New build BSOD and crashes, help diagnose?

[eaclou]

I recently built a new system with the following:

Asus P7P55D
i7 860
G.Skill 4x 2GB 1600mhz 9-9-9-24 cas 1.50V
Radeon 5770
WD Caviar Black 1TB
WD Caviar Green 1TB
Windows 7 Professional 64-bit

I've been having occasional blue-screens or hard crashes, and am having trouble figuring out what the issue is, as they're seemingly random but always happen at the same point in start-up. So i'm unable to reproduce the problem whenever I want.

The first time it crashed was after I had installed Windows on the Caviar Black 1TB, and added the 1TB Green to the system. I got a blue screen one or two times, so I unplugged the caviar green and it booted to windows recovery mode. I restarted and got into normal windows, then turned off the computer and tried adding the caviar green again - this time it worked fine.

I got a similar crash when i plugged in an external hard drive while the computer was off. During startup, i was getting (and still get) a message that the JMicron controller did not detect any drives. This is after a quick flash of other text that i am not quick enough to read, so maybe the no drive found was referring to one of the empty SATA ports, i'm not sure.

So I'm thinking it's a problem with adding new drives, but then I had the same issue after ZERO hardware changes, and on boot-up, sometimes before the ASUS splash screen, it will display a mostly light-blue colored screen with yellow and green dashes that look like it was corrupted or each horizontal row of pixels was shifted left or right - i'll see if i can get a photo of it if it happens again.

I then updated the BIOS to the newest version, but after that, on one start-up, I get a message saying the SixEngine had a problem, and that windows would restart in 60 seconds.

Because of the bizarre corrupted-looking image before the normal ASUS splash screen, and the message from SixEngine, i'm suspecting the motherboard, but is it a hardware or a software issue?

ALL crashes and blue screens (with the exception of the sixEngine windows automatic restart) occurred at the same moment during start-up: Right at the end of the windows loading screen (with the 4 colored dots that form the windows logo) when the screen normally goes black and you see the cursor in the middle of the screen. On restart, i get a message saying windows failed to load, and a choice to start normally or open the recovery mode.

Any ideas? Is it worth re-formatting?

Thanks everyone.

 

[masterasia]

Check your ram timings.

Make sure it is at 1600Mhz 9-9-9-24 with 1.50V.

If you are overclocking, change it back to stock.

Try uninstalling the EPU/six engine thing. You really don't need that unless you're planning to go "Green". If you're overclocking, you definitely don't need it. I hardly ever install the utilies that come on the driver disc. Most of them are useless.

 

[silverron]

Have you checked out Asus website for updates on drivers? I assume since you updated the BIOS, you have the drivers too? Have you called Asus tech, they might have some answers, 64 bit and Win 7? If you purchased Win 7 on retail not OME, you could make a call to them also. As you said, could be a number of issues and might need Pro help. Any comments welcome - Good Luck!

 

[eaclou]

So last night I tried powering up again, and it got stuck on the windows loading screen (I gave it about 6-8 minutes, to be safe, then gave up and restarted, as it normally takes 15 seconds or less).

On restart, I tried ignoring the recovery mode and tried loading windows normally -- it got stuck at the loading screen again, so i restarted and opened recovery mode. In recovery mode, i did the Roll-back settings option, and when it booted again windows worked fine.

I left the computer on for about 5-6 hours, and it never crashed or anything (It's never crashed once it's made it to the windows desktop - with the excpetion of the SixEngine message, but that wasn't exactly a crash.) So the CPU at least must be stable enough. When it booted successfully, there were no errors with SixEngine.


I'll try making sure all drivers are fully up-to-date, and check the RAM settings (I haven't overclocked yet, was planning on a mild one but wanted to get all kinks worked out before doing that.)

It's getting stuck on the Windows loading screen is a problem I had with a computer like 6-7 years ago and I THINK it was a hard drive about to eat it, but i'm not sure. Do any of these problems sound like they could be hard drive related? I want to start narrowing down the list of possible trouble-makers, so far I think it is NOT a problem with:

PSU (really doesn't seem power-related to me, and I have a top-quality Corsair 750HX
DVD (I guess crazier things have caused problems in the past, but i severely doubt this is causing anything)
Heat (It's very cool in the case, i can feel the exhaust, and the crashes happen on startup, never after running for awhile, so it couldn't possibly be too hot)
Graphics Card (It doesn't feel like an issue with the 5770)

So that leaves:
Mobo
RAM
Hard Drive
Software

I'd love to be able to rule out a Hardware problem, any ideas on how to do that if i don't have a spare 1156 Mobo? Also, could scratches or bumps during installation of the motherboard cause problems like this, or is that a more unlikely problem? I guess it's possible that I could have damaged it physically installing the heat sink or something?

Thanks for the help guys, hopefully I can get this worked out with your assistance.

 

[masterasia]

It sounds like a hard drive problem to me. Have you tried running a check disk?
If it's not the hard drive, then I would do clean install and see if the problem is still there.

When Windows is loading long it could be that it can't find the files to start up, or it can't start services and is waiting for a timeout.

Just last week, I was getting a BSOD on a clean install of Windows 7 Home. I ran tests on the hard drive and memory and everything came back negative. I noticed that sometimes when it booted, it would go to the "Safe Mode" selection screen, because it would fail to boot after the Windows loading screen and restart its' self. Here's what I did: I grabbed another hard drive, cloned it, and popped it in. Since then, no more problems.

 

[eaclou]

So i did a bit more testing (more to do tonight).

At first I tried just getting into windows without changing anything, but sometimes going into recovery mode. After 5-6 restarts and it failing in the same place I gave up on that, but it was good to make absolutely sure it was consistently messed up.

So I opened up my case, disconnected the SATA and power cables for the 1TB Caviar Black that had the windows installation on it, popped in the Windows 7 disc, and installed windows on my backup drive, the 1TB Green.

It worked perfectly, and I haven't had a crash yet booting from the Green drive, but I ran windows memory check and it said it found hardware problems (but, despite telling me it would show them to me after it restarted, it didn't, so i have no specifics.)

I spent the night further backing up my data, so i haven't had time to test each RAM stick, or even booting from the Caviar Black (original boot drive) to see if it still has issues.

When i was in windows, i tried d/ling memtest86 and running it, but I couldn't install because it said it couldn't run in a 64-bit environment or something...?

I think the problem is down to one of the following:

Hard Drive (caviar Black)
RAM
Motherboard
conflicting drivers

I HOPE it's not the motherboard, but it's looking less likely that it is, so that's good.

When I get home tonight i'll try testing the RAM sticks individually with windows memory diagnostic.

Any other ideas, or ways to rule out problems? It's possible that BOTH the RAM and hard drive have problems, so i don't want to assume it's only one component.

Thanks for the help so far

 

[masterasia]

You should run Memtest86+ from a boot disc. Don't run it installed on Windows. Just download Ultimate Boot CD and use that. http://www.ultimatebootcd.com/

 

[shortstuff_mt]

+1 to running Memtest86+. Masterasia is right, Memtest does not run inside Windows. You boot from a CD and it runs before the OS loads.

It certainly sounds like a RAM problem. That's not a surprise, since RAM issues are easily the most common cause of BSOD's. It would seem that the faulty RAM has corrupted your Black drive. Once you get your RAM issues solved, I would format your Black drive and do a fresh install on it.

 

[eaclou]

Ok, i made a memtest86 boot CD (version 3.5) at work, so when i get home I can try that out.

I'm just a little stumped why, if it's a memory problem, that I wouldn't have any trouble with my Green drive... oh well, i guess stranger things have happened. Hopefully i'll have good news to report after some thorough troubleshooting tonight.

Thanks again everyone, these forums are far better than most phone tech support.

 

[eaclou]

Well, i really thought i was going to be able to put this to bed last night, but computers are complicated machines, i suppose. Here's the update:

I plop in the memtest86+ CD, and let it run through all 8GB of RAM one time through (took about 45mins). No errors.

So, 'Harddrive!' i think. I unplug the Green backup drive (That i still haven't had any problems with) and try to boot windows again from the original Caviar Black drive. It had still remembered that the last time I had booted it there was a problem, so it loaded up in recovery mode. I let it do its stuff, and it restarted... you guessed it, without a hitch? (Remember the last 4-5 boots it didn't work, and i didn't change anything on that drive).

'Screw it' i thought, and reformatted and reinstalled Windows 7 on it. The first time it was installing, I left the room to start dinner, and when i came back it was on a blue screen with a message about ntfs.sys Amazingly, when i try to install a second time (watching it the whole time now) it works flawlessly.

So i'm in windows now, on the caviar black drive, and i download Western Digital's disk diagnostic (DLG or something, i forget the exact name.) and run the extensive test overnight (took 3 hours or so). Passes with flying colors.

Before i left for work today i started another memtest86+ session and let it run, so when i get back it should have been doing that for a good 10+ hours, we'll see how it looks then, to rule out a RAM issue.


Basically, I'm kind of stumped now. It's looking like either the motherboard or some software/driver issues. Is there a way to test the motherboard to make sure it is fine? That's what i'm most concerned about because a MB RMA would be a pain in the arse compared to Memory or HD.

If it's software that's fine also, as eventually I can find a working combination.

Any thoughts?

 

[masterasia]

Your ram could be fine. Some ram need their timings set manually to run.

Try installing on the Green drive to see if you still get a BSOD.

You're not bending the SATA cables too much? SATA cables aren't meant to be bent.
If you have different SATA cables, try those, it's worth a shot.

 

[eaclou]

masterasia, i'm giving you best answer since you've helped out the whole time.

Anyway, I ran memtest for 14 hours straight, and it came back with 1 error. Is this an acceptable error rate? I don't want to pretend that these things are totally flawless, so i would imagine a certain fail-rate is acceptable, but i'm not really sure what that is.

I reinstalled windows on the Black drive and so far, no problems at all. I'll be keeping a watchful eye though.

Thank you very much for the help during this process. This is the second build i've done and the first one barely counts because I basically got lucky with it didn't even really know about the different CPU sockets, for example, i could have easily gotten incompatible parts.

 

[eaclou]

So I booted the Caviar Black drive up yesterday and it blue-screened with the message IRQL_NOT_LESS_OR_EQUAL. I reboot, run recovery mode, it loads windows fine. (I now have windows installed on both drives, for now - still haven't had an issue with the Green drive's windows)

I restarted it a few times, then tried shut down, let it sit for a few minutes, then turned it back on. Did this about 6 times - every time it loaded windows without a problem. So I installed World in Conflict and played for about 2 hours, no problems. turned it off, booted up the Green drive, ran the check disc utility on the Black drive - no errors.

At this point I'm 90% it's Software related, but I can't pin-point which drivers might be causing trouble. I find it hard to believe that it's hardware-related after running it for over a week and it NEVER crashing anywhere other than during Windows start-up.

I'll see if it works tonight. I hate these inconsistent errors, much harder to find.

 

[masterasia]

I'm just curious. Is your Windows 7 an OEM or RTM or Retail?

 

[newfireorange]

RTM 7600 Final release Windows 7 Ultimate 64 bit runs perfect on my machine so I don't think it would make a difference on any other powerful computer. Unless that is you somehow got a corrupted copy of that Windows 7. Do you have any other operating system you can install to test out your system? XP or Vista for example. Try installing Windows 7 32 bit for testing purposes.

 

[masterasia]

I have the Enterprise version and it's still not stable.

 

[newfireorange]

When you go into recovery mode and get your OS up from there, how long are you able to run without any problems?

 

[newfireorange]

I'm thinking your two hard drives have a lot of bad sectors in them. Do you have any other hard drive you can use?

 

[newfireorange]

Masterasia, is it just my system, or have you noticed Windows 7 sometimes will hang. You go to click a program from the taskbar and it will take like 5 minutes to load. Does the same thing to me even after a brand new format and install. Doesn't seem as stable as Microsoft wants us to believe.

 

[eaclou]

I have Windows 7 Professional x64 OEM.

When i installed it on the green drive, I still haven't had any problems with that installation, so i doubt it's the windows DVD being bad.

I've ran the offending hard drive through the manufacturer's disc-check program's 'Extended' test twice, and both times it didn't find anything wrong.

Memtest86+ ran for 14 hours with a single error - somehow I doubt that is the cause.

Which leaves motherboard and drivers, but if it was the motherboard, why wouldn't it fail at any time other than Windows start-up? I've had the machine on overnight a few times now and when i get up in the morning, there's no blue screen.

I'm going to watch it now and see what happens - it's possible that if it IS drivers, that windows has sorted out the problem and i might never get a crash again (due to this particular issue, that is).


as for your question, newfireorange, after Windows recovery mode does its thing and i'm back up and running, i'll get anywhere from 1 to ~16 (and counting) flawless boots. I have been unable to notice a pattern as to when it blue-screens.

 

[shortstuff_mt]

I would RMA your RAM. Properly functioning RAM will get NO errors running for multiple passes. I consider RAM faulty if I even get one error.

 

[eaclou]

hmm, what's a typical fail rate, then? Surely these things aren't infallible, especially non-ECC RAM.

I.E. how many errors per 100 passes would be acceptable? If it had only run 13 times through, it might not have had that error, and would thus appear perfect -- about how many passes is enough to feel confident the RAM is good?

Also, it said the error occured around the 2700MB area - is there a way to figure out which stick this was, or do i have to run each stick individually?

Thanks

 

[dndhatcher]

RAM doesnt work like HDDs where they can mark a sector bad and stop using it. Any RAM error is fatal every time you hit it. I believe you have to run the RAM individually to find the bad stick.