Various errors! The Saga Continues!

[KeyboardDemon]

@satrow: I have PM'd you some links with the requested reports, let me know if you got them ok.

I was looking through the PerfMon report, there are a couple of things that might be pointing towards a solution or they might be red herrings, first there was this:

tv_ConfigMgrErr43
Cause: A device has a configuration problem that prevents it from working properly.
Details: The device, Intel(R) Management Engine Interface , is reporting "tv_ConfigMgrErr43". This device will not be available until the issue is resolved. The Plug and Play ID for this device is PCI\VEN_8086-DEV_1E3A-SUBSYS_84CA1043-REV_04\3-11583659-0-B0.

Then in the Basic System Checks under Hardware Device and Driver Checks I got a red light which means it Failed the Survey of Windows Management Infrastructure supported devices and when I drill down into that section everything listed looks like it has failed.

The report is massive, it makes War and Peace look like light reading, it even makes War and Peace look interesting, lol.

*** edit *** I just drilled down into my Device Manager and found an exclamation mark that I hadn't seen before and guess what, it was for the Intel(R) Management Engine Interface device, so I right clicked it and updated the driver using the Option to "Let me pick from a list of device drivers on my computer".

*** edit 2 ***

After updating the driver I just ran another report and no longer have any failed hardware, about to test now. Fingers crossed.

But thanks all, and particularly satrow as, unless I'm being a bit premature, I think your advice has pointed me in the right direction, I'll post how well testing went soon.

[KeyboardDemon]

Ok. I successfully managed to play two full rounds of BF4 with no crashes or issues, hooray and then got a BSOD with a SYSTEM_SERVICE_EXCEPTION whilst watching a YouTube video.

So I don't know why, but it's still not fixed!

I think I need to go to bed.

[satrow]

Ha - War and Peace - lightweight

Well done on sorting that Intel/WMI bug, MSInfo32 reported it as being a Problem Device: "Intel(R) Management Engine Interface PCI\VEN_8086&DEV_1E3A&SUBSYS_84CA1043&REV_04\3&11583659&0&B0 This device is working properly." I didn't know whether to laugh or cry - first time in reading hundreds of those reports that I've seen anything as contradictory as that!

Ok, amuse bouche over, let's step up to the bigger plates: Bugchecks appear to be GFX/DirectX -related, with a hint of networking.

Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
Debug session time: Sat May 16 11:40:43.506 2015 (UTC + 1:00)
System Uptime: 2 days 4:26:25.804
BugCheck 1E, {ffffffffc0000005, fffff8019102b316, 0, ffffffffffffffff}
Probably caused by : ntkrnlmp.exe ( nt!SMKM_STORE_MGR<SM_TRAITS>::SmFeAddInitiate+16a )
BUGCHECK_STR:  0x1E_c0000005_R
PROCESS_NAME:  chrome.exe.exe
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``
Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
Debug session time: Sun May 17 23:53:21.699 2015 (UTC + 1:00)
System Uptime: 0 days 1:17:23.392
BugCheck 3B, {c0000005, fffff80021afc881, ffffd00023224cf0, 0}
Probably caused by : ntkrnlmp.exe ( nt!KeTerminateThread+23d )
BUGCHECK_STR:  0x3B
PROCESS_NAME:  bf4.exe
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``
Built by: 9600.17041.amd64fre.winblue_gdr.140305-1710
Debug session time: Sat May  9 23:05:44.223 2015 (UTC + 1:00)
System Uptime: 0 days 5:37:16.924
BugCheck 1000007E, {ffffffffc0000005, fffff80094349535, ffffd000204cba78, ffffd000204cb280}
*** WARNING: Unable to verify timestamp for nvlddmkm.sys
*** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys
Probably caused by : nvlddmkm.sys ( nvlddmkm+149535 )
BUGCHECK_STR:  AV
PROCESS_NAME:  System
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``
Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
Debug session time: Sun May 17 21:58:52.527 2015 (UTC + 1:00)
System Uptime: 0 days 0:35:46.221
BugCheck 19, {d, ffffe000ead47000, ce7c29b53c348ef8, cc7c29b53c348ef8}
Probably caused by : ntkrnlmp.exe ( nt!ExAllocatePoolWithTag+e9e )
BUGCHECK_STR:  0x19_d
PROCESS_NAME:  audiodg.exe
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``
Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
Debug session time: Sun May 17 09:26:09.626 2015 (UTC + 1:00)
System Uptime: 0 days 0:16:18.318
BugCheck 3B, {c0000005, fffff80315bdbf6a, ffffd0002279cd30, 0}
Probably caused by : ntkrnlmp.exe ( nt!SwapContext_PatchXRstor+0 )
BUGCHECK_STR:  0x3B
PROCESS_NAME:  NvBackend.exe
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``
Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
Debug session time: Tue May 12 23:02:44.923 2015 (UTC + 1:00)
System Uptime: 0 days 5:58:43.619
BugCheck 1E, {ffffffffc0000005, fffff801dcc06d7b, 0, ffffffffffffffff}
*** WARNING: Unable to verify timestamp for nvlddmkm.sys
*** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys
Probably caused by : dxgmms1.sys ( dxgmms1!VidSchiProcessDpcCompletedPacket+53b )
BUGCHECK_STR:  0x1E_c0000005_R
PROCESS_NAME:  System
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``
Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
Debug session time: Sun May 10 01:31:05.328 2015 (UTC + 1:00)
System Uptime: 0 days 1:09:11.025
BugCheck 19, {d, ffffe0018150f000, 4a541062b58a226b, 48541062b58a226b}
*** WARNING: Unable to verify timestamp for win32k.sys
*** ERROR: Module load completed but symbols could not be loaded for win32k.sys
Probably caused by : dxgmms1.sys ( dxgmms1!VIDMM_GLOBAL::DestroyOneAllocation+2fb )
BUGCHECK_STR:  0x19_d
PROCESS_NAME:  bf4.exe
¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨``
Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
Debug session time: Sun May 17 13:22:35.340 2015 (UTC + 1:00)
System Uptime: 0 days 1:01:50.030
BugCheck 19, {e, ffffe000952bb000, 99f7460024901850, 9bf7460024901850}
Probably caused by : afd.sys ( afd!AfdPollGetInfo+21 )
BUGCHECK_STR:  0x19_e
PROCESS_NAME:  bf4.exe

To try to reduce interference on the GFX side, I suggest you reduce the current nVidia gfx install to the bare minimum, what we want is clean and simple during testing. (A drivers only install might be best to begin with, remove all then follow the procedure here: http://www.sysnative.com/forums/wind...procedure.html)

There are some drivers that aren't needed, might be a source of interference or relatively old:
IntelRST, only needed for RAID or SSD -caching, not needed for JBOD, uninstall it.
Corsair headset and Roccat Kova+, uninstall their software, uninstall the both from Device Manager, including removal of drivers. You should be able (I hope) to use them as generic devices, if not, try to get hold of basic mouse/'phones to use during testing.

Once you've done that, enable Driver Verifier (http://www.sysnative.com/forums/bsod...87.html#post87 study it carefully, it's designed to crash when it detects a bad driver - but it can prevent a normal startup - workaround provided) and then use the computer as normal, it will feel laggy with DV enabled; once you have got a couple of new crash dumps, upload/attach them for analysis; if DV doesn't flush/crash anything withing 24-48 hours, turn it off again.


EDIT: for the SFC errors in the OP, try SFCFix or follow the full procedure here (later?): http://www.sysnative.com/forums/wind...html#post35484

[KeyboardDemon]

Thanks Satrow. It looks like I'm moving in the right direction, mostly with help from you, I'll try as much of that as I can when I get home from work tonight. With larger files I'll upload and PM the links as before if that suits you.

[satrow]

You're doing fine, KD

Take your time, you'll get a better understanding that way, ask if you're unsure of anything and PMing links to your uploads is OK by me!

[KeyboardDemon]

My main issue as a PC Troubleshooting 'n00b' is that I know what tools are available but I don't have the knowledge or disciplined approach to know how or when to use them. That's changing now. Before my general approach was to get an error, put up with it till I can't put up with it anymore and then reinstall. But lately I actually want to find out what's causing the issues and how I can use the tools to put it right. Strangely, I am actually quite excited about having these issues because I have been able to learn some new stuff as a result. Your help and the suggestion from MaddAussie to use DISM have taken me a lot closer to getting my PC fixed, and this thread is keeping a record for documentation purposes.

[satrow]

Don't look to me for ideas on a disciplined approach to anything, I'm more inclined to be intuitive - if the description includes something I've come across previously, I go there first.

I'd just reached the top of a local ridge ~10 years ago when I got a text 'Help, PC's slow to boot, slow to do anything' or similar, I replied 'Google pio node10' and a short time later got a very grateful response.

I do work off the top of my head most of the time and I'm inclined to forget little bits of info that might later prove to have been important - so it's best (and easier for me) if you try to follow what we're trying to do and why

[KeyboardDemon]

Maybe disciplined is not the word I was reaching for, what I wanted to say was methodical.

My typical approach has been guess at something and try and fix that, what we have been doing here is look at the information available to see what issues need to be addressed. What I've got from this is a little bit more on how to use that information constructively and that's exciting me (I know that sounds sad!) what I want to learn now is how to read through that PerfMon report to establish that the issues might GFX/DX, networking or other drivers etc...

I'm happy with my small victories like when I spotted the issues with my Intel drivers that are being reported as both healthy and not working and fixed the not working part. lol But that wasn't all me, I had a gentle nudge to point me in the right direction.

[satrow]

Yes, always try to get some details from logs, reliability monitor and all the other tools within Windows first - and/or build up a good description of what was happening at the time of the problem - anything really to get a feel for which set of components are most likely to be playing a part. It's usually only later that you might need a few specific external tools to help drill down further.

I hate trying to read PerfMon reports, they never seem to reveal the exact details that I need to understand what was happening on the machine during the test; it really needs to be read on the machine it was done on, in the same session with the same software running for best results imo. Gimme an MSinfo32, an Autoruns.arn and a few crash dumps, along with a good description of what was happening and I can (usually) go a long way on that.

Any progress, anything new learned is great, worthy of excitement; what we strive to do early on is to clear the stage ready for the entrance of the real villain of the piece. Sometimes it's only then that the real work begins...

[shaithis]

Ok. I successfully managed to play two full rounds of BF4 with no crashes or issues, hooray and then got a BSOD with a SYSTEM_SERVICE_EXCEPTION whilst watching a YouTube video.

So I don't know why, but it's still not fixed!

I think I need to go to bed.

Those errors have always been graphics card/driver related when I have seen them....Even to the point where certain flash sites were causing them to happen in browsers......which seems odd as most (if not all) of those driver bugs were fixed a while ago and many people are using Youtube in HTML5 now anyway.

[KeyboardDemon]

@satrow: I have done the following things in the following order.

• Removed Intel RST and Corsair Headset with no issues
• Went into Device manager and found I had about 9 HID devices relating to my Kova [+] mouse (according to the properties) I tried removing them but they re-spawned after a restart so I removed the Kova [+] (my logitech has a broken button so I'm trying to get it RMA'd as if I didn't have enough issues)
• Replaced the Kova [+] with a Logitech V220 laptop mouse so I could uninstall the Roccat PNP drivers
• Removed nVidia Drivers and then installed driver only following instructions in the link provided
• Set a restore point in case I need to rollback
• Before running the driver verifier I used bcdeit /set bootmenupolicy legacy so I can use F8 to get into safemode just in case
• Ran driver verifier following instructions
• Got a System_Service_Exception BSOD during restart so have uploaded minidump which should be in your PM soon

I think verifier is still running so will try and do some typical stuff over the course of tomorrow morning and after work, I might leave it a little longer than the suggested 24 hours.

[satrow]

As per PM reply, no new files in the linked collection, the crash might have happened too early in the boot process to have been recorded, or perhaps you made a slip with the link or file attached?

You're making good progress The next dump should furnish me with details on which 'suspect' drivers are left loading, if it's also Verified and indicates a 3rd party driver, we might have this cracked.

Keep working/gaming as normal, try to remember and describe what other User Mode (stuff you started) processes were also active in the background when a crash happens again. I'm wondering whether there's a program common to all crashes (Chrome?) that could be playing a part.

A 0x3B crash, System_Service_Exception, we've had 2x before, a quick breakdown on the dumps and what might have triggered them:

3x 0x19: Running: not recorded, bf4 (dxgmms1.sys image corrupt), bf4 (win32k image corrupt).
Usual causes:  Device driver.
http://www.carrona.org/bsodindx.html#0x00000019

2x 0x3B: Running: bf4, NvBackend.
Usual causes:  System service, Device driver, graphics driver, ?memory
 http://www.carrona.org/bsodindx.html#0x0000003B

2x 0x1E: Running: Chrome, System.
Usual causes:  Device driver, hardware, System service, compatibility, Remote control programs, memory, BIOS.
http://www.carrona.org/bsodindx.html#0x0000001E

1x 0x7E: Running: System (nvlddmkm.sys image corrupt).
Usual causes:  Insufficient disk space, Device driver, Video card, BIOS, Breakpoint with no debugger attached, Hardware incompatibility, Faulty system service, Memory, 3rd party remote control.
http://www.carrona.org/bsodindx.html#0x0000007E

Running: = the Process that was active at the time of the crash, not recorded might indicate the CPU was switching between Cores/Processes when the error occurred. Note that they all have a heavy bias towards GFX involvement.
Image corrupt: = the file (drivers here) held in working memory wasn't the same as that loaded from disk, 'something', maybe a 3rd party driver, had written or modified some or all of the data.
Usual causes: = set out in order of probability based on studies over a number of years of thousands of cases where causes for a BSOD crash were proven. As you can see, Driver is the only common factor, for 1 type listed, it's the only proven cause.

Yup, the above only illustrates what you've suspected for some time, now we need to pinpoint a driver, or we have to suspect hardware. Hardware trumps software.

[satrow]

Ok, dump collection was successful this time. Bottom line, tl:dr keep on with DV for a day or so, if no verified dumps by then, we should turn off DV and, I think, move on to changing driver versions.

The dump shows no sign of any drivers that were removed, here's the remaining 3rd party driver list, no need to do anything with these yet/info only:

**************************Tue May 19 23:34:32.402 2015 (UTC + 1:00)**************************
e1c64x64.sys                Thu Jun 20 22:59:00 2013 (51C37B24)
intelppm.sys                Thu Aug 22 09:46:35 2013 (5215CFEB)
nvhda64v.sys                Thu Oct  9 18:02:27 2014 (5436BFA3)
nvlddmkm.sys                Tue May 12 02:59:00 2015 (55515E64)
RTKVHD64.sys                Wed Aug 21 05:48:22 2013 (52144696)
TeeDriverx64.sys            Tue Jul  2 19:38:08 2013 (51D31E10)
http://www.carrona.org/drivers/driver.php?id=e1c64x64.sys
http://www.carrona.org/drivers/driver.php?id=intelppm.sys
http://www.carrona.org/drivers/driver.php?id=nvhda64v.sys
http://www.carrona.org/drivers/driver.php?id=nvlddmkm.sys
http://www.carrona.org/drivers/driver.php?id=RTKVHD64.sys
http://www.carrona.org/drivers/driver.php?id=TeeDriverx64.sys


   --- E O J ---   20/05/2015       1:24:39.29     3rdPartyDriversName   Copyright 2012 Sysnative Forums

Basic Windbg details of the latest dump:

Debug session time: Tue May 19 23:34:32.402 2015 (UTC + 1:00)
Loading Dump File [C:\SysnativeBSODApps_2_0_0_0\051915-7234-01.dmp]
Built by: 9600.17736.amd64fre.winblue_r9.150322-1500
System Uptime: 0 days 0:16:00.093
BugCheck 3B, {c0000005, fffff802bbca3881, ffffd00148d05ee0, 0}
Probably caused by : ntkrnlmp.exe ( nt!KeTerminateThread+23d )
Bugcheck code 0000003B
Arguments 00000000`c0000005 fffff802`bbca3881 ffffd001`48d05ee0 00000000`00000000
DEFAULT_BUCKET_ID:  WIN8_DRIVER_FAULT
BUGCHECK_STR:  0x3B
PROCESS_NAME:  svchost.exe
FAILURE_BUCKET_ID:  0x3B_nt!KeTerminateThread
MaxSpeed:     3500
CurrentSpeed: 3500
BiosVersion = 2104
BiosReleaseDate = 08/13/2013
SystemManufacturer = System manufacturer
SystemProductName = System Product Name

MS says:

Cause

This error has been linked to excessive paged pool usage and may occur due to user-mode graphics drivers crossing over and passing bad data to the kernel code.

If it's any consolation/encouragement for you, the only 0x3B with the same trigger as this one I could find as [Solved] (http://www.eightforums.com/bsod-cras...s-backups.html) was apparently cured by a reinstall - but - that was claimed as fixed after only ~3 days of the fresh install.

[KeyboardDemon]

Right a mini update, I spent a couple of hours running Adventure Capitali$t (a Steam game) as it was a game that I could leave running without having to sit by my PC. Anyway the game ran ok for a short time but then started shutting down unexpectedly, I re-tried the game and the crashes persisted, I think I must have got it to crash about 4 times.

I decided I would look at the driver verifier status to see how things were going and given that previous errors have suggested paging issues I was looking at drivers that were reporting high levels of paging activity, am I looking for the right things with that?

This is what I found:

Name: ndis.sys, Loads: 1, unloads: 0
CurrentPagedPoolAllocations: 59
CurrentNonPagedPoolAllocations: 1419
PeakPagedPoolAllocations: 61
PeakNonPagedPoolAllocations: 1446
PagedPoolUsageInBytes: 4538
NonPagedPoolUsageInBytes: 573869
PeakPagedPoolUsageInBytes: 5036
NonPeakPagedPoolUsageInBytes: 592685

Name: nvhda64v.sys, Loads: 1, unloads: 0
CurrentPagedPoolAllocations: 16
CurrentNonPagedPoolAllocations: 42
PeakPagedPoolAllocations: 17
PeakNonPagedPoolAllocations: 42
PagedPoolUsageInBytes: 10048
NonPagedPoolUsageInBytes: 26626
PeakPagedPoolUsageInBytes: 10068
NonPeakPagedPoolUsageInBytes: 26626

Name: rtkhd64 , Loads: 1, unloads: 0
CurrentPagedPoolAllocations: 0
CurrentNonPagedPoolAllocations: 706
PeakPagedPoolAllocations: 2
PeakNonPagedPoolAllocations: 708
PagedPoolUsageInBytes: 0
NonPagedPoolUsageInBytes: 217880
PeakPagedPoolUsageInBytes: 650096
NonPeakPagedPoolUsageInBytes: 218216

Name: dump_storahci.sys, Loads: 2, unloads: 1
CurrentPagedPoolAllocations: 0
CurrentNonPagedPoolAllocations: 0
PeakPagedPoolAllocations: 0
PeakNonPagedPoolAllocations: 0
PagedPoolUsageInBytes: 0
NonPagedPoolUsageInBytes: 0
PeakPagedPoolUsageInBytes: 0
NonPeakPagedPoolUsageInBytes: 0

Name: dumpdumpfve, Loads: 2, unloads: 1
CurrentPagedPoolAllocations: 0
CurrentNonPagedPoolAllocations: 1
PeakPagedPoolAllocations: 0
PeakNonPagedPoolAllocations: 1
PagedPoolUsageInBytes: 0
NonPagedPoolUsageInBytes: 16400
PeakPagedPoolUsageInBytes: 0
NonPeakPagedPoolUsageInBytes: 16400

Name: nvlddmkm.sys, Loads: 1, unloads: 0
CurrentPagedPoolAllocations: 186
CurrentNonPagedPoolAllocations: 16599
PeakPagedPoolAllocations: 210
PeakNonPagedPoolAllocations: 22255
PagedPoolUsageInBytes: 966852
NonPagedPoolUsageInBytes: 12539743
PeakPagedPoolUsageInBytes: 1448276
NonPeakPagedPoolUsageInBytes: 14120579

The other drivers just showed zeroes throughout except in loads and unloads where they were showing 1 load and 0 unloads.

Does any of that mean anything?

Are we going in the right direction?

[satrow]

nvlddmkm.sys is certainly showing a lot of activity but I've no clue if that's normal for the type of load it was under or not.

Can you check the entries in Reliability Monitor? It might give hints that are hard to track down elsewhere. Control Panel\System and Security\Action Center\Reliability Monitor

[KeyboardDemon]

Ok. I'm out at the moment but will check and post back later tonight. Thanks.