Heat, dust, hard drive or memory

[amos]

Win2k FAT32 4x256 PC2100

I have been happily pooting along when

Come in in the morning and find 100 "Unknown Hard Error" popups (no other comments) on the pooter. I clear them all, and log in to the screen saver. They keep popping up, and then comp reboots. After reboot everything seems to be okay until it goes into screensaver, and then boom. Aha I say. SS problem. Loads more "Unknown hard error"s then reboots. But this time machine doesn't reboot properly (not just scan disk). Then reboots but takes a very long time and bosh - another set of "Unknown hard errors" and down we go again (sounds like a Maritime adventure!). Mmm I say, cause I'm not very eloquent like that. Try Safe Mode. Wouldn't boot in safe mode. Try again. Nope. At this point I probably used some word I would be ticked off for if I quoted it on techimo. Safe mode fails. Duh. Open computer.

Look in.

Well as I put all the pieces in there in the first place it was hardly surprising. Although there was a ton of dust. Major amounts.

Well I'm not antipodean, and this is scotland so Jan is not the time of year you associate with overheating. Admittedly the heating in the building is rather overzealous, but even so.

So here's the Q (you have been so patient!)

I've cleared the dust out, and taken the side off the case. The computer hasn't had any problems again in 2 hours. But if the problem is likely to come back it would be nice to know before it did, especially if it is a hard drive failure.

Q1 Could dust cause these problems? Like it is ordinary dust - I'm not wearing chain armour and shedding iron filing fibres or anything.

Q2 I doubt the computer seriously overheated. The vent is pretty good. Certainly neither the processor or system chip boiled - via hm would have logged it. Even so I am running problem free at the mo, and the box is probably cooler as the side is off. So it could have been sensitivity of the memory or hard drive (seagate 3yo) to a little extra heat. If its the hard drive - I want to know now, as it would be easier to pull stuff across _before_ it finally blows than take it all off backup.

Any advice would be welcomely received

[Omardeth]

yes massize amounts of dust , and thats what i am getting outta what you wrote , can cause this problem . dust is a great insulator it holds heat in very well .

can't say for sure if it was the hd or not . heat build up can cause lots of problem . just keep it clean and monitor it . put the side back on , you will be suprised that alot of the times it will run cooler with the side on than off . it has to do with airflow . with the side of there is no real airflow , but with it on there is .

[amos]

Thanks

The dust was in significant quantities, although not around the main hard drive. I'm not sure that it was so much to have main a major insulation difference (I wasn't shovelling it out, just blowing - although it did induce serious breathing problems).

If it all boils down to heat though, it is time I got a little worried. I've been running this beast fine for 3 years, and recent heat has not been out of the ordinary.

My computer does run cooler with the side off. There are 5 HDs in the thing so it can get pretty warm if they are all busy. Even so the recent whole drive backup seemed to work fine, which will have hammered two of them.

[amos]

Mmm now I've got the thing working again, it appears the event log gives load of errors saying

Error was detected on Device\Harddisk0\DR0 during a paging operation.

But I also got
Atapi
\Device\lde\IdePort0 did not respond within the timeout period.

for every one of these errors.

I have also been having some

\device\scsi\Hpt3xxNT1 non responding.

These are two different drives controlled by two different controllers! I think they correspond to my main drive and my backup drive. I'm really not sure what is going down here.

[Omardeth]

like i said there might not ba anything wrong with the drives at all . the heat build up in the heatsink or on the ram might also cause this problem . i would just monitor it now that you have it all blown out and see .

also make sure you blow out the power supply also .

[elroy]

Might d/l the drive utility from the manufacturers website and run a diagnostic test. Just to be sure. In case you have any valuable info on the drive you can't afford to lose. If the test is bad then start backing up if it keeps running.

[amos]

Thanks all.

As the drive is not completely failed, I will work at testing it out.

There is one other issue though. The UDMA100 drive is being accessed through PIO according to hardware resources. I cannot seem to get this to be seen as a DMA drive. The bios is fine, and all the other drives, on the same cables are fine. Obviously I'll do a cable swap to see if that is a problem, but is it true that sometimes the OS reverts to PIO in cases where there are drive errors?