• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Motherboard/RAID trashing my HDs?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

typhoonmike

Member
Joined
Aug 17, 2001
System specs:

Abit KD7-RAID Bios DN
HPT372 Raid Controller(onboard) with Bios 2.32
2 Maxtor 30g 7200rpm ATA100 HDs partitioned into 4 partitions and setup in RAID 0 array. Each on their own cable and both set as Master. Both Hard drives were Diamond Plus 60
Athlon XP 2100+ not overclocked(was OC'd 3+ months ago with no problems but summer time and too warm in room)
Windows 98SE

Actually made a post about the exact same problem previously but I'll restate since this has now happened twice.

About a month ago it seemed that one of my hard drives in my RAID 0 array crapped out on me. Computer was working perfectly fine one evening. Turn on computer the next day and it wouldn't boot into Windows. I could see the information on the drives and was even able to take a file off the drive that I needed. Ran Maxtor's Powermax program while the hard drives were still hooked up to the RAID controller and in the RAID 0 array. The program was able to see both the hard drives and apparently test them each. One hard drive passed all the tests while the other hard drive failed the SMART attribute check and SMART self-test. Wondering if perhaps the RAID setup might be giving me false results, I disconnected them from the RAID array and hooked them up to the normal IDE connectors and ran the tests again. Exact same results. One hard drive showed perfectly fine and other gave the SMART test failures.

RMA'd the apparent bad drive and was sent a refurbished 30g 7200rpm ATA133 drive. This one was a DiamondMax Plus 8.

Tested the replacement drive with Powermax and it showed no problems. Hooked both drives up to the RAID controllers in a RAID 0 array. Tried to test them again but this time the Powermax program couldn't see them. I had flashed the MB Bios to the most recent version but no change was made to the RAID controller version.

Partitioned the RAID array into 5 partitions and installed Windows 98SE on the C: partition. Was testing things out on it last night just installing drivers and whatnot. Came home today and was going to reformat the C: partition and reinstall Windows but I couldn't reformat. Was told the drive had an error on it and the disk was unusable. Tried to run scandisk and was told there was an unfixable error in Sector 8036(something like that). Unpartitioned the drives and took them out of the RAID array and hooked them back up to the regular IDE connectors. Ran Powermax on them both.

Orginal hard drive that tested fine previously and was not RMA'd was suddenly showing the exact same errors as the drive I had just RMA'd. SMART attribute check failure and SMART self test failure.

Replacement drive that tested fine when I got it passed the SMART attribute check but failed the SMART self test.

I think Maxtor is going to question if I try to RMA 2 drives at the same time especially considering one is a replacement for a drive I RMA'd less then 2 weeks previously. I just don't know if I just happen to be getting bad drives or if somehow the motherboard/RAID controller is trashing them. The hard drives were used in my Abit BX133-RAID in a RAID 0 array for quite some time with no apparently trouble. When I upgraded to the XP 2100+, I purchased the Abit KD7-RAID board for it brand new. It was purchased in December of last year so it's not very old. The hard drives were running in it since.

Any help or ideas would be appreciated.


<EDIT> This is kind of weird. The replacement HD that was failing one of the tests, suddenly passed the test after I repartitioned and reformatted it(out of the RAID 0 array). I tried the same on the older hard drive but it still fails.

I found a S.M.A.R.T test utility that gives me all types of information and it shows that the older hard drive has a life critical problem. Something about how many times it has to try and spin up the platters. Showed a few other problems also but it was the Spin retries that were critical. The replacement drive shows fine though. I'm guessing one drive is bad but I'm worried the problem is being caused by my MB/RAID controller.

Another item of interest is I found a patch on the HP site that had to do with S.M.A.R.T monitoring software that apparently came with some of their computers. It was to fix a problem with the computer giving a warning of Maxtor HD failure for attribute 10 which just happens to be the attribute that is showing bad for me. Now this is a homebuilt system so the patch does nothing for me but makes me wonder if I truely have a problem or if maybe the patch provided to those HP owners was just to cover up a potential problem with their drives.
 
Last edited:
There was another member, I don't remember who, that was running past 200MHz/400FSB like me that managed to kill one of his IBMs on the RAID0 Array but he was at 217MHz when it failed & I was only up to 207MHz. I couldn't run stable enough at that BUS speed so I ran at 200MHz for a few months, on a RAID0 Array as well with 2 X 20G 'cuda IIIs, & started to get the data corruption but only after a few months. Those drives are toast now but have had them since 2001 when I got my first 'puter. I had the same problems of not being able to format even after clearing the Array & rebuilding it a few times. I did a low level format & was able to reinstall but after a day the OS was dead again.

My position on this is that I tortured my drives at 40MHz PCI for extended periods of time & that was what killed them & so with the other guy. I ran with KD7_B8.BIN with the HP2.32 during these times but the failure of both drives was with the KD7_DN.BIN HP2.34. Currently I have a pair of 7200.7 ST340014A Seagates on RAID0 & a 7200.7 ST380011A Seagate for backup on the Secondary IDE Controller setup as slave. I'm also running at 13 X 166MHz with the 5:2:1 until I have sufficient funds to run at 200MHz again if ever I have to replace my drives..... again.
 
Hmm...perhaps they slowly failed after my overclock then. The computer was overclocked from probably January until March or April. Only overclocked to something like 176 or maybe it was 183FSB. Can't recall exactly.

I'd heard of people complaining about HD corruption when they were running above 200FSB but I would kind of expect something like that if I had been running that high. Course I guess I can't blame Maxtor either for bad drives since my OC while not very high still put the HD out of spec.

Well thanks for the reply. I'm just really afraid of them dying even if the computer was never OC'd. Don't want it to be the HPT372 or something that's causing the problem.
 
One thing is to not judge this anomaly with only a few instances that point the other direction. I have had 2 instances lately with my new drives that required a einstall because of a failure to read the array. I'm not a 100% sure yet what is causing it but I'm suspecting the HP 2.34 BIOS.
 
I'm about to go crazy now.

The first replacement drive I got is showing signs of dying according to SMARTUDM. Maxtor Powerwhateveritscalled doesn't show a problem but when I run SMARTUDM it says I have about a month left of life on this particular drive. It was replaced less then a month ago and only been turned on about 30 times.

When I received this replacement drive, I made sure to test it out. According to the Maxtor program, it was perfectly fine. After setting it up in my RAID 0 array though suddenly developed a problem according to the Maxtor program. A quick reformat seemed to fix the problem or at least get rid of the error in the Maxtor program. I then tested the drive with SMARTUDM and it was fine. Left the HD alone for about 4 days while waiting for another replacement drive. When it arrived I redid the cables in my case and then installed both new drives. Before setting them up, I connected each to an normal IDE and tested them out. This is where SMARTUDM suddenly showed a problem on the drive that it had previously said was fine.

I find it very hard to believe that in less then 1 month I could have 3 hard drives suddenly go bad on me on their own. I can't keep sending drives back to Maxtor nor do I want to if it's not their fault. Don't know if I could actually RMA the motherboard especially since I can't find any hard proof that it's at fault.
 
Flash to an older BIOS for now, the DN with whatever HP BIOS seems to be the most likely suspect. Does the Maxtor Utility work properly on a RAID Array or even with drives going thru it in non-raid? With SeaTools it fails to see the drives when striped or not if its on the HP Controller, DiskWizard can see the stripe or as single drives on the HP Controller but will fail to diagnose them.

Low Level Format them on just the VIA IDE then stripe then again after you flash to another BIOS see what you get then. I'm grasping at straws here as well:(
 
When the first hard drive went bad, I was using the CS bios and the Maxtor program was able to see and diagnose the hard drives while on the HPT chip. After going to the DN bios, the Maxtor program couldn't see the drives connected to the HPT chip but I don't know if it was the bios flash that did that.

SMARTUDM can see and test drives connected to the HPT and in RAID 0.

I use the Maxtor program to test the drives while they are connected to the regular IDE and neither hard drives show a problem. Using SMARTUDM, it shows a problem on the first replaced drive even while it's connected to the regular IDE. It wasn't even connected to the HPT chip or in RAID 0 for more then a day so it seems odd to think that that could cause damage or problems. Especially since the problem is Spin Up Time which seems to mean its taking longer for the drive to reach it's rpm then it should.

One thing I have read a few times is that some Maxtor drives were/are known to go through a breaking in period in which their Spin Up Time goes all goofy before eventually getting normalized. I don't know if that is an existing issue or perhaps only for older drives. I'm going to just keep testing the drive each day and see if it gets better.

<EDIT> It just keeps getting better. Had the HDs setup in RAID 0 and was in the process of formatting the partitions. While formatting the last partition, my monitor flashed and suddenly half of my screen is filled up with a green screen and the other half with flashing letters,images,etc. Turned power off and made sure vid card was seated properly. Turned back on and started to reformat the last partition again and now it's hung up at 15%.

I thought computers made your life easier. BAH!
 
Well now I can't seem to format one of my partitions past 15%.

I dropped down to the CY bios instead of the DN. I did a low level format on one of the drives but didn't want to wait around for another 5+hrs to do a low level on the 2nd drive. All I did was hook them up to the normal IDE, partition and format them with no trouble.

Removed the partitions and hooked the drives back up to the Raid Controller and put them in RAID 0. Partitioned into 5 seperate partitions. I can format the C,D,E,G partitions but everytime I try to format the F partition it just hangs up at 15%. The F partition is the largest partition at around 29k Mbytes.

One thing weird I've noticed is that the Raid Controller shows the size of the array as 61.5GB. FDISK shows the size as something like 59k Mbytes or 57.5GB. I wonder if that could be the reason why it's having problems.
 
I'm at a complete loss. If I set the hard drives up in RAID 0 but just make one partition then I can format the entire array. It's only when I partition it does it give me a problem on one partition everytime.

No clue what to do now since I want those partitions in there.

<EDIT> Well changed the size of a few partitions and everything suddenly worked. The F partition was normally 29000Mb but set it as 28000Mb this time and used up the extra space elsewhere. Was able to format it. Now just hoping Windows installs alright and doesn't crap out on me later.

<EDIT>Blah. Changed partition sizes again and back in the same boat. Only it's a different partition that is screwing up. About the only thing that appears constant is the fact that it's happening near the end of the drive. I'd say it was a bad part of the drive but this problem only happens in the RAID 0 array and not when the drives are seperate.
 
Last edited:
Well I've decided it was my motherboard or at least the RAID chip on it.

Hooked the 2 drives up to my old BX133-RAID motherboard and set them up in a RAID 0 array. First thing I noticed is that it was reporting about 58Gb as the size of the array which was the same as what FDISK was saying. The troublesome KD7-RAID was reporting about 61.5Gb. I partitioned the drives into the exact same size partitions as the ones I was having trouble with earlier. Was able to format them with no trouble at all. Didn't hang up on the F: or G: partition like it was on the KD7-RAID.

Not sure if that is basis enough for an RMA but I'm hoping. Unfortunately I don't think I'm going to have much luck with it. The place I bought it from wasn't a big name online dealer like Googlegear or Newegg. If I get a response at all from them then most likely I'll send them my motherboard only to have them say they couldn't find a problem or just to never see it again.

Hate to think I'm going to have to buy another motherboard but perhaps it's time to start looking.
 
You can RMA to ABIT & not get charged a fee if you have the original reciept of purchase.

I have fixed my problem so it seems because I have not had a cold boot nor have I seen anything out of the ordinary with my array. What seems to have fixed it is my return to the KD7_C8.BIN but I'm curious if the stock KD7_DN.BIN will pose any problems. Next week I'll flash to it.
 
Back