for an hour or maybe 10 minutes).

Do you think it's just bad luck and the HD died again? Or is something else happening?

Thanks for any clues that you guys can provide.

--Garfield3d

 

[XxPrOdiGyxX]

If chkdsk is telling you that there are bad sectors then it's probably the hard drive again. Go to the website of the hard drive manufacturer and download their diagnostic utility. Run that. It should give you a more definitive answer if you aren't sure.

Edit: That is if you have a second computer that you can run diagnostics with.

 

[cprince]

WAIT!!! Don't run the manufacturer diagnostic software until you have all your data safely back up. The diagnostic software stresses the hard drive, and the drive may fail during the test with no cheap way for you to recover data.

 

[Decoy84]

Here's a good tool for hard drives http://www.hdtune.com/ but it won't try to repair bad sectors you'll have to get the manufacturer software for that.

 

[XxPrOdiGyxX]

Originally posted by: cprince
WAIT!!! Don't run the manufacturer diagnostic software until you have all your data safely back up. The diagnostic software stresses the hard drive, and the drive may fail during the test with no cheap way for you to recover data.

I assumed since it had only been five days since he reinstalled everything that there would be no data worthwhile that would require a backup.

 

[Garfield3d]

Thanks for the response guys. I actually have a bit of data on the drive right now. I asked for Advanced Replacement on the first hard drive, and so I moved all the data from that first drive to the current drive (after a clean install of Vista).

I've actually been thinking... is it possible that the power supply is the problem? I went into the BIOS and I was watching the voltages, and the 12V line was fluctuating between 10.2V and 11.1V. The 3V and 5V lines seemed perfectly fine.

Atop of this, I replaced my old video card with a eVGA 7800GS (AGP 256MB). I assumed it wouldn't have caused any problems because I used a friend's BFG 7800 GS (AGP 256MB) card as well without running into any problems. Is it possible that the eVGA card is just more strenuous on the power supply, or that I just have a particularly bad card?

Although, I'm still kinda sketchy on that being a problem. There are still occasionally bad sectors, and it always crashes with the HD light on. Atop of that, a couple of crashes occur when everything else on screen will stop, and I can move my mouse for a scant few seconds until the mouse freezes up too (this happens a bit when I'm playing World of Warcraft).

--Garfield3d

 

[RebateMonger]

10.2 volts is pretty low for that 12V line. I had an Abit board that was only getting around that voltage and the built-in sound stopped working. The minimum allowable voltage under the ATX specification is around 11 volts. While there can be errors in the motherboard voltage sensors, if it's nearly that low, it's not a good sign.

 

[Garfield3d]

Originally posted by: RebateMonger
10.2 volts is pretty low for that 12V line. I had an Abit board that was only getting around that voltage and the built-in sound stopped working. The minimum allowable voltage under the ATX specification is around 11 volts. While there can be errors in the motherboard voltage sensors, if it's nearly that low, it's not a good sign.

Aside from buying a new power supply altogether, is there any way to alleviate the load or somewhat rectify the voltage on the 12V line? If I buy a weaker card, like a 7600GS or maybe a X1650 Pro, could that take some pressure off the 12V line?

--Garfield3d

 

[GAZZA]

I don't think getting less power consuming 12v devices will alleviate the problem, and 10.2 volts points to a PSU going bad IMO>
Better to just get a better quality PSU

 

[RebateMonger]

Originally posted by: GAZZA
I don't think getting less power consuming 12v devices will alleviate the problem, and 10.2 volts points to a PSU going bad IMO>
Better to just get a better quality PSU

Yeah. I'm not a member of the "only a 1000 Watt PSU will do" club. But I have found that a well-built, stable, power supply is pretty important. The Antec TruePower USED to be considered such a power supply (I own several of them), but Antec's capacitor problems have made the TruePowers a possible problem.

I believe you ARE having hard drive issues, but that low voltage makes it difficult to assign the cause to the PSU or to that refurbished hard drive.

 

[Muse]

The HD light being on, doesn't tell me that there's a problem with the HD. There may be, anyway. Yes, your PSU could be the problem, especially that 12v fluctuation seems very excessive. To me, the PSU in your system seems a weak link. Truepower 380 isn't that great. Were I you, I would use more than one HD. I have 4 HDs in my current main system and I boot to the smallest. If I suspected a problem with one HD, I'm in a position to test it, back up my data, etc. Don't believe my system rigs pages, they are antiquated. One day they'll fix Anandtech... Right now I'm running a Corsair 520HX PSU, a serious contender for bang-for-the-buck winner.

 

[xgsound]

The 12v line should not drop below 11.4v. This will corrupt data if you only consider the 12v driving the platters. 10.2v sounds bad if it's accurate.

One possible way to lower the stress on the 12v is to remove any nonOS hard drives or optical drives temporarily leaving only the OS drive. If the problems disappear, you need more power.

You could also try trading video cards with your friend, or revert to your old video card and see what the 12v does.


Jim