How do I find/fix a failing HD in a RAID array?

[cpjust]

Hi,
I'm not sure if this is the best place for this question, but the RAID forums hardly have any activity...

I'm running Vista x64 on an ASUS P5K3 Deluxe motherboard and I have 3 Seagate Barracuda SATA drives in a RAID 5 array.

It's been running fine for the last 1.5 years I've had it (except for 1 drive that died about half a year ago), but now all of a sudden my system is freezing at random intervals (about once every day or two). This is what happened before when I had a dying hard drive.

The Seagate SeaTools diagnostics are useless, since they don't find any problems with my drives (and they didn't even find a problem with my last drive that died), and since it hasn't died to the point of making noise or not working at all, I have no way of telling which of the 3 drives is the one causing the problem.

Does anyone know how I would go about tracking down this type of problem? Right now I'm guessing it's probably freezing because of excessively long read/write timeouts dropping the drive from the array... but that's just a guess. I see nothing in the Event Log about any RAID or drive problems.

 

[BadBigBen]

if you value your data, then I would do the following:

1. get at least TWO drives the same size as the others...

2. take out one of the HDDs that you had not changed yet, and replace the drive on it, then let the RAID rebuild itself...

3. repeat step 2 for the last OLD drive...

4. use the other two for storage (until they fail) or as door stoppers (your choice)...

Ben
"If it works don't fix it! If it doesn't use a sledgehammer..."
How to ask a question, when posting them to a professional forum.
Only ask questions with yes/no answers if you want "yes" or "no"

 

[linney]

If you want to check out other possible causes for freezing these XP threads may offer you a few clues as they mainly refer to hardware and other non-Windows causes.

Windows random hang
thread779-1423743

xp freeze troubleshooting
thread779-935320

PC Freezing - Hard to explain (see details)
thread779-1398265

 

[cpjust]

Thanks.
I've got 2 more of the exact same drives. I use one for external backups and the other as a spare in case one dies.

My data volume doesn't change that often. The most I'd lose is some E-mail, but I can start backing that up more regularly. This is my home PC, so I'd rather not spend more money than I absolutely need to.

I never had these kinds of problems when I was using a single non-RAID drive. I'm wondering if these Seagate Barracuda drives suck or if the RAID is shortening the life of the drives quite a bit?

I'd still like to find a way to determine which drive is causing the problem if it's possible.

But who knows. Maybe it's not even a drive problem at all. I'm just guessing, since the symptoms are similar to what happened the last time a drive died.

 

[BadBigBen]

I'm wondering if these Seagate Barracuda drives suck or if the RAID is shortening the life of the drives quite a bit?

there has been a problem with Seagate drives at the end of last year and the beginning of this year, but I do not know if your drives fall in that category, see for more info:

this talks about MAC's and Seagate failure, but you will get the gist:

Take two of the Seagate saga

http://www.tuaw.com/2009/01/23/take-two-of-the-seagate-saga/

and here the KB articles covering the drives:

http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207951

http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207963

Ben
"If it works don't fix it! If it doesn't use a sledgehammer..."
How to ask a question, when posting them to a professional forum.
Only ask questions with yes/no answers if you want "yes" or "no"

 

[cpjust]

If you want to check out other possible causes for freezing these XP threads may offer you a few clues as they mainly refer to hardware and other non-Windows causes.

Thanks, but the in first link the guy could still move his mouse after the 'freeze' and in the 3rd it was only pausing for a second or two at a time. Mine is completely frozen until I do a hard reboot.

there has been a problem with Seagate drives at the end of last year and the beginning of this year, but I do not know if your drives fall in that category...

Thanks, my hard drives are ST3500320AS at BIOS SD15 & AD14. It doesn't really mention what the problem with those BIOS revs are though.

The thing that confuses me is, the first time my hard drive died I had my OS volume in a RAID 0 configuration, so if one drive dropped out of the RAID array my system was obviously useless, but now I'm using RAID 5, so if one drive dropped out, I would expect it to hop along in degraded mode with the other 2 drives; and I would expect it to write an Event Log error about it being in degraded mode. I would find it highly unlikely that 2 drives drop out at exactly the same time all the time.
Could this mean that maybe it isn't really a hard drive problem?

 

[BadBigBen]

Have a read:

thread751-1525155
thread602-1524979

Ben
"If it works don't fix it! If it doesn't use a sledgehammer..."
How to ask a question, when posting them to a professional forum.
Only ask questions with yes/no answers if you want "yes" or "no"

 

[cpjust]

I think I figured out a way to find out which of the drives is bad.
I noticed that when I ran HD Tune to check the speed of my RAID 5 array, I was only getting about 37MB/s (but I was originally getting an average of 130MB/s from this chart). So I could take one drive at a time out and test it's speed on my other computer and see which one is slow... At least it'll give me something to do this weekend after I see Harry Potter. ;-)

 

[cpjust]

OK, I've run a thorough set of tests on all the drives and I can't see anything wrong with any of them, and their speeds are exactly what I would expect.

Is it possible for a loose cable to cause the slowness & lockups that I've been experiencing?

 

[BadBigBen]

Yes, that is a possibility, and a dying PSU could cause that as well...

Ben
"If it works don't fix it! If it doesn't use a sledgehammer..."
How to ask a question, when posting them to a professional forum.
Only ask questions with yes/no answers if you want "yes" or "no"

 

[gbaughma]

Normally, your RAID utility will tell you if a drive has failed. Additionally, if the drives support S.M.A.R.T., that should also tell you which drive is acting up.



Just my 2¢

"What the captain doesn't realize is that we've secretly replaced his Dilithium Crystals with new Folger's Crystals."

--Greg

http://parallel.tzo.com

 

[cpjust]

I don't know why it never occurred to me to try this, but based on the symptoms I was convinced it was a hardware problem. But a colleague told me to try reinstalling the RAID drivers before doing anything major; so I did, and my HD speed went from 50MB/s to 150MB/s and I haven't had any more lock ups since then!

I don't know how the drivers could have gotten corrupted, especially on a RAID 5 volume? Makes me wonder if anything else got corrupted also.

 

[firewolfrl]

I would look into ditching the seagate drives...I have had nothing but issues with that brand for raid systems....I have replaced about 9 drives in a year in 3 servers...the one server that runs western digital has been good for 2 1/2 years without any issues at all ...the original failures were all the 500 gig drives

To note: the 750 seagates were the replacements and have been stable

seagates are great drives without a raid system

 

[cpjust]

I'm not running a server, so maybe I won't have as many problems as you. So far I've only had 1 drive die.

 

[BadBigBen]

cp - it would not have dawned on me to do so either, reinstalling the drivers that is... glad that you got it...

and I can only confirm firewolfrl statement, about having issues with Seagates drives (though 1TB ones) conking out, and these where not in a RAID nor Server...

Ben
"If it works don't fix it! If it doesn't use a sledgehammer..."
How to ask a question, when posting them to a professional forum.
Only ask questions with yes/no answers if you want "yes" or "no"