System crashing (BSOD) within 5 mins of start up

[foaf]

Hi, my 5 year old custom build has recently developed a massive problem. It is BSODing within 5 minutes of start up, sometimes quicker, sometimes slower. Often it seems like it will crash faster when a few programs are open. It started crashing intermittently about 2 weeks ago, by 1 week later it was constant. The error codes on the Win 8 BSOD seems to be different each time, but include "PAGE_FAULT_IN_NONPAGED_AREA".

At first I thought it was the RAM. I've tried each RAM module in each RAM port and some combinations do make it crash faster (sometimes at BIOS and during startup). One of the modules of RAM seems to work the best (consistently loads Windows), but all of them crash in the end.

Then the network port has stopped working (there are two on the MoBo and both have the same effect). Whether set to auto IP settings or manual, it's not setting a Default Gateway or DNS. So it fails to connect to the network and internet. This behavior has only just started 4 days after the crashing started. I have no idea if it's related, but nothing else has changed.

If not the RAM, then could it be the MoBo? Any ideas on how best to test?

The system is Windows 8 Pro (x64)PC, with wired connection to a router then the modem.
- 3x1Gb Corsair DDR2 XMS2-6400 800Mhz [originally 4Gb and one of the modules fried a few years ago]
- 2.66Mhz Core2 Quad
- MoBo: ASUS P5N32-E-SLI, LGA 775/Socket T
- Graphics Card: Nvidia Gforce 8800 GTS
- HD: Wester Digital Raptor WD15
- PSU: Zalman 1000W [ZM1000-HP]

Any help appreciated!

------------

EDIT: Solution seemed to be:
1. Replace RAM
2. Physically removed Hauppauge WinTV TV Card
3. Power down PC (remove plug & LAN). Boot up, plug in LAN

[MrJim]

It sounds like it could be RAM issues, or possibly PSU related. You could try testing the RAM with Memest downloaded from here: http://www.memtest.org/

You don't mention the model of PSU you have, but if it's 5 years old, it could be on its way out. Do you have another PC you can swap the RAM & PSU with, to see if it solves the problem? It's always handy to have spare bits lying around for troubleshooting

Edit
----
Just re-read your original post - you mentioned one DIMM being fried a few years ago, which could result from unstable voltages from the PSU. If it's a relatively low-spec unit, my guess is that's the culprit...

[foaf]

Hi Jim, thanks for the reply. I have added the Zalman 1000W PSU to the specs above, it wasn't cheap, but as I understand it the quiet design was always more likely to result in errors down the line. I don't have a spare PSU with the same power connections unfortunately.

Today I have tried memtest, but can't get it to boot from USB which is annoying. The PC also seemed to crash more when hot.
Tempted to buy a new PSU to see what happens.

[MrJim]

Well, it's certainly not an under-powered PSU. I would have thought the Zalman PSU to have been of reasonable quality, but you never know.

It's possible of course, that the instability is due to software problems - have you installed any new programs or drivers recently? You could try using System Restore to revert back to an earlier state in order to see if that helps? It's probably a major hassle, but before buying another PSU, you could always try re-installing Windows...it'd be cheaper, at least

Regarding memtest - you can always burn it to a CD & boot from that, if you have a CD writer?

Edit
----
A few further thoughts: Is your system overclocked? Have you check obvious things like air-flow (the CPU cooler fan, system fans, etc)? You could check your BIOS settings and make sure the voltages to the CPU, RAM etc are all correct? What are the temps like in your system both when idle, & under load?

[big_hairy_rob]

If you have a spare hard drive, consider doing a fresh install and see if this experiences the same problem. This could also be a hard drive on it's way out, but usually you would be able to hear that as the majority of fails have been mechanical

If this isn't software then I would suspect that if this is happening after a certain point (after boot), then something either requires getting warm or power and takes that long to actually replicate the fault, either memory, cpu, north/south bridge(mobo) hard drive or gpu would be the most likely culprits. I suggest also borrowing a different GPU and ram and try and replace a part at a time until you think you have located the failed module.

[big_hairy_rob]

If you have a spare hard drive, consider doing a fresh install and see if this experiences the same problem. This could also be a hard drive on it's way out, but usually you would be able to hear that as the majority of fails have been mechanical

If this isn't software then I would suspect that if this is happening after a certain point (after boot), then something either requires getting warm or power and takes that long to actually replicate the fault, either memory, cpu, north/south bridge(mobo) hard drive or gpu would be the most likely culprits. I suggest also borrowing a different GPU and ram and try and replace a part at a time until you think you have located the failed module.

[Willzzz]

Use HCI Memtest to check memory in windows.

[shaithis]

Are you sure 1 ram module behaves differently to other? Especially if you have tried them on their own in the same slot each time.
If you are, then then that is your avenue of investigation. They should behave the same, which would imply at least one of them is faulty.

If they are more random than you are implying and it does happen faster when hotter, I would suspect the CPU, which is where the memory controller is located.

[flearider]

if it's when the pc is getting hot!! then try reapplying your tim arctic silver or ceramic does not go hard .. could be that .. clean all fans ..
take everything apart and put it back together .. clean behind the board .. use just one stick of mem does it still happen use the other .. try error checking hard drive with bad sectors checked ..

[cptwhite_uk]

http://msdn.microsoft.com/en-us/libr...(v=vs.85).aspx

Suggests it's memory related. Unfortunately this could be CPU / GPU or the L2 cache on the Processor according to the link:

Cause
Bug check 0x50 usually occurs after the installation of faulty hardware or in the event of failure of installed hardware (usually related to defective RAM, be it main memory, L2 RAM cache, or video RAM).
Another common cause is the installation of a faulty system service.
Antivirus software can also trigger this error, as can a corrupted NTFS volume.

My money is on the GPU, it's nVidia 8 series which had reliability issues. Since it's one of the easiest things to check I'd try and borrow a known working GPU. Given the crashing nature it's probably safer to switch the cards before attempting to uninstall drivers (disconnected for the internet to stop Windows auto-installing drivers on load up).

The system is also 5 years old, so unless you've cleaned it since then could probably do with a thorough cleaning anyway. Blow out all the dust, specially from the GPU and CPU heat sinks.

[jigger]

As the network ports have gone down and you get random but regular crashes I would bet the nvidia 680i chipset has burnt out.

[foaf]

Hi everyone OP here. Sorry I've neglected this thread, I've been actioning some things...
1. Tried to reinstalled Win 8 Pro, the system crashed before getting anywhere.
2. Replaced RAM modules with 2 x 1Gb (similar spec)
3. Reinstalled Win 8 Pro - system is now stable, but dual LAN ports do not work.
4. Reinstalled Win 8 one more time to double check LAN ports were not a software issue.

The network ports seem to have a DHCP error, registering an IP address of 169.x.x.x regardless of what I do. After researching and trying everything I found, it won't work. So a wireless adapter is on the way to me. I am confident the wireless will work as tablets, laptops and mobiles are all using the wireless on the same network currently.

So, with the above in mind, I am thinking that the PSU has somehow damaged 3x1Gb RAM modules and the LAN ports. Does this sound likely? Is the PSU likely to do it again?
(NB: The PSU is connected to a power spike protected [UK] plug socket, so I also think it's unlikely to be anything related to the mains electricity)

Thanks so much for all your replies, I really appreciate it.

[MrJim]

Probably an obvious question, but have you checked that the ethernet adapter drivers are installed properly? They should be available on the Asus website. You should check the Device Manager & see if it's reporting any errors? The IP address problem is a strange one, as the network should default to obtaining the IP address automatically via DCHP. If you're sure the ethernet adapters are damaged, you could always add another one like this:

http://www.scan.co.uk/products/dynam...i-express-card

It's difficult to say for certain whether your PSU caused the apparent damage to your RAM & LAN ports, unfortunately.

[aramil]

Easy way to test the lan ports is to boot off a live linux disc and try accessing the network.

[MrJim]

I found this thread on the Tomshardware forum which relates to Windows 7, but might be worth a try:

http://www.tomshardware.co.uk/forum/...-port-detected

They suggest unplugging the ethernet & power lead, switching on the PC & allow it to boot up, & then plugging ethernet back in. Might be worth a try?

Also, is your BIOS up to date? You could check that both network adapters are enabled in the BIOS.

All obvious things I know, but sometimes it's the obvious things that are the cause of such problems

[foaf]

Back again with a few more tries:
1. Removed the LAN ports from Device Manager. Windows seemed to just reinstall them immediately.
2. I've flashed the BIOS with the latest update and still no changes. But I did turn on a LAN test setting in the BIOS and both ports register the following error message on boot:

*** Warning: LAN Cable Fail ***
Please check the LAN cable has been connected to Onboard LAN correctly

Just in case I tested with 3 different Cat6 LAN cables (which I know to work on a personal server).
3. For fun, I installed Windows 8.1 Consumer Preview, again, no changes.
4a. In device manager I had a Hauppauge WinTV Nova-T 500 DVB T reporting an error, I uninstalled this and physically removed it.
4b. I came back to this thread, and noticed Jims simple power down idea. I shut down the PC, removed the power lead and LAN cable. Left it off for over 30 seconds and pumped the power button 6 times (I read this might remove voltage from the MoBo). I powered on the PC, went into windows and then plugged in the LAN cable.

Bingo, the internet came back on

So I am pretty damn happy, and thanks very much to everyone, but in particular Jim who's been replying at every stage with ideas.

Just as a final summary, this still makes me think that both the problems were to do with the PSU in some way. However, the TV Card does seems to be causing an issue, as trying to use it again resulted in the LAN ports failing. I hope this thread can guide anyone else with similar issues. Thanks everyone!