Solved Random BSODs

[Faintuk]

Just had another BSOD browsing the web. Crash dumps attached.

 

[Arc]

I asked for some more help. It reporting the same thing in every occasion, which is not the real cause.

 

[Vir Gnarus]

I personally see a single 0xC9 crash which was caused by Driver Verifier. However, part of me thinks it's a false positive. Regardless, it was caused by your Razer Deathadder drivers, which are very stale (dated Aug 2, 2007). You'll want to install updates for it as well as any updated firmware. I personally cannot begin to explain how many crashes and other instability woes people have had with Razer products. They code their drivers very poorly. If updating drivers and firmware doesn't fix things, your only choice is probably to uninstall the drivers and just rely on the basic mouse drivers Windows has available.

Again, however, this may be a false positive. While the driver itself did cause this crash, it's because of a check in Driver Verifier (Force Pending I/O Requests) that, unless specifically anticipated, will create false positives. Make sure to follow the instructions here to the letter. Again, provide us any new crashdumps from crashes that may occur after Driver Verifier is setup and the system is restarted.

Analysts:

3: kd> !analyze -v
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

DRIVER_VERIFIER_IOMANAGER_VIOLATION (c9)
The IO manager has caught a misbehaving driver.
Arguments:
Arg1: 000000000000023e, A driver has marked an IRP pending but didn't return STATUS_PENDING.
Arg2: fffff88004a0f4e0, The address in the driver's code where the error was detected.
Arg3: fffff980081baca0, IRP address.
Arg4: 0000000000000000, Status code.

Debugging Details:
------------------

TRIAGER: Could not open triage file : C:\Program Files (x86)\Windows Kits\8.0\Debuggers\x64\triage\modclass.ini, error 2

BUGCHECK_STR:  0xc9_23e

DRIVER_VERIFIER_IO_VIOLATION_TYPE:  23e

FAULTING_IP: 
[COLOR=Red]dadder[/COLOR]+24e0
fffff880`04a0f4e0 53              push    rbx

FOLLOWUP_IP: 
[COLOR=Red]dadder[/COLOR]+24e0
fffff880`04a0f4e0 53              push    rbx

IRP_ADDRESS:  fffff980081baca0

DEVICE_OBJECT: fffffa8009df4b30

CUSTOMER_CRASH_COUNT:  1

DEFAULT_BUCKET_ID:  VERIFIER_ENABLED_VISTA_MINIDUMP

PROCESS_NAME:  System

CURRENT_IRQL:  2

LOCK_ADDRESS:  fffff80002cd5b80 -- (!locks fffff80002cd5b80)

Resource @ nt!PiEngineLock (0xfffff80002cd5b80)    Available

WARNING: SystemResourcesList->Flink chain invalid. Resource may be corrupted, or already deleted.


WARNING: SystemResourcesList->Blink chain invalid. Resource may be corrupted, or already deleted.

1 total locks

PNP_TRIAGE: 
    Lock address  : 0xfffff80002cd5b80
    Thread Count  : 0
    Thread address: 0x0000000000000000
    Thread wait   : 0x0

LAST_CONTROL_TRANSFER:  from fffff80002f603dc to fffff80002ada1c0

STACK_TEXT:  
fffff880`035af1c8 fffff800`02f603dc : 00000000`000000c9 00000000`0000023e fffff880`04a0f4e0 fffff980`081baca0 : nt!KeBugCheckEx
fffff880`035af1d0 fffff800`02f6a47a : fffff800`02f5e9f0 fffff880`04a0f4e0 fffff980`081baca0 00000000`00000000 : nt!VerifierBugCheckIfAppropriate+0x3c
fffff880`035af210 fffff800`02f6b34e : 00000000`0000023e 00000000`00000000 fffff980`081baca0 00000000`ffffffff : nt!ViErrorFinishReport+0xda
fffff880`035af260 fffff800`02f7671f : 00000000`00000000 fffffa80`0607d990 fffffa80`0607da48 fffff880`031f1c68 : nt!VfErrorReport6+0x6e
fffff880`035af330 fffff800`02f76b63 : fffffa80`0607d990 00000000`00000002 fffffa80`09df3dc0 00000000`00000000 : nt!IovpCallDriver2+0x13f
fffff880`035af390 fffff800`02f7cc2e : fffff980`081baca0 fffff980`081baca0 00000000`00000002 fffffa80`09df3dc0 : nt!VfAfterCallDriver+0x353
fffff880`035af3e0 fffff800`02f7f52a : fffff980`081baed8 fffffa80`09df3dc0 fffffa80`09df3b80 fffffa80`0607d990 : nt!IovCallDriver+0x57e
fffff880`035af440 fffff800`02f7cc16 : fffff980`081baca0 00000000`00000002 fffffa80`09df3b80 fffffa80`095dd410 : nt!ViFilterDispatchPnp+0x13a
fffff880`035af470 fffff880`031f94cf : fffff980`081baf20 fffffa80`09df4b30 fffffa80`09df4f08 fffffa80`095dd410 : nt!IovCallDriver+0x566
fffff880`035af4d0 fffff880`04701517 : fffffa80`09df4b30 00000000`00000001 fffff980`0800ae10 fffff980`081baca0 : hidusb!HumPnP+0x193
fffff880`035af540 fffff880`047045e7 : fffff980`081baf68 00000000`00000001 fffff980`0800ae10 fffffa80`095dd410 : HIDCLASS!HidpCallDriverSynchronous+0x4b
fffff880`035af5a0 fffff880`04701ccd : 00000000`00000008 fffff880`046fe300 fffff880`04704c60 fffff980`081baca0 : HIDCLASS!HidpStartDevice+0x13f
fffff880`035af620 fffff880`0470164a : fffff880`046fe300 fffffa80`09df4c80 fffffa80`09df4c80 fffff880`035af6e8 : HIDCLASS!HidpFdoPnp+0x20d
fffff880`035af650 fffff880`046f3805 : fffff880`046fe3b8 fffff880`046fd4b0 fffffa80`09df4c80 fffffa80`00000001 : HIDCLASS!HidpIrpMajorPnp+0x8a
fffff880`035af6c0 fffff800`02f7cc16 : fffff980`081baca0 fffff980`081baca0 00000000`00000002 fffffa80`09df4b30 : HIDCLASS!HidpMajorHandler+0xf5
fffff880`035af730 fffff800`02e9003e : fffff980`081baca0 fffffa80`09dd2820 fffffa80`09df4b30 fffffa80`0957d260 : nt!IovCallDriver+0x566
fffff880`035af790 fffff800`02bc7e6d : fffffa80`06087440 fffffa80`09dd2820 fffff800`02bd1590 00000000`00000000 : nt!PnpAsynchronousCall+0xce
fffff880`035af7d0 fffff800`02e9f386 : fffff800`02cd5940 fffffa80`09dd4d90 fffffa80`09dd2820 fffffa80`09dd4f38 : nt!PnpStartDevice+0x11d
fffff880`035af890 fffff800`02e9f624 : fffffa80`09dd4d90 fffffa80`06070024 fffffa80`0607c5d0 00000000`00000001 : nt!PnpStartDeviceNode+0x156
fffff880`035af920 fffff800`02ec2d56 : fffffa80`09dd4d90 fffffa80`0607c5d0 00000000`00000002 00000000`00000000 : nt!PipProcessStartPhase1+0x74
fffff880`035af950 fffff800`02ec32e8 : fffff800`02cd3500 00000000`00000000 00000000`00000000 00000000`00000000 : nt!PipProcessDevNodeTree+0x296
fffff880`035afbc0 fffff800`02bd3ed7 : 00000001`00000003 00000000`00000000 00000000`00000001 00000000`00000084 : nt!PiProcessReenumeration+0x98
fffff880`035afc10 fffff800`02ae3841 : fffff800`02bd3bb0 fffff800`02dcc601 fffffa80`06063000 00000000`00000000 : nt!PnpDeviceActionWorker+0x327
fffff880`035afcb0 fffff800`02d70e6a : 00000000`00000000 fffffa80`06063040 00000000`00000080 fffffa80`053e8040 : nt!ExpWorkerThread+0x111
fffff880`035afd40 fffff800`02acaec6 : fffff880`03364180 fffffa80`06063040 fffff880`0336efc0 00000000`00000000 : nt!PspSystemThreadStartup+0x5a
fffff880`035afd80 00000000`00000000 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : nt!KiStartSystemThread+0x16


STACK_COMMAND:  .bugcheck ; kb

SYMBOL_NAME:  [COLOR=Red]dadder[/COLOR]+24e0

FOLLOWUP_NAME:  MachineOwner

MODULE_NAME: [COLOR=Red]dadder[/COLOR]

IMAGE_NAME:  [COLOR=Red]dadder[/COLOR].sys

DEBUG_FLR_IMAGE_TIMESTAMP:  46b1a4cf

FAILURE_BUCKET_ID:  X64_0xc9_23e_VRF_dadder+24e0

BUCKET_ID:  X64_0xc9_23e_VRF_dadder+24e0

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

3: kd> !irp fffff980081baca0
Irp is active with 9 stacks 8 is current (= 0xfffff980081baf68)
 No Mdl: No System Buffer: Thread fffffa8006063040:  Irp stack trace.  [COLOR=SeaGreen]Pending has been returned[/COLOR]
     cmd  flg cl Device   File     Completion-Context
 [  0, 0]   0  0 00000000 00000000 00000000-00000000    

            Args: 00000000 00000000 00000000 00000000
 [  0, 0]   0  0 00000000 00000000 00000000-00000000    

            Args: 00000000 00000000 00000000 00000000
 [  0, 0]   0 10 00000000 00000000 00000000-00000000    

            Args: 00000000 00000000 00000000 00000000
 [  f, 0]   0 10 fffffa8009c75050 00000000 00000000-00000000    
          fffffa8009c75050: Could not read device object or _DEVICE_OBJECT not found

            Args: 00000000 00000000 00000000 00000000
 [  f, 0]   0 10 fffffa8009dd7b80 00000000 fffff88004a0d620-fffff880035af270    
          fffffa8009dd7b80: Could not read device object or _DEVICE_OBJECT not found
    [COLOR=Red]dadder[/COLOR]
            Args: 00000000 00000000 00000000 00000000
 [ 1b, 0]   0 10 fffffa8009df3dc0 00000000 fffff80002f62540-00000000    
          fffffa8009df3dc0: Could not read device object or _DEVICE_OBJECT not found
    nt!ViFilterStartCompletionRoutine
            Args: 00000000 00000000 00000000 00000000
 [ 1b, 0]   0 10 fffffa8009df3b80 00000000 fffff880031f1c68-fffff880035af500    
          fffffa8009df3b80: Could not read device object or _DEVICE_OBJECT not found
    hidusb!HumPnpCompletion
            Args: 00000000 00000000 00000000 00000000
>[ 1b, 0]   0 e0 fffffa8009df4b30 00000000 fffff880046f3694-fffff880035af570 Success Error Cancel 
          fffffa8009df4b30: Could not read device object or _DEVICE_OBJECT not found
    HIDCLASS!HidpSynchronousCallCompletion
            Args: 00000000 00000000 00000000 00000000
 [ 1b, 0]   0 e0 fffffa8009df4b30 00000000 fffff80002bd1590-fffffa8009dd2820 Success Error Cancel 
          fffffa8009df4b30: Could not read device object or _DEVICE_OBJECT not found
    nt!PnpDeviceCompletionRoutine
            Args: 00000000 00000000 00000000 00000000

Now that I scrutinized over this a bit more, this to me actually does not appear to be the typical fake IRP that Driver Verifier will occasionally press onto drivers when doing the Force Pending I/O Request check. Rather, this seems to be a legitimate PnP start device IRP, which ended up being pending but the DeathAdder driver mishandled it and DV caught it in the act. Note that the pending on the IRP has already been returned, which is good, but I guess it failed to complete it in a proper manner, which would be explained by the fact that the driver is pre-Windows 7. So with this in mind, this may very well be cause of his problems, moreso than I originally perceived, but it's still TBD at this point.

 

[essenbe]

hello Faintuk, I will throw my 2 cents in. First, please completely fill in your system specs. You can simply go to your last post, in the bottom left click on my system specs, you should see edit system specs. Fill them out completely, please. If you want to know what we need, look at mine. We need a complete description of your hardware to help diagnose the problem.

If you are overclocking your CPU, ram ot GPU please stop. Go into bios and set everything to optimized defaults, save and exit. If you haven't already, go to your motherboards web site and update all of your drivers.

Also, to check your CPU and to see if your hardware is stable please run Prime95 hardware Stress test. Be sure to have a good CPU monitoring program like Core Temp or Real Temp open during the test and monitor your CPU temps. It will get real warm. If it gets too hot, stop the test. Let us know how that comes out.

 

[Faintuk]

Reinstalled Windows and updated it from a clean boot state. The updates have all been successful and a BSOD didn't occur with the installation of SP1 as it did at the last attempt. I'm using the default Microsoft driver for the mouse (though it is still recognised as a Deathadder). The only GPU driver installed is the latest available one. I'll post again if a crash occurs and also update my system specs.

Thanks for your help.

 

[Faintuk]

Things were looking good (something like 8 hours without a crash), however the PC restarted this morning with no information other than a critical Kernel-Power log in Event Viewer. Event viewer also indicates errors regarding a 0xC000000D error with security essentials. The reboot took place when browsing the web.

I've re-enabled verifier to see if anything shows up, but I've installed very few drivers at this stage. Provided I can get the PC to BSOD I'll upload the dumps and in the meantime I'll run Prime95 to see if it turns up anything. Should verifier be run alongside Prime95, or is that problematic?

 

[Vir Gnarus]

Interesting. What was the most recent drivers/software you installed on the new Windows?

For Prime95, you'll want to start with a several-hour (overnight) test with Blend settings, then follow up with the same lengthy test but on Large FFT settings, regardless if the first one passes or fails. This can help us discern if it's either the CPU or some other component.

Verifier should not pose an issue with Prime95 unless Prime95 has some driver that does something unscrupulous, and I have no knowledge of it ever doing such (unless something on hardware corrupts it during its testing). Verifier will help ascertain problems faster.

 

[Faintuk]

I installed the AMD 12.8 Display Driver (Along with Catalyst Control, the Audio Driver etc.), the drivers for the Asus Xonar DG soundcard and the driver for my USB wireless adapter. Verifier lists 12 non-Microsoft drivers.

I had another crash in the last hour. Same thing happened as the last one, reboot with no BSOD. Prior to reinstalling windows this would sometimes occur, but I would usually see a BSOD.

I will run Prime95 tonight and post the results tomorrow. In the meantime, is there some software I can use to collect all my system specs so that I can post them here?

 

[essenbe]

See this http://www.sevenforums.com/tutorials/180324-system-info-see-your-system-specs.html. Please be as detailed as possible. If you are having a hardware problem, the more we know about what hardware you have the better we can help.

 

[Vir Gnarus]

If you had a crash with no BSOD, either the BSOD appeared too quickly (as in it failed to dump a crashdump file and just restarted immediately), or your system restarted from reasons besides a BSOD, which could only mean power supply failure, CPU overheating or some kind of short circuit.

We can make sure to determine if this was a BSOD or not by turning off automatic restart on BSOD. To do so, type "advanced system settings" in Start Menu, then go to Startup & Recovery settings, then uncheck the automatic restart option. Save and restart the PC. Now if the PC does BSOD, regardless if it fails to produce a crashdump or not, it will stay on the blue screen. If your system still reboots, and you can confirm it saved your changes, then you have one of the aforementioned hardware failures.

 

[Faintuk]

System Specs updated. Let me know if you'd like any more info.

I had turned automatic restart off prior to the last reboot. It isn't the power supply as I upgraded from 380 to 600w this week to try and correct the issue. I've been monitoring CPU temps and they are never above 30 and rarely beyond 25. I'll give the heat sync a clean anyway.

 

[Vir Gnarus]

While you're at cleaning the heatsink, you'll want to go ahead and re-seat all your cards and reconnect all your cables, making sure that there's no material (like dust) that got into the slots. Also make sure the motherboard is sitting correctly on the standoffs in your PC chassis. All of these can cause shorts which would cause your system to spontaneously bug out.

You mentioned you replaced the PSU. When did you replace it? Did you suffer the non-BSOD reboots before or after your replaced it? Did you turn off automatic restarting before or after you replaced the PSU?

There's always the chance that you could've replaced a dud with a dud. Again, not conclusive, but shutdowns like this can only occur from hardware failure of specific types.

 

[Faintuk]

Reboots were occurring previously, but as far as I recall they only occurred with verifier enabled. In an effort to fix the issue I opened up the case, checked the connections and removed the sound card (before re-enabling on-board sound). Everything seemed to be going ok so I turned verifier off. However, I just recieved an IRQL_NOT_LESS_OR_EQUAL 0x0000000A BSOD. It failed to write to the crash dump.

I'll get Prime95 up and running again. I did a blend test overnight but the system locked up (everything turned on, but unresponsive) about 2 hours into it. Despite that, I looked at the results document and there are no failures.

 

[Faintuk]

To answer your question, I had suffered non-BSOD reboots before I replaced the power supply. As I said, I think these were generally when verifier has been enabled.

Edit: BSOD occurred when gaming.

 

[Faintuk]

I've been running Prime95 with large FTTs since my last post, but CPU temps aren't exceeding 35 (according to SpeedFan). Should it take this long to increase temp?

Edit: Rebooted (No BSOD) shortly after I posted this. The browser was still open and verifier was enabled (unlike at the last crash where there was a BSOD). Could be a shot in the dark, but I'm going to run the computer without the wireless adapter connected for the next 24 hours and try and force a crash. All reboots and BSODs have occurred with a browser window open or in a game running through Steam (while connected to the internet).

 

[essenbe]

No. Prime95 will usually reach it's maximum temp in less than 1 hour. Why not try Core temp to monitor temps and see if you get different readings.

EDIT: you may want to see if there is an updated driver for your Network Adapter.

 

[Faintuk]

Just had another BSOD in-game (without the network adapter plugged in). This time the BCcode was 101 and it stated that "A clock interrupt was not received on a secondary processor within the allocated time interval". I haven't seen this particular BSOD before. I'm going to try and run the Prime95 Large FTTs test again to see if any of the cores run above 35.

Any thoughts on why it only reboots without a BSOD with verifier running?

Dump attached.

 

[essenbe]

See if your computer is set up to create mini dump files. http://www.sevenforums.com/tutorials/174459-dump-files-configure-windows-create-bsod.html

 

[Faintuk]

It was set to create a kernel memory dump rather than minidump.

 

[Vir Gnarus]

Setting it to minidumps will ensure that it'll write the dump file regardless of the system not meeting the requirements necessary to make a kernel dump. If the dumps are still not written, then it's because of a failure of writing to the drive when the BSOD occurred, which could be caused by many various and sundry things.

If you've made sure the system is still setup to not do automatic restart with BSODs, and the system is still rebooting instantly, then that just has to be some form of hardware failure, as even if Driver Verifier was involved it would not automatically reboot like that.