Tom's Guide | Tom's Hardware | Tom's Games
 |
 |
 |
This is now the second hard drive of a 6 drive system to come up corrupted. First drive was the slave on the first channel of an Asus P4P800-E Deluxe motherboard, and suddenly just stopped responding. Upon rebooting, chkdsk popped up, and, in the third part of the process, went into a scrolling of what seemed to be every file on the drive (as in went on for minutes), THEN launched into scrolling random characters that literally filled the screen. This went on seemingly wothout end til I couldn't take it anymore and did a hard shutdown. I wound up formatting that drive.
This second drive, running off a Promise controller card was being defragged with Perfectdisk when I got an error message that told me one of the directories was unreadable and to run chkdsk. It's still running as I write this.
After the first drive folded, I checked the IDE cables, which are new and were definitely seated, but I still checked for continuity and shorts with a multimeter. No problem there. Drivers are all up to date. Scanned for viruses, nothing coming up. Checked the voltages on all the rails of the Antec 550W PSU - good there too.
I'm relatively new to using the NTFS file system, having held onto FAT32 since forever, only "upgrading" at the insistence (beratement?) of one of my PC gurus and because I'd read that larger drives don't handle FAT32 too well. Is there something to loading up a drive with a bunch of files all at once that would cause this mess? I'd never seen this much corruption with FAT32. I know even a new drive can be a lemon, but the drives in question are a brand new 750 gig Seagate and a less-than-a-year-old 400 gig Nitachi and this corruption has now hit me twice in four days.
I'll probably have more to report on this after this chkdsk gets done. It's only at 5% after 40 minutes and I really want to get a jump on finding out what I can do to STOP THIS! Anyone have any ideas of what else to look at here?I love Humanity - It's the people I can't stand
Nope - not overclocking. Despite the assurrances I got here, didn't like the temps on the Prescott so I'm keeping it at 3.0
To add to the frustration, the second chkdsk finished without indicating the errors that kept Perfectdisk from finishing. Tried running the defrag again and it still says there's a directory (not specified) that's unreadable and it's telling me to run the chkdsk again.
I love Humanity - It's the people I can't stand
0 
Drives were formatted in Disk Management, in Administrative Tools. Didn't do a quick format - suffered through a complete formatting. Well... didn't actually suffer - actually got some much needed housework done while I was waiting. Even managed to find over two whole dollars in change and a lost set of keys.
I appreciate the follow ups here - I think I'm almost Googled out, trying to get to the bottom of this. It's got me kinda leary of committing anything to a drive without instantly burning it to disk.
I love Humanity - It's the people I can't stand
0
OK, I did some research for you. Your MBoard is based on the Intel 865PE chipset. According to Seagate the Intel 875P chipset is required to support this drive. I could be wrong here as I assumed this was a SATA drive. The PATA may be something different. Look here for further info on the Seagate. I never heard of the other HD brand. http://www.seagate.com/products/interface/sata/compatibility.html
0
Again, I thank you for staying with me on this.
Yeah - I've been to the Seagate site. Was going to post that I've run the online diagnostics there and the drive does show up with errors in the file system. So, their recommendation is to run chkdsk to fix them. I've run chkdsk from Windoze and it comes up every time I reboot - and it comes up clean!
I went into event viewer to see what's what, then found a site to explain the error code(55)and, the best I can tell, it's an I/O error causing this grief. What that means exactly, I'm not sure. I'd figure it means something in my hardware and/or driver config ain't quite right. (duh) I just don't understand how diagnostics can say there are problems and chkdsk doesn't catch 'em and/or fix 'em!
From Microsoft:
If the NTFS 55 message appears regularly, for example daily or weekly, run Chkdsk using
the /R command-line option. This option allows Chkdsk to locate bad sectors on the hard disk.
Currently there are no Microsoft Knowledge Base articles available for this specific error or event message. For information about other support options you can use to find answers online, see http://support.microsoft.com/default.aspx.Is there ever an answer to a problem on their site? Anyway, as far as the I/O error goes, all I can think of is a driver problem. That is a problem because I've seen quite a few driver install issues raised for this motherboard on a few of the sites I'd run across.
Any more thoughts here?
I love Humanity - It's the people I can't stand
0
Oh - and my drive is listed on their compatibilty chart
I love Humanity - It's the people I can't stand
0
Your drive is on WHAT compatibility list? The chipset on your motherboard ISN'T on Seagate list of chipsets that are compatible with the 750GB drive.
"then found a site to explain the error code(55)and, the best I can tell, it's an I/O error causing this grief. What that means exactly, I'm not sure."
I/O is one of the functions of the MBoard chipset. Check with Seagate and find out if your Seagate drive IS compatible with an Intel 865PE Chipset.
0
I didn't get riled when you assumed my drive was SATA, even though my original post said it was IDE. Why get riled at someone who's simply trying to to keep up here? I'm not young. I didn't grow up with computers in school and there's no one here to show me anything. I've been trying to learn what I can over the past few years or so through trial and error and forums like this. As usual, I'm in over my head, but that's how I've learned whatever I've learned. Forgive me if I looked at a list and mistook it for something it's not.
You said this drive needs the Intel 865PE chipset to run it, right? I don't know where you got the information that my motherboard isn't based on that chip. It is.
P4P800-E Deluxe
So... I don't understand why, if that's the chipset I need and that's the chipset I have, why all this is going on.
I sent an email to Seagate support. Hopefully they'll have something for me.
I love Humanity - It's the people I can't stand
0
I read all your posts over and I don't see any reference to ATA or SATA. I got the chipset model from the ASUS site using your MB model. I assumed you would buy SATA with such a large capacity.
From MY response #6 above:
"According to Seagate the Intel 875P chipset is required to support this drive."
Your MB has the Intel 865PE, which is an older chipset.
0
Third paragraph of my original post on this thread - "checked the IDE cables..."
And, yes, I missed the lack of the letter E in the chipset ID... What a fine time to find myself short of cash. Looks like I need either a new board, or some more, smaller hard drives.Damn.
But, thank you yet again for seeing this through with me.
I love Humanity - It's the people I can't stand
0
If you reread my response #6 I stated that your MB chipset didn't support the Seagate 750GB SATA drives. I don't know if there is an issue with the PATA version or not.
Your Chipset is the 865PE - The supported chipset is 875P. Note 865/875. The suffix is less important than the actual series.
In my defense, you originally stated you had 6 drive installed in this rig. I had no way of knowing what was what.
Your problem COULD possibly be a Win2000 issue too. I don't really know. There have been issues with Windows being shutdown before the large cache in these huge drives has completely flushed. This causes eventual corruption. I believe there may have been a patch for this behavior. Are you using the SATA channels or the RAID functions at all? Check the controller card specs to be sure IT can support the drives.
One other thing of note. I asked how the drives were formatted earlier. I asked because I have read that if you format NTFS drives using non standard cluster size (Std 4kb) and then use third party defrag or scantools you can have problems. I have no personal experience with this issue.
0
The drive I'm concentrating on right now is the one that's running off the Win2000 Promise Sata378 IDE Contoller. That, in itself, makes no sense to me - a SATA controller for IDE drives? The 250's on the same controller ran fine. Beyond enabling LBA and making sure the BIOS is set to run the controller in IDE mode, is there anything else I should be doing to use a drive this large?
I love Humanity - It's the people I can't stand
0
Howdy, Hill-
On and off... been stopping by every once in a while. : - )
The latest I've got is that Seagate has nothing for me beyond their Seatools and/or chkdsk.Something else to add maybe?
I love Humanity - It's the people I can't stand
0
Yea, I was wondering why you are using the Promise Controller card. You really don't need it do you? You are aware that Maxtor drives are the only drives that are rated at ATA133? All others are ATA100. This is meaningless anyway because NO drives run anywhere near that fast. I was thinking that part of your problem may be the Promise card. You need to check that the Promise card is 48BitLBA compliant. That means it is capable of handling those large drives. If it isn't then the drive WILL get corrupted when Windows writes to a section of the drive that the controller can't use.
0
My current configuration is made up from what's left of my old PC - the one with the dead motherboard. I had the 133 cards because I used to have Maxtors (ages ago). I know the cards are LBA compliant - was running 250's on 'em for at least a couple of years without problems. At least nothing to this degree.
I can't seem to find anything to verify it, but it seems to me that there's yet another cap to drive size. Otherwise, I don't understand why some of the external hard drive enclosures are advertised as being for "up to 300 gigs", or "up to 400 gigs". Or is that getting into bridgeboards and a whole 'nother mess? Seems to me, once you get past 137 gigs, you're... ummm... past 137 gigs.
And thanks for coming back (He laughs) Now I'm back to 4 people scratching their heads on this.
I love Humanity - It's the people I can't stand
0
I would try pulling the controller cards anyway. When the MB manufacturers are configuring the BIOS for a new MB I bet they don't even consider the odd situation. At any rate the two things I KNOW can cause data corruption are the two I mentioned already. Shutdown before the drive cache has flushed itself or writing to a section of a drive beyond what has been partitioned. I suggest you try setting up one of these drives using the tools from the disk manufacturer instead of Windows.
0
So you're suggesting I reformat the drives using, in this case, the Seagate utilities that came with the drive? I'm not saying that that's not a reasonable idea, but a format is a format, right? I guess I'd just like to understand how a drive could be formatted differently. I mean I know you can set cluster size and such, but wouldn't the default settings be the same in both the Seagate tools and in the Windows Computer management?
0
If you are correct about an upper limit on drive size, the Seagate utility will address that issue. I am wondering if Win2000 may have some limits on disk capacity. Doesn't seem logical but something has to be going on. I have had issues installing Win2000 on newer hardware but you have a successful install don't you? As you know, Win2000sp4 & WinXP are both 48bitLBA compliant. In the case of WinXP I believe that using the manufacturers utility can overcome the 137GB hardware barrier. I don't believe it is a true drive overlay either. Kink of like a HDrive BIOS update. I could be allo wet with that analogy though as I don't use WinXP.
You could also try one of the drives directly on the MB ports using Windows to configure the disk. If the drive doesn't configure correctly it will give you some incite as to what is going on.
0
Part of the problem is, I've done just that - in one of the (now) three formattings of this drive, I did let Windows install the drive - and right off the motherboard. The install appeared to go fine, next thing you know, I'm still looking at all this corruption. So I guess it's debatable whether I've had a successful install or not. I think I'm going to have to do a side-by-side comparison between my current mobo and the old one. I had a P4P800 Deluxe before - this one is a P4P800-E Deluxe. I knew I lost an IDE channel and got SATA right off the board with the new one, but I thought the rest was the same. But I've found other differences - like different integrated sound, a different ethernet adapter, etc. So I agree - there's gotta be something going on here.
I'm even thinking I might just format the drive again and split it two partitions to see what happens. Was also a thought of RMAing the drive...
0
Now I hope you're still here
An interesting update, at least I think it is
Went to run msconfig and found something called PtiuPbmd. Rundll32.exe ptipbm.dll, SetWriteBack.Never having seen it before, I looked it up and there seem to be plenty of tech forums identifying it as both a component for Promise controller cards and a threat. I also see I can disable it if I want to disable write back caching on the drives. Isn't write back caching what "secures" the data to a drive? Then again, if I'm reading the article here correctly
Disabling write back caching
it seems that disabling it might remedy my situation here.Any thoughts here?
0
This is from my response #13.
"Your problem COULD possibly be a Win2000 issue too. I don't really know. There have been issues with Windows being shutdown before the large cache in these huge drives has completely flushed. This causes eventual corruption. I believe there may have been a patch for this behavior." What you really want is for Windows not to shutdown before the cache is flushed. Disabling the cache could hurt performance.
0  |
 Virtual memory settings
|
NT Authority shutdown!
|
This post is quite old and has been locked from receiving new replies. Please create a new posting instead.
| Ads by Google |
