Constant BSODs consistently a minute or two after starting Windows

O

Onyx

Member
Joined
Sep 10, 2017
Posts
6
· OS - Windows 10, 8.1, 8, 7, Vista ?
 - Windows 10 x64
· What was original installed OS on system?
 - Windows 10 x64
· Is the OS an OEM version (came pre-installed on system) or full retail version (YOU purchased it from retailer)?
 - Retail.
· Age of system (hardware)
 - ~ 3 months
· Age of OS installation - have you re-installed the OS?
- I reset my Windows 10 install yesterday. Prior to that, the OS install was as old as the hardware.
· CPU
 - i7 7700k
· Video Card
 - GTX 1070
· MotherBoard - (if NOT a laptop)
 - Gigabyte Z270X Gaming K7
· Power Supply - brand & wattage (if laptop, skip this one)
 - EVGA SuperNova 850W G3
· System Manufacturer
- Myself.

· Laptop or Desktop?
 - Desktop

I have already tried asking on Microsoft's forums for some support, and there are additional details of what I've experienced in that topic (including several minidump files): Sudden increase in Windows 10 BSODs after updates around - Microsoft Community

(Apologies for the long link.)

I updated my BIOS from F3 to F7 this morning, in the hopes that it would fix my persistent BSOD issues, but doing so has only made it worse. I cannot go more than two minutes without my computer BSODing, even if I idle at the Windows login screen and do nothing. Usually the errors are KMODE_EXCEPTION_NOT_HANDLED, and the "What failed:" message calls out intelppm.sys. The other most common is an IRQL error. Occasionally there will be a page fault exception or an out of bounds memory exception. I also observed a KERNAL_AUTO_BOOST_INVALID_LOCK_RELEASE error message this morning, but that has not recurred.

I have had no errors in safe mode thus far after at least an hour in this.

Some searching on the Internet indicated that Gigabyte had issues with its F5 and F6 BIOS versions for my mobo (related to XMP), but that they should be resolved in F7. My BSODs are occurring with XMP both enabled and disabled. When it's disabled, my mobo defaults to running my RAM at 2133 MHz. With XMP enabled, it runs it at the rated 3200 MHz.

I cannot downgrade the BIOS normally below F6. In desperation I also tried F6 but had no better success there.

I have tried to run the Sysnative collection app in safe mode, but I was hung with a message "Waiting for tasks to complete..." for more than 30 minutes. I will try to run it in regular mode, but I fear that the application won't complete before an inevitable BSOD. I will try now and add the ZIP if it succeeds. Can some debugging be done without that data?
 
View attachment SysnativeFileCollectionApp.zip

Success! The processing eventually did finish in safe mode.

Edit: So one thing is that there is an Intel INF update available from Gigabyte. I looked at my CPU in my Device Manager and saw that the driver version was from 2009 (version 10.0.15063.0). The INF update is a newer version number (version 10.1.1.38). I tried to install it in safe mode, but got an unknown error in the Intel install wizard. Should I attempt to install it in regular mode? My fear is that I will BSOD before the install completes and somehow completely bork everything.
 
Hello Onyx :)

Two of the mini-dump files reveal that the following driver is causing problems:

e1d65x64.sys driver is dated:Tue Jul 26 17:48:21 2016
Driver c:\windows\system32\drivers\e1d65x64.sys (12.15.23.7, 530.45 KB (543,184 bytes), 7/26/2016 11:48 AM)


Code: 
ffffe081`ecf117c8  fffff803`0ed18f0d*** WARNING: Unable to verify timestamp for e1d65x64.sys
*** ERROR: Module load completed but symbols could not be loaded for e1d65x64.sys
 e1d65x64+0x18f0d

This driver is an Intel network driver (Intel(R) Ethernet Connection (2) I219-V, Adapter Type Ethernet 802.3)

and you should be able to update it directly via the Intel download centre.
 
philc43 said:
Hello Onyx :)

Two of the mini-dump files reveal that the following driver is causing problems:

e1d65x64.sys driver is dated:Tue Jul 26 17:48:21 2016
Driver c:\windows\system32\drivers\e1d65x64.sys (12.15.23.7, 530.45 KB (543,184 bytes), 7/26/2016 11:48 AM)


Code: 
ffffe081`ecf117c8  fffff803`0ed18f0d*** WARNING: Unable to verify timestamp for e1d65x64.sys
*** ERROR: Module load completed but symbols could not be loaded for e1d65x64.sys
 e1d65x64+0x18f0d

This driver is an Intel network driver (Intel(R) Ethernet Connection (2) I219-V, Adapter Type Ethernet 802.3)

and you should be able to update it directly via the Intel download centre.
Click to expand...

Hi Phil.

Thank you for digging into this! I have updated that driver, and it's now up to 12.15.25.6712 dated 7/12/2017. Unfortunately, it looks as though that may not be the only cause of my problems (did the other minidumps indicate any other issues?). The first restart after I updated the driver, it seemed to be more stable for a bit longer. However, it soon BSOD'd again and I'm back to the now normal BSOD after a minute or two.

I've run the Sysnative collection tool again and I have 5 more minidumps. Could you look through them again?

View attachment SysnativeFileCollectionApp_postIntelDriverUpdate.zip
 
I could not find any reference to other misbehaving drivers. After having had so many crashes I would make sure all your drives are error free with CHKDSK C: /F (replace C: with whatever drive you are checking) and also run a sfc /scannow command to check that all your system files are free of any corruption.
 
I've run both chkdsk and SFC, no errors were detected. I ran Memtest86 earlier and no errors were detected after 2 passes. Before that, I swapped my RAM sticks to the other pair of slots to make sure it wasn't the slot that was bad. Is it possible that this has something to do with the second Ethernet port my mobo has? That somehow the two drivers are interfering with each other?

It's probably not going to show anything, but there's 3 more minidumps that I got from BSODs with different errors than the normal KMODE_EXCEPTION. Should I try to generate a full blown dump instead of these minidumps?

View attachment SysnativeFileCollectionApp_round3.zip

Is this at the point where I just need to buy another mobo and rebuild my PC? I'm going to try doing a full reinstall of Windows, not a reset, but I'm not hopeful that's going to do anything.
 
Unfortunately I've already begun the process of trying to reinstall Windows, so I'll run those if/when I get back into Windows.

The problem now is that I've gotten 4 BSODs during the install process, one of which was a DPC_WATCHDOG_VIOLATION stop code (which I believe is related to SSD drivers?). The first and third I did not catch. The fourth was another KMODE_EXCEPTION_NOT_HANDLED. I am attempting to install via USB.

I'm going to try one last time and then run Memtest86 overnight to see what happens.

Failing that, I have a new motherboard arriving tomorrow (Monday). If I can't make any progress with the Windows install, I guess I am going to go with re-assembling this thing from the ground up and probably RMAing a bunch of stuff.

I really appreciate the help and time y'all have spent so far. Thank you. I'm sure I'll be back tomorrow with more problems.
 
It's been a while, but I did manage to get my problems resolved. (I thought I'd post because xkcd: Wisdom of the Ancients)

I did run 10 passes of Memtest. No issues were found. I tested using my new motherboard, but was not able to even get through a complete Windows install most of the time. Of the one time that I did, after trying about a half dozen different USB sticks, Windows was actually somehow installed incorrectly; Edge would not open any tabs successfully, and I couldn't change some Windows settings without further BSODs.

After that, I decided an RMA of my CPU was my next step. Three weeks later, my refurbed replacement CPU arrived from Intel and I rebuilt my computer again. Reinstalled Windows and everything worked like a charm. I haven't had a blue screen yet. *fingers crossed*

Thank you all again for looking into my BSOD logs.
 

Has Sysnative Forums helped you? Please consider donating to help us support the site!

Back
Top