Hard Drive access light on constantly 2

[adale]

The hard drive light is on constantly, except when occasionaly the pc is crashing whilst trying to run Battlefield 1942.

Does the hard drive light stay on when the Pc goes through the initial BIOS test/boot up stage with other people as I'm starting to think it has something to do with a conflict?

The PC specs are:-
MOTHERBOARD - ABit 8363 KT7[A] Version v1.0 ~ v1.2
BIOS Award - Version VT8371 - 42302e31 6.00PG 11/07/02
Processor - AMD Athlon(tm) 1.2Ghz Socket A
Memory - RAM 256 MB [Pagefile 384 MB]
Graphics - NVIDIA GeForce4 Ti 4200 64 MB [Driver 6.13.10.4072]
SOUNDCARD -Creative Audigy
MONITOR - Dell D1226H
OS - Microsoft Windows XP Ver 5.1.2600 SP1
File system - NTFS [System Restore=Off]
HARD DISK - Maxtor IDE 32049H2 19.08 GB
DVD - LG DVD-ROM DRD8160B
CD/RW - Creative CD-RW RW2024E

 

[gargouille]

thread751-363931
is a thread addressing the issue...just down a ways from your post...very similar topic...very good answers...

 

[edfair]

Is this the hard drive light, or the HDD light on the front of the case?
With some systems the HDD lite stays lit most of the time but it doesn't mean that the drive is actually being accessed. It is more an error in the controller chip programming.
Lite should flicker on once during post then be on while the OS is loading then off till the drive is used. Ed Fair
unixstuff@juno.com
Any advice I give is my best judgement based on my interpretation of the facts you supply. Help increase my knowledge by providing some feedback, good or bad, on any advice I have given.

 

[adale]

How can an error in the controller chip programming be rectified as it was not an issue when I had Win98 on an older bios version. Or perhaps there is some diagnostic app that you may have heard of to track down the offending code.

Thanks

 

[edfair]

Probably not a problem with your controller chip programming if 98 didn't have problems. May very well be a problem with XP and how it accesses the chipset.
First step probably to reinstall 98 to see if the controller has gone bad. Then reinstall XP to see if the problem remains.
There are no diagnostics or repairs available. You have a red light, you live with the red light until you replace the M/B.
You could cripple the onboard controller and install an expansion port controller and see if the symptom changes. Ed Fair
unixstuff@juno.com
Any advice I give is my best judgement based on my interpretation of the facts you supply. Help increase my knowledge by providing some feedback, good or bad, on any advice I have given.

 

[adale]

I know you're right but I get that sinking feeling about re-installing Win98 then XP as XP took ages to get installed.

Thanks for the help.

 

[Kocky]

Have you installed your motherboard into a new case ?
If so then check if you installed the cables that set the lights on and off are installed properly.

 

[adale]

The case and MB haven't been touched, and I double checked the HD light cable had not been touched accidentaly.

Thanks

 

[roamer1]

ALRIGHT, I AM STICKING MY NECK OUT HERE- I have run into this a time or two, but I don't know if it applies to you. The instances I have experienced came with the drive getting really hot and making noticeably more noise (rattling)and being noticeably slower than it did in Win98SE. If these things apply to you, then maybe this is the ticket.

The reason for the rattling and heat is that your ATA66 (or bigger) drive is stuck in PIO mode instead of ATA66

Here's some stuff to try:

!!!!! MAKE SURE SURE SURE you are using Proper drivers for your MB rated for 2k or XP !!!!!

Dwnld a utility to access the IDE controller Sorry, I can't remember what I used but probably there is something that will work in the Utility disk / website for your MB. look for readmes,RTFM,etc. You should then be able to determine if your HDD is indeed in PIO mode. If it is, change it to the proper ATA mode it should have. If it won't go there or goes back to PIO after a cold boot, look to see if that HDD is SL or MSTR to any ATAPI device as they can cause this. Remove the ATAPI device and try again. if it works after a cold boot you are fixed, but can't use ATAPI against your OS drive anymore. VIA's utility will run TSR and allow ATAPI to be on the same IDE port as ATA66 (I believe it is an install feature in the Autorun). Don't know about others. this is really common on Athalon slot boards running a VIA southgate.

If that don't work disable SMART in CMOS and try again

Hope this works for ya. Lemme know too. I got a half a bag of Cheetos (Puffy, slightly stale)invested in this

Bruce

 

[gargouille]

Interesting stuff, roamer, but shirley in the part about
"Via's utility will run TSR" you meant to say "and NOT allow ATAPI to be on the same IDE port as ATA66." (picky, picky...)

And a harumph to those we hear saying, "I didn't load the chipset drivers for the board and it runs just fine".
(not suggesting you, adale)

Adale, you say it took ages to install XP (I've told you a thousand times to quit exaggerating!) and in the accompanying statement, there's a hint that you may not have clean installed XP. Zatso? IF SO, could that be the unintended consequence that's causing this?
Also, have you considered running HD benchmarking utilities?

 

[adale]

When I installed XP I had reformatted the HD first, but didn't realise I had to update the MB bios and it took a while for the penny to drop on that one. Since then, I've been trawling the net to find the most up-to-date drivers for everything. I don't know why I haven't tried HD benchmarking as that will tell me if the data rate etc is correct. After that I will try the IDE controler thingy above. Tks.

 

[roamer1]

You are right Gargouille. I should have said: VIA's TSR IDE Utility allows ATAPI to run on the same ribbon as an ATA/66 in 2k without forcing the drive to PIO. I stand corrected. I don't remember benching anything. Must have gave me 33, or I moved the ROM to IDE2 and scrapped the TSR, which is more likely. It was long ago, and a painful memory. I had to clean out about half my cobwebs and an old sock just to get that info to boot up. I do remember I had to physically set the drive with that utility to get it outta PIO. would not come back on it's own.

And by the way, quit calling me Shirley.

Bruce

 

[adale]

I've benchmarked the pc using PerformanceTestV4.0 and everything looks about ight except the HD.
Sequential Write=7.7mb/sec
Sequential Read=11.4mb/sec
Random Seek & RW = 1.2mb/sec
Whereas the specs say it should be:
To/From Interface up to 100 MB/sec
To/From Media up to 46.7 MB/sec

Strangley, the first time I ran the test the HD light decided to work properly. And the same happened when I tried unsucessfully to run SiSoft Sandra (which crashed).

I'm not sure about your instructions to do with the PIO mode. The bios says it will look for PIO automatically, and there is no reference to SMART in CMOS that I can see, but I may be looking in the wrong place.

I haven't been able to find a utility that will access the IDE controller have you got some more suggestions as to what I should look for ?

Thanks

 

[roamer1]

ADALE
I haven't dealt with this for a while, and as I said before, I can't remember what I used (from my toolbox). I know it is gone though, as I lost it in the Great Root Beer Disaster (note: (215Floppy+1CanRootBeer)=(212Coaster+3Floppy)). I would be very reluctant to recommend (or use) a "universal" controller interface without having prior good experience with it. Best to find one tailored to your MB. or at least to your chipset maker. Let me go look.

A few other ideas:
I wonder if the drive still "accesses" if you boot to a 98 floppy? just to be clear that it is a MB/HDD problem and not a NT driver thing that I am missing.

Also try to boot to floppy w/ HDD in a different system if possible. The theory being to further isolate the fault to the drive.

Resource: Maxtor has some little gizzies to manipulate ATA33/66/100. never used them, but they may do the trick if the HDD is doing it. While you are there you might pick up the powermax utility to diag the drive.

If the Controller is doing it you are probably screwed without access to the controller.

I don't know ABIT very well, tend toward MSI, so i don't know if your board is SMART compliant or IF there is a switch in CMOS to turn it off with. so don't sweat that.
Lemme look about a bit...

Bruce

 

[roamer1]

Sorry I took so long Adale.
Goto Device Manager>IDE/ATA/ATAPI Controllers

Make sure you are seeing VIA on the Titles or you didn't install them yet.

Choose PrimaryIDE Channel (I assume thats your OS Drive)
ADVANCE SETTINGS TAB has access to the controller and will show current mode. If current mode is "PIO only" hit the pulldown and choose "DMA if available"
Current transfer mode shows what the board thinks it is transferring at.

IF you are in DMA Mode already it will be a HDD Problem
IF you are in DMA Mode in the ComboBox but it reads PIO in the current mode it will be an HDD Problem
Repair with Maxtor Powermax.exe

If you are in DMA Mode and is not showing at least UDMA Mode5 I might try the proper ATA utility from Maxtor If that isn't it, Back to PowerMax.exe

ABOUT POWERMAX.EXE
Obviously read instruction and sequentially (& incrementally)apply tests from least to most destructive. Back up first as the most destructive will be LLF and 0write

http://www.viaarena.com/?PageID=66

Explains the IDE driver set from VIA.
If you continue to have problems you may want to look at the MINIPort driver (which is the controller driver I was talking about as an install option in my initial post). But the page lists only a few instances where it should be installed instead of the filter driver that is default in VIA4in1

Good Luck

-Bruce

 

[adale]

I have 'Ultra DMA Mode 5' in the current mode so that makes it a HDD issue then I guess. I'll get the ATA utility from Maxtor next.
Did you see the comment I made earlier about the drive acting normally when I did the benchmarking ? Is that a clue of any sort ?

 

[roamer1]

ADALE Said- "Did you see the comment I made earlier about the drive acting normally when I did the benchmarking ? Is that a clue of any sort ?"[COLOR=/]

Adale- not that I can put a finger on, unless cmd is different, like booting from floppy- or it is looping on WriteVerify or SMART is doing something stupid at idle that is put to the side because of the demand the benchtest calls for. The Write Verify would slow it way down because everything it writes it reads. But not as bad as the specs you are showing.

ECC,CRC Errors?? CHKDSK should pick that up and it would have to be re-ocurring. But Powermax will prove that any way.

How about PowerMax? or the ATA utilities? Any luck there?

About PIO vs. ATA(UDMA)[COLOR=/]

I may have lead you a bit astray, and I may even be full of (@*&#%$. Lets think this through.

I am sorry I didn't (and can't)explain this better, and before anybody gets tense, this simple analogy is not exactly correct but it serves the purpose:

Imagine your HDD is a modem- 9600Baud would be about PIO Mode2(Modes are 1-4 I believe), and UDMA (ATA) is 14.4 and up(UDMA Modes 1-5). Your Controller is a switch at the phone company- and your data is coming / going beyond.

Data is like sending a fax- Both sides of the BUS "shake hands" and usually agree on 14.4- but if they have trouble talking they will AUTOREDUCE to insure they can understand each other. They talk slower. The problem is determining why. you don't know if your fax is broke, the other guy's is broke or the switch is broke. All you know is that you paid yer d**n phone bill and you ought to get better service than 9600Baud! And if there is static in the line WriteVerify turns on and slows it down sommore.

Now lets make it a little worse yet. You live on Maxtor Drive. You send your fax. It goes through the switch and into the phone company's Server (CPU), and right back through the same switch to be received by your neighbor across the street (Mr. N.T. Kernel)!!!
The bench you ran is still talking through the bus. You are sending the same fax.

Now, if you finally get fed up, you figger, Hmm I'll just walk on down to A Drive and stop off at that little DOS Boot repair and copy shop. I will send a fax from there over to my place to see if my machine picks it up ok.
This isn't quite as bad, but it's still the same phone company, and even though A Drive is on a different circuit it still needs to go through the same old switch to get to your house on Maxtor Drive.

So what can you do?
Well, if you are really sick of it I guess you could just pick up and leave. Just move. You hate it here anyway- It's much better over on West Erndigital blvd. The road there is brand new! But West Erndigital is on the same phone company and circuit- so how do you know you can fax from there? you might have the same problem... and the cost of the moving truck... and all that packing and unpacking...

Do you think Mr. Kernel would let you come over and fiddle with his fax to see if it is the problem? Not too likely! I know Mr. Kernel, and he is surely one grumpy old fart. He don't open his door to NOBODY! Besides, he's richer than anyone I know. He could probaby buy the phone company (mebbe he already did)! Surely he has the best fax you could possibly get!

Well what about the phone company or the sub-contractor who installed the switch? Nope. Union Shops. if you ain't got a card you can't do squat.

So heres yer choices:
You really gotta take your fax machine (HDD) to another town (box). You gotta find someplace where you know they got clean lines.
Then, if you find out your fax machine is broke, Go back home wrap that fax around a brick and lay in the ditch till Mr. Kernel comes out to go to work....

See what my thinking is?? Ain't it scary?

Logical progression (Process of Elimination)[COLOR=/]

Boot from a floppy and observe the drive if it winds up good and the light goes off... Good.
Run Powermax. if there is error, fix it, but don't yet assume that it is the drive or controller (or both)(or either). If it is fixed, yay. If the problem returns PROVE it.

Boot it from a floppy in another box if it shows the same behavior on a different MB there is problem with the drive. Repair it there. If it develops the same probem again (on the guest MB) then the drive is faulty and needs harder repair or replaced. If the repair sticks in the guest box, but returns when you bring it home then the controller (or something else) is wrong and causing the error to manifest on the drive.

If you are not fixed at this point you have proved the drive itself and the issue is in the board controller, OS or CMOS.

So...

With the rates you have there is NO WAY that HDD is in UDMA Mode5- or -the bench is losing data across the bus and giving you inacurate data -or a CMOS or soft setting is the culprit forcing low rates for some reason. Something is lying to ya (barring physical damage which I don't buy, worked good before install).

Try going back to the controller and choosing to DISABLE DMA (set it to PIO) /restart/ and choose DMA again. /restart/ If that don't fix it I will quit beating that dead horse. Look to see if you meet the specs to need the MINIPORT driver instead of the stock filter driver.

Nuther thought: with the earlier install probs- are you certain PNP is OFF and ASCPI is ON (if ALL are ASCPI compliant, otherwise PNP OFF ASCPI OFF) in CMOS? I believe that 2k failsafe needed CMOS set up that way. Not gonna cause DRIVE probs particularly, but can tend to spawn an occasional poltergiest... Please prove before action here- I don't know if XP is as picky. Also, I don't know that changing them now would fix it w/o re-install... ANYBODY ELSE?

Hope I didn't just muddy it up some more.

Sorry so long winded... and I did it all with just these two fingers...

Bruce

 

[adale]

Well I've cut down the areas to look at now as I've bought and installed a new Western Digital HD and re-installed WinXP which now thinks I've activated it too many times.

Guess what, the light is still on. So I guess that leaves us with the mother board. Oh joy!

 

[roamer1]

This HAS to be soft- same reason as not suspecting the drive- it all worked fine before.

Well, ssince youve got it, throw 98 on whichever is loose to see if it all works in 98- alot less complicated.

then it almost has to be OS / CMOS related. IF yournew

drive is ATA133 it needs the miniport driver
DID you look at the bottom of last post- PNP ASCPI...etc?

is the new drive benching poor too?

 

[JAFrank]

Two possibilities that I wonder about - you said that you didn't have this problem in Win98? How much memory do you have installed in this pc? If you have less than the recommended RAM then the system will thrash your hard drive as it constantly accesses the swap file. 64mb is the minimum for XP, with 128+ recommended.

Another less likely possiblity is that Indexing service is turned on, and that's what's beating up on your hard drive all the time.

If you look at Task Monitor, is your cpu activity high while the HD light is on? Also, if you leave the computer on overnight without using it, for example, does the hd light eventually go out?

JAF