Random lockups part 2: The Reckoning

[nanaki333]

Absolutely amazing. Ok, so here's the deal. My machine randomly locks up. sometimes 3 hours, sometimes 8, *maybe* a day. I've reinstalled windows 3 different times and ran repairs twice on different configurations. I've done this on 3 different harddrives. 2 WD 120GB and 1 Seagate 120GB. I've replaced the CPU, was an xp2400, now it's an xp2800. I replaced the cd drive in there, even took it out completely. Lets see. I got brand new pc2700 512MB memory. I bought a brand new motherboard. Went from an Asus a7v8x to an a7n8x, so it's a completely different chipset complete with the same errors. I went from a 500watt Raidmax PSU to a 420watt Antec. I bought a UPS backup in case i was just getting bad power. I've replaced everything in there but the case. I am completely stumped. I've run the comp with bare minimum in there to run windows. Basically just the harddrive as the only device plugged in. Does anybody have any idea what it might be? I have a brand new computer and a bunch of leftover hardware now.

Oh, I also took out my radeon9700 pro and put in my radeon7200 AIW.

 

[Mnementh]

Blimey, sounds like you've exhausted everything now, there's only one thing left... Software.

What are you running on your machine? Have you run a base winXP install (ie with no 3rd party software installed) have you tried another OS? Have you run Spybot and Adaware and a couple of different online virus scanners?

I assume you've tried all the different drivers and numerous versions? I can't remember if I asked you this before but were there any events in the eventlog?

Also just because a piece of hardware is new doesn't always mean it works unfortunately I would recommend you run memtest for at least 10 passes with no errors, if you get any errors then you need to RMA your RAM.

Let us know what happens and we'll look further if necessary.

Mnementh

 

[nanaki333]

hehe.. That's the really laughable thing. I used to get events in the event log, now i don't. When i reboot my comp now, I don't get the message the system has recovered from a serious error blah blah blah. I haven't tried spybot or adaware or anything like that recently. Last time I reinstalled, I didn't even have it on the network for about a week. I'd just go home and reboot it whenever I saw that it had locked up. Then taking pieces of hardware out here and there. I guess I may have to try and dig up my old Windows2000 disc and product key and try a new OS.

I've run burn-in test already on every component for a couple of hours, I don't know how reliable those readings are though. My last memory worked perfectly fine and I still had this problem. I don't see how one person can have so much bad luck with hardware. What are the odds that every component I've replaced is also bad.

I'm sorry, there were 2 things that I haven't replaced. The whole case, and the CPU fan. I had a volcano9 on my old processor, I have a volcano11 on there now. So I don't know, maybe the fan's vibration are shaking up the processor too much?

I'm going to install windows2000 on it this weekend when I have time. I'll set memtest to run when I get home and let it run while I'm at class tonight. Anyone think of any other little things I can do before the weekend?

 

[imported_raybay]

Memory defect, video card, sound card, or modem problem... I would exchange all of them with any other known working machine. Bad case fan or cpu fan, or something as simple as a defective IDE cable or power switch connector at the motherboard end. Also would again carefully check all PCI and AGP sockets to be certain you don't have debris. Strangely, our shop sees these problems caused by modem electronics even when the modem is not used.
You didn't mention the brand or model of the computer, and whether you have downloaded the latest BIOS and hardware fixes.
The standard fix is to remove absolutely everything you can... down to one memory module and floppy drive. Boot to floppy. Test. Then keeping a written record, add and change one component at a time, working from most simple on up. If it doesn't work with only motherboard and floppy boot disk,that tells you a lot. If it does work in its simplest state, figger out which added device reinstitutes the problem. Be sure to trade out memory modules first, and do not depend on memory testers.

 

[nanaki333]

heh.. I replaced all the IDE cables in there already with ones that were still in the wrap. I have the latest BIOS. I don't have a modem. The NIC and sound probably have little to do with it, since when I swapped motherboards, they have different hardware on them. Different NICs and soundcards on both, not to mention completely different chipset. As I said, I have my old radeon7200 in there now, just to rule out my video card as a possibility.

I mentioned the brand of the motherboard is asus. I build all my own computers, so it isn't a dell or gateway or anything like that. This is the first time I've ever had this much trouble. Amazing how I can build and fix everyone's computer but my own.

I stated already that I had everything down to the bare minimum to run windows with no luck. I don't even have a floppy drive in my computer (what's a floppy? hehe).

Like I said, I've replaced every single component in my machine. There's absolutely no reason it should be doing this.

 

[Mnementh]

Originally posted by: raybay
Memory defect, video card, sound card, or modem problem... I would exchange all of them with any other known working machine. Bad case fan or cpu fan, or something as simple as a defective IDE cable or power switch connector at the motherboard end. Also would again carefully check all PCI and AGP sockets to be certain you don't have debris. Strangely, our shop sees these problems caused by modem electronics even when the modem is not used.
You didn't mention the brand or model of the computer, and whether you have downloaded the latest BIOS and hardware fixes.
The standard fix is to remove absolutely everything you can... down to one memory module and floppy drive. Boot to floppy. Test. Then keeping a written record, add and change one component at a time, working from most simple on up. If it doesn't work with only motherboard and floppy boot disk,that tells you a lot. If it does work in its simplest state, figger out which added device reinstitutes the problem. Be sure to trade out memory modules first, and do not depend on memory testers.

Please please read the whole thing before giving out rote phrase advice.

You started out so well and then had to end it badly, if you'd have read the OP you would know that the "standard" (there's no such thing in the PC world btw) fix does not apply in this case, your "standard" fix is for machines that don't boot or are having problems staying up for more than a few minutes, not applicable in this case as the machine stays running for up to a day sometimes. The four main causes for this to be happening are power, software, memory and heat. We've already ruled out power and most probably heat, that leaves software and memory. Memtest86 is imo (and in the opinion of most people who frequent these boards) the best program for testing for memory errors, as the OP has already stated he has swapped out his RAM so rather than keep doing that and hoping for the best a 10 pass run of memtest86 is the way forwards.

Your advice about cables and debris was top notch nobody else has mentioned it just please please think about what has been written before answering with the tripe given out in stores...

Mnementh

 

[nanaki333]

well said. i probably should have also mentioned, my case sides are off and my temperatures run pretty cool. last night, i think that my fan was s till flucuating in speed. do you honestly thing my cooling fan on the CPU could have anything to do with it? i have all my other fans unplugged right now. i don't know, just trying to eliminate more sh!t before i rip the whole thing apart and inxpect my case to make sure there's nothing connecting to my motherboard to cause a short. probably nothing there cause i've taken the old board out to put the new one in and it looked fine. memtest shall be the first thing i run when i get home.

 

[PingSpike]

What temperatures are you getting from your PC Health status section in the bios?

I second running memtest-86 at this point. Run it overnight just in case, then you can rule out bad/incompatiable memory.

What components have you NOT replaced? Is there a sound card...a harddrive? Anything at all?

And just for the hell of it, replace all the IDE and Floppy cables. You probably have 10,000 extra lying around so it can't hurt.

 

[nanaki333]

already said i replaced the IDE cables and don't even have a floppy drive installed.

i've used 3 different HDDs, 2 different brands. also stated i swapped the mobo already, so it's a completely different chipset, onboard lan, sound, the works. i'm gonna go ahead with memtest most definately. just trying to get some insight on anything else i can try. only thing not replaced has been the case. spooky, eh?

 

[PingSpike]

Originally posted by: nanaki333
already said i replaced the IDE cables and don't even have a floppy drive installed.

i've used 3 different HDDs, 2 different brands. also stated i swapped the mobo already, so it's a completely different chipset, onboard lan, sound, the works. i'm gonna go ahead with memtest most definately. just trying to get some insight on anything else i can try. only thing not replaced has been the case. spooky, eh?

Oh man....LOL. My official diagnosis is you have a ghost in your case. Contact your local exorcist.

I had a similar deal like that this morning. I had a known working motherboard with just a pci video card and drives hooked up to it. It kept locking up at the IDE HDD detection, sometimes before it. I unplugged the drives, it still did it. I tried a different PSU, it still did it. I tried a known working stick of RAM, it still did it. Reset the bios, still no go. Checked all connections, looked for shorts...there were none. I was pissed, there wasn't anything left to try!

I was about to say I'd zapped the board while putting it in, then I swapped the PCI video card out for a different one. WTF...thats the first time a video card has caused hard locks like that before windows even starts to load. Live and learn.

Anyway,
You're sure there's nothing common? Not even that stick of ram in question? If thats the case, then we must be looking at a software problem, not a hardware one.

I have a slipstreamed copy of windows 2K SP4 that works fine usually, but the install was wonky on a few asus boards I tried. If I installed from a standard disk and put SP4 on after the fact the boards ran like a dream, but all four clone boxes hated the install from the slipstreamed disk. So try a different OS or at least install disk I guess, since thats the only remaining common factor.

 

[Mnementh]

Originally posted by: nanaki333
well said. i probably should have also mentioned, my case sides are off and my temperatures run pretty cool. last night, i think that my fan was s till flucuating in speed. do you honestly thing my cooling fan on the CPU could have anything to do with it? i have all my other fans unplugged right now. i don't know, just trying to eliminate more sh!t before i rip the whole thing apart and inxpect my case to make sure there's nothing connecting to my motherboard to cause a short. probably nothing there cause i've taken the old board out to put the new one in and it looked fine. memtest shall be the first thing i run when i get home.

The only thing I would recommend for the HSF is to remove it and clean any thermal paste of it and the CPU and then reapply with some fresh paste.

Install asusprobe (it's on the CD or you can download it from the website) it should give you the temps for your CPU and the ambient temp if the CPU is running at less than 60 deg C it's fine up to 70 deg is OK too but I like to keep it as close to 50 or below as possible.

Mnementh

 

[nanaki333]

wow, just had something really funny happen. think i caught the ghost in the act. i was sitting here after i rebooted and my case suddenly got quiet. i look in there and my CPU fan just stopped spinning. i don't know if the mobo told it to stop or if the fan itself is going bad. doesn't surprise me though, it is pretty old. few weeks, which seems to be the average lifespan of components lately.

 

[Mnementh]

Originally posted by: nanaki333
wow, just had something really funny happen. think i caught the ghost in the act. i was sitting here after i rebooted and my case suddenly got quiet. i look in there and my CPU fan just stopped spinning. i don't know if the mobo told it to stop or if the fan itself is going bad. doesn't surprise me though, it is pretty old. few weeks, which seems to be the average lifespan of components lately.

Did the fan stop after the PC crashed or did it stop while it was still running?

If the machine was still running then turn of Q-Fan support in the BIOS (I've never managed to get it to work reliably on my A7N8X-Deluxe) that should stop the mobo from attempting to control the CPU fan.

We will solve this we're just getting to the stage now where we've nearly eliminated all the possibles so whatever is left however improbable must be the cause (Yes I'm sure I've mangled the quote, Sherlock Holmes iirc, but you get the point)

Mnementh

 

[nanaki333]

i ran memtest and it went about 12 passes with no errors. yeah, the computer was logged into windows when it stopped. i was opening up mozilla and then, it just got quiet. really really quiet. cause them volcano HSF are loud folkers. looked in the case and it was stopped.

 

[nanaki333]

ooops.. oh yeah, mbm5 reports my cpu fan spinning at like 8000rpm which doesn't seem quite right, especially considering it wasn't turned up all the way either. these boards have that feature that if the processor overheats, it just kills the processor but leaves everything else on right? i don't wanna disable that in case the fan never starts spinning again and my processor melt. i mean, the fan like cut off completely. not spinning at all.

 

[huesmann]

how about you replace your fan and get back to us?

 

[psiu]

The Q-fan on Asus boards is just their attempt at fan control....Basically it tries to control how fast your CPU fan runs. You can turn it off, it's a seperate feature from COP (Cpu Overheating Protection).

 

[nanaki333]

w0rd. tune in next week for part 3...

 

[PingSpike]

Do all fans stop or just the CPU fan?

Replace CPU fan.

 

[Ryoga]

Definately replace the CPU HSF. It should never, ever stop. It might switch from low speed to high speed, but it should never stop.

What does it do when it "resets"? Do you get a blue screen? Does the system completely freeze? Or is it a "welcome to your video BIOS show"?

If it's the latter, maybe there's a short in the power or reset switch in the case. Try disconnecting the reset switch wire from the mobo.

Is the motherboard touching the case? Maybe it's grounding somehow. Standoffs properly installed and everything?

 

[nanaki333]

it doesn't reset, it just locks up. the cpu fan just stops spinning completely. usually when it locks up, all devices are still on in the case, even the cpu fan (guess it starts spinning back up eventually), but lately it's been just cutting off and staying off. all the components in the comp are on when it's locked up with or without the cpu fan spinning. i'm going to go to the store this weekend and get a new one. no time with work and class during the wee unfortunately.

as i've said, i've taken out the old mobo and put in a new one and there was no screws or anything like that between the case and mobo. there's plenty of space to fit your pinky between the case and board. a lot of my power cables are just kind of laying on the bottom of the case though, would that cause a short?

 

[PingSpike]

Originally posted by: nanaki333
it doesn't reset, it just locks up. the cpu fan just stops spinning completely. usually when it locks up, all devices are still on in the case, even the cpu fan (guess it starts spinning back up eventually), but lately it's been just cutting off and staying off. all the components in the comp are on when it's locked up with or without the cpu fan spinning. i'm going to go to the store this weekend and get a new one. no time with work and class during the wee unfortunately.

as i've said, i've taken out the old mobo and put in a new one and there was no screws or anything like that between the case and mobo. there's plenty of space to fit your pinky between the case and board. a lot of my power cables are just kind of laying on the bottom of the case though, would that cause a short?

That does sound like a heat problem related to the CPU fan failing. You usually get hard locks like that when the processor gets a little to toasty. The good news is since it has a sink on it I doubt it did any permanment damage to the CPU.

 

[huesmann]

once again, have you tried replacing the CPU fan???

 

[Mnementh]

Originally posted by: huesmann
once again, have you tried replacing the CPU fan???

Did you not read the thread?

Originally posted by: nanaki333
i'm going to go to the store this weekend and get a new one. no time with work and class during the wee unfortunately.

Yes you're quite right he needs to change his HSF but there's no need to get on his case when he's already answered your question if you'd just been bothered to actually read before jumping in.

Mnementh

 

[nanaki333]

well, just bought a brand new HS/F, and surprise surprise, it didn't work. every last component has now been swapped out and i'm still getting lockups. this is becoming a thorn in my side. i guess i'll play musical chairs with the harddrives again. that's the only other thing it could be. i haven't been having too much luck with WD drives and i just happen to have a couple in there. even if they are new, i don't trust em. unless anyone else can think of something i can do.