Diagnosing W10 BSOD (CRITICAL_PROCESS_DIED) without a dump

1

I've been getting BSODs (CRITICAL_PROCESS_DIED) on one of my Windows 10 machines on a "regular" schedule of about once in 15 days. One of the problems I face in diagnosing this is that there are no crash dump files. I leave my machines on 24/7 and all of my other machines typically run for a few months before I manually initiate the update/reboot cycle.

Things I've tried

  1. Replacing the OS hard drive (I figured it could also be a HDD issue if the OS was unable to write the crashdump)
  2. Ran memtest overnight several times
  3. Reinstalled Windows (After #1)
  4. HW Temp monitoring
  5. Reduced third party kernel mode components to a minimum

What I would like to do is have some kind of continuous kernel monitoring running with a modest performance impact that lets me work. Any ideas are welcome.

Hardware Config:
Win10 Pro-64bit
Intel i7-4790K
Crucial 16GB DDR3
ASUS Maximus VII Hero
ASUS GTX 980
Intel SSD - SSDSC2BB800G7
Seagate 3TB - ST3000VN000

Peripherals:
Logitech Webcam C615
Oculus VR Headset

Mark

Posted 2017-02-12T17:42:05.277

Reputation: 11

this is also hard to debug with a dump. check the HDD/SSD and RAM for issues and make sure you use the latest firmware for your SSD – magicandre1981 – 2017-02-12T17:50:37.777

@magicandre1981 Thanks for the tip. I am using the latest firmware. One thing I did was change my SSD to a 'data center' SSD which is designed for 24/7 operation. – Mark – 2017-02-12T20:51:02.727

Answers

0

It turns out that the problem was my CMOS battery. I've replaced it and I'm back to my usual several months uptime before I reboot for updates.

Mark

Posted 2017-02-12T17:42:05.277

Reputation: 11