Windows 7 x64 Crashes on Startup When Powering On System

Specs:
Mobo: Asus M4A89TD Pro USB3
RAM: G. Skill Ripjaws x2 2GB
CPU: AMD Phenom II Black x4
PSU: Corsair HX 650W
HDD: Western Digital Caviar Blue 500gb SATA 6.0 x2
GPU: Sapphire Radeon ATI HD 5770 PCIe
Case: Cooler Master CM690II Advanced

I've built two computers in the recent past, one for myself and one for my boyfriend. I've noticed that when shutting down my boyfriend's machine, and powering it back on, Windows will crash to a BSOD or get stuck on a black screen. The suspicious thing is, this will ONLY happen when shutting down and powering back on. If the system is simply restarted, Windows will start up every single time without a crash. Restarts don't cause problems, only powering on.

Usually it will hang right on the splash screen and not advance beyond that point, and after a short while after, it will usually crash to a BSOD or black screen.

Strangely, his computer will not record the BSODs that occur when they happen at the startup logo. This means that I cannot post the analyzed BSODs here with BlueScreenView or any other program, so PLEASE DON'T ASK ME TO. They do not show up on any BSOD reader. The crashes DO NOT appear in the minidump folder. Please DO NOT ask me to supply the minidump folder.

I have run a Memtest on each individual stick, each with 6 passes with zero errors.

I wrote down the most recent errors that popped up when his computer crashed on starting up, since I can't look at his blue screens otherwise. The first said SYSTEM_SERVICE_EXCEPTION, BCC was 3B.

The second crash didn't give an error name/message, it just gave a STOP error ending in 7F and the related file was ahcix64.sys, which I have seen multiple times on various BSODs on his system, as well as my own.

Another recent one was DRIVER_IRQL_NOT_LESS_OR_EQUAL with Stop error D1.

ntoskrnl is also a file I see frequently during BSODs.


Each time the system is reset after powering it on, it gets farther and farther in the startup process until eventually it will give in and start up without a crash. Once Windows is up, it is, strangely, stable. Games will run perfectly without errors. Programs will function normally. Performance is exceptional, and the computer is very fast. It will run perfectly for weeks without a single crash, hangup, or problem.

I have not done a startup repair yet. The main reason for that is, his hard drives are set up in RAID 0. I'm afraid that it may damage the array somehow. If I get reassurance that it won't, I might try to fix it.

All of his drivers are up to date. All of his programs are up to date. His Directx is fully functioning with no errors. His Windows 7 is fully updated. Everything else seems to be A-Okay, except for the startup problem.


I also scanned his PC for viruses and he doesn't have a single one.
This problem was happening long before I had installed anti-virus software on his machine. Anti-virus software is not the culprit.

I've also cleared his CMOS.

I've already done a SFC Scan and no corrupt files were found. I can't run a chkdsk because he has RAID 0.

I have cleared all of his temporary files.

I have defragmented his disks.

Knowing I've taken all of these measures... please don't ask me to repeat them. I've done everything I can. I've downloaded every tool I could to help. I have no answers.

A most recent development has me thinking it could be a hardware problem. When touching the front/top USB ports on the case, the display crashes to a black screen. The rear USB ports don't cause this to happen, only the top/front ones. The USB ports are directly next to the power and reset switches.

I can provide System Info and Event Logs as needed, if some of you want to actually help, or even can help if your knowledge is applicable. I've been posting in other forums for over a month now and I get very little help, in fact I've received more questions than answers.

 

the only time I've seen 0xD1 later turned out the computer had an infected mbr. not saying you do, just speculating.

can you boot into safe mode?

if you can,

try this:

Download MBRCheck to your desktop

• Double click MBRCheck.exe to run (vista and Win 7 right click and select Run as Administrator)
• It will show a Black screen with some information that will contain either the below line if no problem is found:
  • Done! Press ENTER to exit...
• Or you will see more information like below if a problem is found:
  • Found non-standard or infected MBR.
  • Enter 'Y' and hit ENTER for more options, or 'N' to exit:
• Either way, just choose to exit the program at this point since we want to see only the scan results to begin with.
• MBRCheck will create a log named similar to MBRCheck_07.16.10_00.32.33.txt which is random based on date and time.
• Attach this log to your next message. (See: HOW TO: Attach Items To Your Post )

 

Yes, I can successfully boot his computer into safe mode.

Since I'm not at his house currently, I'll have to do all of that when I see him again. We live kind of far apart, so I have to wait until this weekend to do some more troubleshooting.

What's MBR?

 

It stands for Master Boot Record.

 

Ah, okay, that actually makes a lot of sense. I'll definitely make sure to check that when I see his computer next.

Thanks for the helpful advice and I'll post again whether I find a problem or not.

 

Hello,

I have exactly the same issue but this on a new build (totally different hardware then yours) of 1 and half weeks old, however the first two days it worked fine. This gave me the idea it might be an update that is the problem.

In short: Restoring doesn't solve the problem when allowing windows to update again but does if you force it not to update. So far it remained stable (1 day with several power downs).

To exclude the MBR (didn't think of it cause I tried two different HDD/SSD's with same issue) I'll check this evening and update this thread for you.

 

To add to my last post (sorry the edit button is not there).

Some people claim (see the net and a friend reported this also) that a similar issue occurs due to the Catalyst 11.6 drivers. So if you troubleshoot it might be wise to roll back to 11.5 drivers which are claimed to be stable.

I want to point out however that rolling back to the 11.5 drivers didn't solve this issue for me but it did for my friend but it is very hard to compare issues of which BSOD info isn't stored.

 

These crashes on his particular machine have been happening long before I gave him the latest driver. Before, he had 11.2, I believe. I only gave him the latest one just last week. I haven't tried shutting down and powering on since then, but he said yesterday after powering on that he could at least enter safe mode without a crash so that he could restart into normal Windows.

And since these computers, mine and his, are both about five months old, I don't think I can restore before the updates, and he's done many things on his machine since those updates that he might lose. Kind of an annoying process, really.

If it's an update that caused it, maybe it's possible that a startup repair might fix it? I haven't tried that yet.

 

It's quite possible that you each have different reasons for what appears to be the 'same' problem, especially as the hardware is "totally different" and should be in separate threads so as to avoid any confusion.

Please ensure that the computers are set so that they create minidumps and do not overwrite them. Also check that there is a functional paging file on the system drive.

The issue may be hardware related, please give BIOS revisions and exact CPU and RAM specs.

Once in Windows, run DriverView to see which drivers are loaded and check their ages, versions and dates. You can change the output so that only the name, version and dates columns show, then select all non-Windows drivers and save the list to be uploaded for checking for later versions, versions that may be incompatible, etc.

And be prepared for more questions, it's part of the diagnosing process

 

Minidumps are turned on. He still has dumps from over a month ago from problems that I seemed to fix. Ever since then, the crashes only happened on Windows startup and ever since then, it stopped recording them.

What CPU and RAM specs do you mean? Do you mean their clock speed or if I OC'd them or not? They're at their default speed and voltage, and when checking, they were normal.

I'll have to run Driver View when I see his computer next. I'm not around it right now.

 

Check that there is a paging file on the System drive, ok?

Full make, model, serial # etc. - the above is not enough to check with the validated list at Asus and cross-reference against the BIOS revision the 'board is currently on .

 

I'll have to provide those later, I'm currently in between addresses and all of my computer stuff is packed up. But I'll post those when I get a chance to.

 

Here, I at least found the RAM info:

Here's the serial number
F3-12900CL9D-4GBRL

One way I know that the parts are compatible is because a friend of mine is a computer technician and he's the one that suggested these parts to me, and he and some mutual friends have the same hardware.

 

Ok, CPU and BIOS version still needed for cross referencing.

Are your friends all using the same mainboard and BIOS revision as well as CPU, RAM and Windows versions, if not, all bets are off and that's before we move on to drivers ^^.

 

I really don't know.

The BIOS version is BIOS 2001. That's the most recent update to it and he has it, or he used to. I think it reverted back to the original when I cleared his CMOS. The problem persisted after clearing it, and I haven't re-updated it.

http://support.asus.com/Download.aspx?SLanguage=en&m=M4A89TD+PRO/USB3&p=1&s=24

Here is the downloads section for Asus, the OS is Windows 7 64-bit, if you need to look it over.

As for the CPU, I'll have to provide that later.

Also, you'll have to tell me how to check for a paging file.

 

The RAM is not on the QVL, so not tested/approved by ASUS. It might improve compatibility if it was dropped to 10666 but what the timing and voltage would need to be is beyond me.

Clearing the CMOS does not revert the BIOS revision to an earlier level; equally, setting the BIOS to defaults or optimized doesn't mean it'll suddenly work with all components; often there are many more trial and error settings needed to gain stability.

As far as I can tell, all Phenom II x4 CPU's should work with 2001 BIOS.

 

if it helps any, my computer had the exact same specs and I didn't have this problem that he has. Same RAM, same CPU, everything is identical, except that I have RAID1 instead of 0.

My motherboard has died so I have to RMA it, but before that, there were no problems with shutting down and powering on.

 

And all the drivers are identical?

 

They were exactly identical, up until my motherboard died, of course. I've continued to update his drivers myself since then despite my computer dying. But yes, they were exactly the same.

We also both had SP1 and identical updates to the machines.

 

Well, there's not much left except drivers (rollback), file/driver corruption (chkdsk, SFC, reinstall drivers) or malware.

 

I did SFC and it found nothing corrupt.

I scanned for malware, and there isn't any.

Can't run a chkdsk on a RAID array, it could damage the array.

It's more than likely his RAID driver in combination with a hardware problem. I'm going to test some things over the weekend.

 

I attached the txt file. Didn't say anything was wrong with the MBR but the scan might reveal something else.

 

Is it possible that flashing/updating the BIOS may have damaged his RAID array or something?

 

That's not something I would expect to happen.

 

All right, just making sure.

However, I do see his RAID/AHCI drivers aren't updated. They actually used to be the official Asus drivers, but I think since I flashed his BIOS it reverted the RAID driver back to Microsoft version. It used to say AHCI RAID controller but now it says something else and says it's not Asus-issued, but it's Windows-based.

I'm going to update his drivers for that, since I think they're causing some problems. In most of his crashes I see the achix64 file involved.

 

Involved in the crash or just loaded?

If the driver wasn't loaded, then there'd be no boot.

The ahcix64s.sys is the AMD RAID driver.

 

It's hard to say. Because when Windows will refuse to boot, most of the time it crashes to a BSOD and it will mention the RAID driver. It will try to boot and crash. Well, I'm sure I don't need to tell you all of that, it's in my first post.

If it crashes trying to start up in Safe Mode, I've seen it stop after trying to load ahcix64, it's one of the last drivers I see. It doesn't happen often when trying to enter Safe Mode, it's almost always in normal Windows.

The details for the RAID driver says AMD AHCI Compatible Controller Driver, but for some reason that's not what his HDDs are using. They're using some Windows-based one. It says it was last modified on the first of June, and he nor I never modified or changed it then. But I think that was when I flashed his BIOS. That was when BSOD's during system usage stopped, but after that, they began happening when powering on the system, from how it looks on his event log.

 

If you only check the drivers in Safe Mode, you will see the 'fail safe' drivers, not necessarily the drivers loaded during normal Windows boot.

 

Ah, I see, makes sense, then.

 

His computer crashed during gameplay and when trying to start in safe mode it gave the stop error code 7E.

That's the only info I have.

But it just happened a moment ago.

It happened a second time and gave the stop error 1E
Both while trying to enter Safe Mode.

Each time it does this, it's after ahcix64s.sys and storport.sys are loading.

After the third reset, it finally gave in and entered Safe Mode successfully. From there I could restart the machine and enter normal Windows.

 

Note: His system didn't crash, just his game did, which needs to be patched. He had to force a shut down.

 

Hi

As much as the following probably sounds like a royal pain in the you know what, it's also the most efficient troubleshooting strategy. Also, since you've already done a fair amount of frustrating things, I can only imagine the size sledge hammer you'd like to use on anyone giving "random" advice. However, I assure you this isn't random, and could likely help you resolve the issue, or at least have a better idea of what the issue is. One key thing you're missing that I noticed is you're seeing a driver load in Safe Mode before a crash. Unfortunately, it's the NEXT driver you don't see loading that's usually the culprit IF it's actually a driver issue.

Often, after I've spent hours on a hardware/software problem, and I meekly resort to this, it does wonders:

First of all, get a copy of a live Linux distro, and try booting to that. Mint's good, Puppy's not bad either. Any live distro will do. See if you have any issues when using that, and if you still are, then:

Work from basics up. Start with the absolute minimum hardware for the computer to post, i.e. motherboard, CPU, and one stick of RAM. Disconnect everything else (if the motherboard doesn't have onboard video, leave the GFX card in... otherwise, take it out.), INCLUDING everything connected externally (except for USB keyboard and mouse). Once you've ensured that works, add the rest of the RAM one stick at a time. As long as you still have success, add the hard drive (or in this case, the HDD RAID array, though if you have an extra hard drive, it'd be helpful to keep the RAID array out of the equation as of now and add a blank hard drive) and install Windows on it. It doesn't matter if you have a license, as you won't need more than 3 days to test everything with it anyhow... so no need to activate this "temporary installation".

Once you've successfully booted into an OS on the hard drive, add all other hardware one piece at a time, rebooting after each add and testing a few shutdowns/power ups to make sure nothing is wrong. I think the last thing you should add is the RAID array, as while that may be the issue, it could very well not be the only issue. If things work without a hitch WITHOUT the RAID array, after several shutdowns/power ups, then obviously that's the issue to address. If things aren't going right without the array, then you have something else to worry about. Again, take things one step at a time, and worst case, you'll have to reinstall the unlicensed Windows on the hard drive a time or two if it takes longer than a few days to test everything. If you don't have a spare drive to replace the array with, then obviously the troubleshooting's going to be quite hampered.

The major problem with this type of troubleshooting, is that the system will be inop for regular use until you're done.

Another thing I'd HIGHLY suggest, either before or after this troubleshooting method, is to run the hard disk manufacturer's Short and Long test on the array drives to ENSURE the drives aren't faulty. Faulty drives can cause all sorts of weird issues, including constant BSODs, especially if hard boots work better than a "restart", as hard drives don't really completely power down in a restart.

 

Will the tests on the drives work on a RAID array? It won't damage the array at all, will it? Also, what can we do if they are faulty, or corrupted? Would he have to break his array?

 

No, it won't damage the array or affect the data on the disks in any way. And yes, it will work on any hard disk in the system made by the manufacturer that makes the diagnostic software, as it detects the drive through strictly hardware means. It has no ties with Windows, drivers, etc.

It's just simple manufacturer diagnostics. If they are Seagate drives, you'll be using SeaTools, if they're WD drives, you'll be using WD Data Lifeguard. Run the Short AND Long tests on each drive, and if there are any errors, the drive(s) with errors need RMA'd pronto.

This is why I ALWAYS run the Short and Long tests on any new drive I purchase, even before using it.

 

If one of the drives is faulty, will the array get broken? Since he has RAID 0, it's not a mirror, so I don't think his data will copy to the new drive, will it?

 

From Carrona.org:

Memory crops up in both; let's assume it could be Virtual Memory = page file and get that removed, set it to zero then reboot. Then set it for 4092MB min. and max. This should at least allow the creation of a minidump, assuming that the services needed to create it are loaded when it BSOD's.

 

IMO, RAID 0 is always asking for trouble, as all data is split between disks, and if one disk is faulty, the data on the other disk is useless.

However, IF one of the drives is faulty, you can copy the drive data to another drive (same make/model, otherwise the RAID might not work) using something like this and swap out the drives. If you don't have a spare drive, obviously get one and do the data copy before sending the faulty drive in.

We can just hope for now though that his drives are both good.

 

Interesting that remote control is mentioned in both as well.

Also, wouldn't virtual memory be a hard drive issue?

 

Device Driver is mentioned both times, too.

 

Yeah... Microsoft's error code system doesn't exactly help much, unfortunately. :-D

EDIT: Also notice the part we're supposed to skip over -- "USUAL CAUSES" hahaha so obviously there's unusual causes for these error codes as well. ><

 

Yeah, I know it's not much information, and it actually makes me feel at a loss when it's the only information I can provide. Even with dumps enabled and automatic restart turned off, the computer just won't tell me what's wrong. It's a giant pain in the... well, the you-know-what. ;P

I've been troubleshooting this problem for over a month now, and I've been running tests until I'm blue in the face, but I want so much for his computer to work, especially since I'm the one who built it.

Since he's using the machine for the rest of the night I'll have to run HDD tests either later or tomorrow morning, and use Driver Verifier.

 

Hah, yeah. You never know the true cause unless you can analyze the info.

Speaking of analyzing info, I don't know if anyone could detect any issues by looking at his event log or system info, but I could provide that in the meantime.

 

Yeah, that could definitely help.

 

Excellent.
They're evtx and nfo files, and I have them uploaded to the Microsoft SkyDrive, if you're able to use that, if not, I'll have to find a way to copy them into text files or something.

https://skydrive.live.com/?cid=E4DE9A8BE33ED49B&id=E4DE9A8BE33ED49B!126

Here's the skydrive link.

 

Ok... is that event log from April, or is his system clock awry?

 

He should have events all up until early July, unless you mean something else.

You know what? They are different dates on my machine, too. They say that they're up until today for some reason, even though I made that log about a week ago.

But the latest errors/events are from early July, regardless.

 

Doh. It goes from April through July. It's 2:30AM here... I think I need to call it a night before I start making no sense whatsoever! I'll check back in tomorrow. Meanwhile, hope you figure something out.

My gut feeling is that one or both of the hard drives are faulty... which would suck. The event log seems to support that theory in a general sense, but it could also support many other theories so I'm not saying by any means that's what's going on.

Does he notice any other problematic hard drive symptoms? Things like sometimes opening/saving files takes a strange amount of time compared to normal, Windows Updates taking forever or needing a few tries to install, programs hanging at odd times for random amounts of time, etc.

Anyway yeah I'm calling it a night.

 

Thanks for you help and advice
I asked him, and he says "no" to all of the above. The hard drives actually run quite exceptionally. They're very quick. The only time they strain is when starting up Windows after powering on.

Again, thanks for the help, and I'll continue to post here if I get any more information.

 

Good news, HDDs passed both tests. They're clean as a whistle.

 

So I was going to update his RAID drivers because it looks like his system is using that Microsoft one instead of the Asus one, but whenever I try to get it to update with the Asus RAID driver I downloaded, the update screen gets pissy at me and says, basically, "cut it out, that driver is already updated," even though the file I'm trying to give it is completely different.

When looking at the "volumes" tab under the driver information, it's all blank.

Apparently the RAID is using two driver files I've never heard of involved in a RAID. disk.sys and partmgr.sys.

I realize the drivers for Asus are from last year. But I don't think I care how recent the Microsoft one is. I have a bad feeling that the computer is trying to run the wrong driver or even two at the same time for the RAID. Since it won't allow me to update the ahcix64s.sys file, which is the old version from April off of the Asus disk, I can't even determine if the update can fix the problem.