Solved Random BSOD - error 1E

[tgmorris99]

As expected, no virus or other unwanted infections.

At the moment it appears that removing Intel Rapid Storage has fixed the problem as I've got all the items I normally let run from Startup running and no BSODs after forcing many, many automatic backups over several days. I'll let this go a bit longer before marking it finally solved.

 

[Arc]

As expected, no virus or other unwanted infections.

At the moment it appears that removing Intel Rapid Storage has fixed the problem as I've got all the items I normally let run from Startup running and no BSODs after forcing many, many automatic backups over several days. I'll let this go a bit longer before marking it finally solved.

Both are good news.

 

[essenbe]

That's great news. I am glad to hear it is running right for you. If you have further problems, we will be right here.

 

[tgmorris99]

Sad to say I had another BSOD on a wakeup. The machine had done a normal sleep on it's own from being idle and I went to wake it up and was greeted with the bios screen.

I'm really not sure what to think this time around but with all the crashes I'm not sure that something didn't get corrupted as I had several errors in the event log related to .NET errors which seem to have been fixed by running the repair tool. However I'm probably going to do another reinstall this weekend and bypass the Intel Rapid Storage this time around.

 

[essenbe]

This is all it's telling me

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************
[COLOR=red]MEMORY_MANAGEMENT[/COLOR] (1a)
    # Any other values for parameter 1 must be individually examined.
Arguments:
Arg1: 0000000000000411, The subtype of the bugcheck.
Arg2: fffff6800000b608
Arg3: 8bb00001b75d0a84
Arg4: fffff8a001617ef9

If we haven't already Please download CPUz (CPU-Z CPUID - System & hardware benchmark, monitoring, reporting), install it and post screenshots of the Memory tab and The SPD tabs. On the SPD tabs, in the upper left you can select the Dimm Slots, I just need the ones with ram in them. This will tell you how to post screenshots, if you have forgotten, which I doubt.
http://www.sevenforums.com/tutorials/9733-screenshots-files-upload-post-seven-forums.html

The reason I asked is if you will look at the screenshot you took of memtest in Post 7, it shows your ram running at timings of 11-11-11-28. The Ram you have is listed as 9-9-9-27 at 1.65V here HyperX 4GB 240-Pin DDR3 SDRAM DDR3 1600 Desktop Memory Model KHX1600C9D3/4G - Newegg.com. We have had Memory Management and Memory corruption several times in this thread.

I'm also unsure of why you would wake up to the BIOS screen. That just doesn't sound right to me.

 

[essenbe]

OK, I researched it a little bit and it will run at the timings and frequency you have it at @1.5V or run at 9-9-9-27 at 1.65V.

 

[tgmorris99]

OK, I researched it a little bit and it will run at the timings and frequency you have it at @1.5V or run at 9-9-9-27 at 1.65V.

^this

I had to physically power on the computer this morning and had another BSOD. Dang this is frustrating. The event log doesn't show anything from last night other than waking up and shutting down and the backup completed normally.

 

[Golden]

Steve, can I take a wild guess based on something I have been reading recently?

Start Menu\Programs\Samsung Magician	Public:Start Menu\Programs\Samsung Magician	Public

Uninstall Samsung Magician and reboot. Monitor for further BSOD's.

 

[tgmorris99]

Magician is not running - unless it's hiding as a service as I removed it from the startup in msconfig and also disabled it in the task scheduler as it would start anyway even if disabled in msconfig. Also, RAPID Mode has been turned off.

I'm beginning to think that this last set of problems are related to it being a PSU problem in that it's not coming up to voltage in the proper amount of time and the faster HW is exposing the situation where the old mobo/CPU was a bit slower. Since the problem does seem to be related to waking up and somewhat random yet everything tests fine when running so that's where I am now - though I'll likely still do a reinstall once things are sorted.

 

[Arc]

Please disable Search Index.
Index - Enable or Disable

It will have a cost .... some functionalists of the library will not be accessible. But we want to see how it goes with search index turned off, as disabling IRST gave it a bit stability.

Let us know for any further BSOD after doing it.

Also, uninstall Microsoft Securities Essential. Get a fresh copy of it and install it.

 

[tgmorris99]

Please disable Search Index.
Index - Enable or Disable

It will have a cost .... some functionalists of the library will not be accessible. But we want to see how it goes with search index turned off, as disabling IRST gave it a bit stability.

Let us know for any further BSOD after doing it.

It's turned off on the SSD already. So you want me to turn it off on the HDD as well, correct?

Also, uninstall Microsoft Securities Essential. Get a fresh copy of it and install it.

It's a new install after I uninstalled AVG 2 weeks ago so it's likely the same version but I'll do it anyway.

I'm still betting it's either a rise time issue on the PSU at this point or corrupted files from the earlier crashes. I'm starting to search for a replacement PSU just in case.

 

[Arc]

It's turned off on the SSD already. So you want me to turn it off on the HDD as well, correct?

Correct.

Also, uninstall Microsoft Securities Essential. Get a fresh copy of it and install it.

It's a new install after I uninstalled AVG 2 weeks ago so it's likely the same version but I'll do it anyway.

May be the same version, but a clean install will be needed, because it is failing to get the updates. Perhaps a fresh install of the program will solve the issue.

I'm still betting it's either a rise time issue on the PSU at this point or corrupted files from the earlier crashes. I'm starting to search for a replacement PSU just in case.

Good idea, obviously. I still go with the second option.

 

[essenbe]

Are you on a UPS? If you are, take it off and plug the computer into a wall socket, just to test.

 

[tgmorris99]

I decided to start over again and reloaded things from scratch. This time around I did not load Samsung Magician or the Intel Rapid Storage software. After successfully installing all of the Windows updates I did a full system backup to my WHS 2011 server (which took what seemed like forever) I finally let the system go to sleep.

When I went to wake it up an hour later (via the mouse) I got a black screen so rather than force a reboot I decided to see if it would go back to sleep on its own - which it did. When I tried to wake it again via the power button (wouldn't wake via the mouse) it produced the attached dump.

Indexing is turned off on all the drives. Also, I'm currently plugged into the wall instead of the UPS so that's out of the picture. If there's no clear reason in the dump file I'll put in the new PSU in the hopes that may fix it.

ETA: essenbe, I see in another thread you mentioned Haswell certified PSU. Is it possible my situation is related to that as well?

 

[Golden]

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 9F, {3, fffffa800686ea10, fffff80001b5a4d8, fffffa8008f58ca0}

Probably caused by : pci.sys

Followup: MachineOwner
---------

0: kd> !irp fffffa8008f58ca0
Irp is active with 4 stacks 3 is current (= 0xfffffa8008f58e00)
 No Mdl: No System Buffer: Thread 00000000:  Irp stack trace.  
     cmd  flg cl Device   File     Completion-Context
 [  0, 0]   0  0 00000000 00000000 00000000-00000000    

			Args: 00000000 00000000 00000000 00000000
 [ 16, 0]   0  0 fffffa800686ea10 00000000 fffff88000d765e0-fffff88000d7a3c4    
	       [COLOR="Red"]\Driver\pci[/COLOR]	ACPI!ACPIBuildRegOnRequest
			Args: 00000000 00000000 00000000 00000002
>[ 16, 2]   0 e1 fffffa800686e7f0 00000000 00000000-00000000    pending
	       [COLOR="red"]\Driver\ACPI[/COLOR]
			Args: 00014400 00000001 00000001 00000002
 [  0, 0]   0  0 00000000 00000000 00000000-fffffa800841c7e0    

			Args: 00000000 00000000 00000000 00000000

Is Windows 7 fully updated?

 

[tgmorris99]

Is Windows 7 fully updated?

Yes, all the updates completed successfully.

ETA: OK, I've disabled the C6/C7 states in the BIOS which, if I understand correctly, should eliminate the PSU incompatibility related to Haswell if that's it.

 

[essenbe]

I really can't answer as that power supply is no longer made, so it is not on Cooler Master's list of 'Haswell Ready' Power supplies. But, most are. The only issue, even with the ones that are not is, when entering idle, sleep or Hibernation, most systems have always gone down to around 6 Watts, Haswell will go down to somewhere around 1 Watt. On some older power supplies, when it goes that low, it causes the PSU to think the system is off, or won't wake from sleep. Sound familiar? Try hooking an extra fan directly to the power supply so that even in sleep the fan will run. That should keep the current high enough to solve the problem. If that works (and supposedly does) you will know that is your issue.- Your power supply is not in production any more, so it is difficult to find any current information on whether it is or is not. It looks from the dates I saw on some of the articles about yours, it hasn't been made since before Haswell was released.

 

[tgmorris99]

I've already got a couple of fans that run off the PSU molex connectors but they turn off when the system enters sleep mode. If disabling the C6/C7 mode in the BIOS works then I guess I've also got my answer. It's just here's no real way to prove it except for time.

If that does solve it I'll swap in the new Antec 620C and move the CM supply and my old mobo into a spare case along with the other HDDs since I know that combo works. It'll make a decent spare system.

 

[essenbe]

As best as I can find out, the only difference is going from 5-6W at idle to 1 W. so 4-5 Watts savings is all you are talking about with a Haswell ready VS a not Haswell ready. If disabling C6 and C7 works, it's up to you. I wouldn't worry about it if it works.

 

[tgmorris99]

Setting the C6/C7 sleep to disabled in the BIOS didn't stop the problem so I just swapped the PSU. Now it's a wait and see situation. At least now, other than the case, all of the HW (mobo, SSD, HDD, memory, PSU) is new. If it still happens about all it can be is an incompatibility somewhere, or a bad mobo.

ETA: One things that's definitely different is that it now sleeps almost instantaneously (<2 secs) when manually putting it to sleep. It wakes up a lot faster as well. Hmmm.