• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Help Louis diagnose his system lockup problem?

L

LOUISSSSS

Diamond Member
system is exactly as in signature. everything is running at stock speeds

situation:
PC would lock up mostly during idle, but it has happened at least once during desktop usage, and once while playing CS.

Stability Testing Results:
CPU: passes 24 hours p95, 2 hours OCCT, 50 passes LinX (512mb ram usage)
RAM: passes memtest86+ for 24 hours, HCI Memtest for 12 hours.
GPU: passes 30 mins of ATI-tool error checking
HD: passes HD Tune Error Check

temps are WELL within range and PSU is brand new (1 month old; and same thing happened with old psu)

BIOS settings are all set at fail-safe defaults with latest BIOS. only things changed are:
loadline-calibration: disabled (from auto)
audio codec: disabled
memory timings: manually set to manufacturers settings: 1000mhz, 5-5-5-15 at 1.9vRam
CPU voltage: 1.325 (bumped higher lately just to ensure maximum stability)
pci-e freq: 100

is this problem related to the suspend settings? (str, pos?) right now i set it to str, but it has happened under pos also.

i know this is a tough problem...help will be gladly appreciated by a member who often helps everyone out in these forums
 
Some history might be useful. Did this build work for a while with no changes? What changed - s/w and/or h/w, power surge from a storm, was the case physically moved, etc? How much time passes between lockups? Does it ever suspend and not lock up? Is there any consistency? Can parts be swapped with another build to develop a baseline of goodness?

Most inexplicable lockups are related to ram in my experience. I've had memtest run for 40 hours successfully but replacing ram made a problem go away. I'd try swapping the sticks or, even better, try others.

If the cpu is truly at stock, then default settings for the rest of the mb settings (including ram) should suffice. In only one build did I have to up mem to 1.9v when everything was at default.

Another possibility is a marginal (soon to fail) cpu if it's been stressed for a while. I assume you've visually inspected the mb capacitors.

Have you tried reloading the os? It would be ideal if you had a free drive to do this on so as to not disturb the current drive contents in case this doesn't reveal anything. I know it's a pain to reload but it's solved problems for me in the past. If this does solve the problem, it can be looked on as a forced house cleaning...

The cpuz validator indicates win xp sp3. Have you tried vista? how about linux?

And last, but not least, what's the chance of a malware infestation?

It's pretty big of you to admit being stumped especially since it seems that you are usually solving problems for others here...
 
heres some information that i can give to out before i pass out from being awake til 4:30 am

- lockups happen every 8-48 hours. more towards 8-12.
-i don't exactly have more ram to test in my system, but i trust memtest86+ to tell me that my ram is working when both sticks pass it for 50 passes individually (and together)
the cpu was overclocked for about a year but under 1.4vCore and 8 x 400@ 3.2ghz.
-system has been through a complete reformat and has happened on both winxp installs (i do not have access to vista)
-there is a slim to none chance of malware. this is a fresh install.
-i have the pc always hooked up to an APC back-ups so i dont think any surges would've gotten to my pc
-i dont know how to visually inspect mb capacitors**

out of ideas..
 
Sounds like you've done your due diligence. Swapping ram though is a good idea though if you can track some down.

Uninstall your Audigy.

OS stuff:

Check your power saving settings. Make sure it's set to whatever "Always On" and "Shutdown: never" for your HDDs.

Make sure your paging file is managed by windows.

MS debugging tools - pretty complicated, I've never successfully troubleshooted anything using them but worth a shot:
http://support.microsoft.com/kb/315263
 
Your board is dying.

I dont think theres any other explaination.

Or its degrading.

If i remember correctly you F@H'd no?
 
By board, do u mean motherboard?
I ran fah for about 6 months on this thing overclocked. How can I fully ensure that it's my mobo??? I pass all stability tests...
 
I had a problem with my E2140 @ 3.2Ghz system, it was stable for 24hr Prime95, 24hr Memtest86+, etc.

It would run for up to a month at a time, but then it would either cause MagicJack or SeventeenorBust to crash, or it would spontanously reboot.

It did it since it was built. I have it plugged into a ES-550 APC UPS. Originally, it would reboot within a 24hr period. Finally chose to plug it directly into the wall, and it stopped rebooting. So I replaced the UPS. Now it only reboots every week or month. So I think that the second UPS is faulty. I eventually built a F@H rig, and I hooked that up to the UPS, and it was at stock speed, but it rebooted too on the UPS. So now I have a beefy 1200VA/720W Ativa UPS (mfged by APC for Ativa), and so far, so good. I'm trying to exceed one month of uptime on the new UPS. If that works, then I know that the computer is good, and it was the UPS that was bad.


I have a different system, an E4400 on a Conroe865PE board, that was overclocked to 2.8Ghz, but is now running at 2.0Ghz. (Stupid BIOS removes the FSB selection option when you have a storage controller card installed, unless you reflash the BIOS.)

It runs for multiple months on end, running SeventeenorBust, and it is connected to a BX1500 APC UPS (1500VA).
 
Originally posted by: VirtualLarry
I had a problem with my E2140 @ 3.2Ghz system, it was stable for 24hr Prime95, 24hr Memtest86+, etc.

It would run for up to a month at a time, but then it would either cause MagicJack or SeventeenorBust to crash, or it would spontanously reboot.

It did it since it was built. I have it plugged into a ES-550 APC UPS. Originally, it would reboot within a 24hr period. Finally chose to plug it directly into the wall, and it stopped rebooting. So I replaced the UPS. Now it only reboots every week or month. So I think that the second UPS is faulty. I eventually built a F@H rig, and I hooked that up to the UPS, and it was at stock speed, but it rebooted too on the UPS. So now I have a beefy 1200VA/720W Ativa UPS (mfged by APC for Ativa), and so far, so good. I'm trying to exceed one month of uptime on the new UPS. If that works, then I know that the computer is good, and it was the UPS that was bad.


I have a different system, an E4400 on a Conroe865PE board, that was overclocked to 2.8Ghz, but is now running at 2.0Ghz. (Stupid BIOS removes the FSB selection option when you have a storage controller card installed, unless you reflash the BIOS.)

It runs for multiple months on end, running SeventeenorBust, and it is connected to a BX1500 APC UPS (1500VA).
Click to expand...

so all of your PCs (except for the last) reboots randomly when you have it conencted to your APC UPS? hmm that is strange, why do you guys think its doing that?

i'll try plugging my pc into a normal surge protector right now and report back if it freezes again
 
I had a similar problem with my mom's computer. After many, many hours of searching for a solution I tried disconnecting the reset switch. It worked. Its a long shot but free to test.
 
You don't have a wireless card installed, that has a RaLink chipset, do you?

I was just reading last night that apparently they cause hard freezes on systems, with more than 4MB of RAM, during heavy usage.
 
Originally posted by: VirtualLarry
You don't have a wireless card installed, that has a RaLink chipset, do you?

I was just reading last night that apparently they cause hard freezes on systems, with more than 4MB of RAM, during heavy usage.
Click to expand...

i think i might, althought i'm not sure how i can find out.

my wireless card is USB interface model number: Linksys WUSB54GP.
i know theres more than 1 version but i dunno how i can find out...
 
link

Apparently v1 used PrismGT chipset (I think I have that version somewhere), and v4 uses RaLink chipset.

But if the adaptor is USB, I don't see how it could have the 4GB RAM issue (64-bit DMA issue), since it's not a PCI or PCI-E adaptor. Eg. everything has to go through the chipset's USB ports, therefore the chipset is responsible for DMA, therefore it's unlikely that that is causing the problem.

How much RAM do you have?
 
Originally posted by: VirtualLarry
link

Apparently v1 used PrismGT chipset (I think I have that version somewhere), and v4 uses RaLink chipset.

But if the adaptor is USB, I don't see how it could have the 4GB RAM issue (64-bit DMA issue), since it's not a PCI or PCI-E adaptor. Eg. everything has to go through the chipset's USB ports, therefore the chipset is responsible for DMA, therefore it's unlikely that that is causing the problem.

How much RAM do you have?
Click to expand...
yea i highly doubt it is my usb wireless related, i've been using that for years and this problem only happened in the past months.

i have 2 x 2gb of ram
 
guess people ran out of ideas... how about i throw some stuff out there:

i did run my system at 100% full load (CPU + GPU) for about 6 months, both CPU+GPU overclocked. the load was from Folding@Home.
the highest i've ever set my vCore was 1.4v and the highest i've ever seen my core + Motherboard temps was 75c. i always stop the test and back off after that.

the motherboard + CPU is about 1 year old. and the computer remains on for pretty much 24/7

i cannot attach the 8-pin power to my cpu due to its design where one of the heatsinks blocks 4 of the 8 pins, so i only have a 4-pin in there and thats all i can do.
view pic here: http://s47.photobucket.com/alb...w&current=dsp38pin.jpg
the ones that are not plugged in are the 4 plugs that are covered by the black plastic cover (what are these for?)

without just saying "YES" because there is no other obvious explanation, who thinks my motherboard has a problem?
 
I think your hardware is probably fine, try taking out both hard drives, put a new drive in with fresh os install, do all your stress testing/gaming, whatever is making it lockup the most frequently (you said mostly while idling??). If that fixes it it's some odd software issue and you just need to format, and if it doesn't, then it sounds like its motherboard... Don't wanna buy a new motherboard if you just need a format.

Only other option you have is to get your hands on more components to do testing with and that's a lot of hassle and likely not feasible.
 
louis, keep in mind just because you ran it within acceptable temps OC'd for 6 months, does not mean you didn't stress the hardware overclocking it.

People seem to think just because it passes stress tests and runs cool that it doesn't damage (long term) your HW to OC. My bet is your MB.
 
Originally posted by: AdamK47
Remember that stress testing thread I created and you posted in? Yeah.
Click to expand...

LinX AND Prime95 at the same time? Why? Wouldn't they fight each other for CPU time, and thus not stress the system out as much as one of them alone?
 
Originally posted by: TidusZ
I think your hardware is probably fine, try taking out both hard drives, put a new drive in with fresh os install, do all your stress testing/gaming, whatever is making it lockup the most frequently (you said mostly while idling??). If that fixes it it's some odd software issue and you just need to format, and if it doesn't, then it sounds like its motherboard... Don't wanna buy a new motherboard if you just need a format.

Only other option you have is to get your hands on more components to do testing with and that's a lot of hassle and likely not feasible.
Click to expand...

i have tried a fresh format already.. the same problem happened with the new install. if i do it again it'd be my 3rd or 4th time since the problem has been happening.
i've tried it with 2 x different hard drives too.
 
Originally posted by: xxceler8
louis, keep in mind just because you ran it within acceptable temps OC'd for 6 months, does not mean you didn't stress the hardware overclocking it.

People seem to think just because it passes stress tests and runs cool that it doesn't damage (long term) your HW to OC. My bet is your MB.
Click to expand...

i am in no way defending that overclocking doesn't hurt componenets even if its within range.
i was actually trying to tell you guys HOW I HURT my pc by running it overclocked for that period of time. so u guys can help me figure out what component MIGHT e hurt my the overclock and stressing period
 
Originally posted by: VirtualLarry
Originally posted by: AdamK47
Remember that stress testing thread I created and you posted in? Yeah.
Click to expand...

LinX AND Prime95 at the same time? Why? Wouldn't they fight each other for CPU time, and thus not stress the system out as much as one of them alone?
Click to expand...

exactly, thats a horrible stress test. why would u run 2 programs that would just fight each other for cpu cycles. in the end, they'd just take a longer time, can give false errors, don't find potential errors and other bad bad things...

and did i post in that thread? link to the thread?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top