Solved Random BSODs, now hard waits, file attached

[buckyswider]

Hi, I had a few BSODs and random times, and now there are hard waits at random times, solved only by power off/on.

1- The first BSOD was while awaiting reboot after I had increased virtual memory in an attempt to fix chrome tabs from crashing due to out-of-memory.
2- after the first BSOD I found some hits on the code so I updated my NVIDIA drivers
3- somewhere along in on of the many reboots I noticed Intel Rapid Storage Technology had marked one of my Samsung EVO SSDs as failed- I have two in a mirrored volume. I thought haha, that's it- so I removed the one bad drive and sent it back for warranty replacement (still pending) but now the hang has re-occurred. (So Intel RST is showing the volume as degraded at this point).

I have a few funky USB hardware devices connected (UPS, data slicer). I run dual monitors (one DVI one HDMI).

Gonna hit submit before I get another hang...If I can think of other pertinent info I will reply.

 

[buckyswider]

Well, typing that all out proved a bit cathartic, as I realized that I probably turned my BSODs into hangs by "updating" my NVIDIA driver. So I went in to do a rollback...and the "update" it had given me was from 2015!! I did a rollback, and it gave me a March 2018 driver. I had windows updates fail on this update yesterday, saying a "more recent" version was installed.

So now let's see if the driver update takes us back to the BSOD from the "hangs". Or (wishful thinking) the defective SSD was causing the BSODs and now all will be well... :roflmao:

 

[axe0]

Hi buckyswider,

Has any hardware been changed in the past years?

I'm a bit worried about the memory, I had a look at the specs of the memory and it doesn't look like they should run together.
You have memory from Crucial and Hynix running at the same speed, but according to the specs the timings are different when they're running at the same speed which is quite a risk for system stability.

 

[buckyswider]

Hi axe0! Thanks for the reply. I added that crucial memory in November 2015 (thanks Amazon for the order history!). Although that would seem to rule this out as the current problem, if you think it's something that should be addressed I will go ahead and remedy the situation. I assume the Crucial is the "faster" memory and the best move would be to remove the Hynix and replace it with identical Crucial sticks??

 

[buckyswider]

Well, so much for my theory. Hangs continue with the updated NVIDIA driver.

 

[buckyswider]

Hi again, no progress in this quest. I checked the memory specs on the different pairs of memory- they look exactly the same to me. I'm still open to removing/replacing a pair in an effort to troubleshoot. I did run memtest which came back clean. I have disconnected both USB devices that are attached. All that's left is the keyboard & mouse. I found that NVIDIA has a newer driver than windows update was giving me, so I installed that. It didn't fix anything- if anything, it made the hangs more frequent. So I've ordered a new video card of the same model (GT 635) and I'll swap that out in an attempt to isolate bad hardware.

The hangs have been about 2 or 3 a day. I did have a BSOD overnight last night (first one of them in a couple weeks) so I've re-uploaded the dumps.

my EVO SSD is back from Samsung- they gave it a clean bill of health & upgraded the firmware. So I re-installed that and I re-built the Intel RST RAID-1 volume. Other than replacing the video card, I don't know what else I can try. Could my keyboard be causing this? I can swap that out as a test. It has an integrated USB hub which the mouse attaches to. I have two monitors off the GT 635- could a monitor do this? Other than the two EVO SSDs in the RAID-1 volume, I also have a 1TB spinny disk on the internal SATA, and a DVD+R/W. I'm at a point where I'm thinking of just ditching this hardware and starting fresh (although it's a good beefy machine that runs great & fast except for, you know, that hanging thing ).

Thanks for taking a look....

 

[axe0]

Odd, a search for HMT351U6EFR8C-PB shows the following on all sites I checked (including but not limited to HMT351U6EFR8C < PRODUCTS < SK Hynix & HMT351U6EFR8C-PB from SK Hynix)
Speed: 1600
Timing: 11-11-11

Crucial's memory shows the following
Speed: 1600
Timing: 9-9-9

 

[buckyswider]

Wow, thanks- I learned something new today. I thought that if I had the same type RAM sticks with the same frequency ratings that all would be compatible (especially if they're installed in the properly paired slots). But apparently not. I will remove the two 4GB sticks and see if that makes a difference. I have been running this way for 3+ years with no issues, but hey, until a cause is found anything can be suspect. Did you see anything in any of the dumps that lead you to this or was it just a 'best practices' check of my RAM specs??

Thanks again!

 

[buckyswider]

I still haven't removed those RAM sticks yet (waiting for a drive sled to arrive so I can do a couple tasks while I'm disassembled). BUT...what I've done recently was to use DDU to remove ALL versions of my video drivers and do a cleanup. And then I re-installed the latest driver versions from NVIDIA. So I've taken my hangs to like once a day now, which is a decent improvement. What I'm contemplating is using DDU to clean up THAT latest version of the NVIDIA driver, and then installing the latest "supported" version from Dell. I wonder if while cleaning up the video drivers I solved one problem while introducing a new one in the latest and greatest drivers? I also have a replacement video card on the way. Throw enough mud at the wall, maybe some will stick!!!

 

[buckyswider]

I removed the 2 4GB memory sticks, leaving a matched pair of 8GB sticks...hangs continue.

 

[buckyswider]

New logile attached...

 

[axe0]

Wow, thanks- I learned something new today. I thought that if I had the same type RAM sticks with the same frequency ratings that all would be compatible (especially if they're installed in the properly paired slots). But apparently not. I will remove the two 4GB sticks and see if that makes a difference. I have been running this way for 3+ years with no issues, but hey, until a cause is found anything can be suspect. Did you see anything in any of the dumps that lead you to this or was it just a 'best practices' check of my RAM specs??

Thanks again!

It's definitely best practice, but that's not why I mentioned it.
I mentioned it, because it's not uncommon for system incompatibility to cause issues.


Please update the Intel usb chipset drivers from Dell, it looks like they are repeatedly causing issues and they're a little outdated.

Bugcheck code 0000009F
Arguments 00000000`00000003 fffffa80`172f4bf0 fffff800`04b53798 fffffa80`0cf5ec60
Debug session time: Mon Jan  7 05:17:46.872 2019 (UTC + 1:00)
System Uptime: 0 days 2:14:33.684
0: kd> !irp fffffa80`0cf5ec60
Irp is active with 7 stacks 5 is current (= 0xfffffa800cf5ee50)
 No Mdl: No System Buffer: Thread 00000000:  Irp stack trace.  
     cmd  flg cl Device   File     Completion-Context
 [N/A(0), N/A(0)]
            0  0 00000000 00000000 00000000-00000000    

			Args: 00000000 00000000 00000000 00000000
 [N/A(0), N/A(0)]
            0  0 00000000 00000000 00000000-00000000    

			Args: 00000000 00000000 00000000 00000000
 [N/A(0), N/A(0)]
            0  0 00000000 00000000 00000000-00000000    

			Args: 00000000 00000000 00000000 00000000
 [N/A(0), N/A(0)]
            0  0 00000000 00000000 00000000-00000000    

			Args: 00000000 00000000 00000000 00000000
>[IRP_MJ_POWER(16), IRP_MN_SET_POWER(2)]
            0 e1 fffffa80172f4bf0 00000000 fffff88007b8ba2c-fffffa80172e91a0 Success Error Cancel pending
	       \Driver\iusb3hub	RtsUStor
			Args: 00000000 00000001 00000001 00000000
 [IRP_MJ_POWER(16), IRP_MN_SET_POWER(2)]
            0 e1 fffffa80172e9050 00000000 00000000-00000000    pending
	       \Driver\RSUSBSTOR
			Args: 00000000 00000001 00000001 00000000
 [N/A(0), N/A(0)]
            0  0 00000000 00000000 00000000-fffffa801d283400    

			Args: 00000000 00000000 00000000 00000000
0: kd> lmvm iusb3hub
Browse full module list
start             end                 module name
fffff880`04e3d000 fffff880`04e9c000   iusb3hub T (no symbols)           
    Loaded symbol image file: iusb3hub.sys
    Image path: \SystemRoot\system32\DRIVERS\iusb3hub.sys
    Image name: iusb3hub.sys
    Browse all global symbols  functions  data
    Timestamp:        Fri Mar 29 13:36:15 2013 (51558ABF)
    CheckSum:         00064C0C
    ImageSize:        0005F000
    Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4

 

[buckyswider]

Thanks again! Is this the one you're referencing? It's the only USB driver on the dell support site for my service tag #:

Access Denied

OK, it won't let me paste the link, but it's:

ntel USB3.0 xHCI Driver

This package provides Intel USB 3.0 xHCI Driver Installation and is supported on XPS 8700 running the following Windows Operating systems: Windows 7.
Fixes & Enhancements
Fixes
- Not Applicable

Enhancements
- Enhanced Intel chipset compatibility
Version
Version 2.5.0.19, A02
Category
Chipset

 

[axe0]

Yes, that's the one.

 

[buckyswider]

OK, it's been updated, thanks! Will keep you posted on the hang/crash situation. I have also noticed what may be abnormal behavior on my 2nd monitor (or maybe just things I never noticed before) so that is also disconnected.

 

[buckyswider]

No good, another hang overnight.

 

[axe0]

Could you uninstall the device in device manager, including the drivers, and reboot your system.

 

[buckyswider]

Thanks once again Axe0. I have done that and after reboot it automagically re-installed with the latest NVIDIA driver version 25.21.14.1735 dated 12/11/2018. I think this is one I had installed manually before, but maybe this different method of installation will yield better results.

Yesterday I started getting the hangs more frequently, and this morning I had a BSOD which implicated NVIDDMKM,SYS (bccode 116) so I did replace the video card this morning with a new piece of the same model- but still had a hang a few hours later.

Will update when the machine next hangs/BSOD (or if it doesn't in the next 24 hours).

 

[buckyswider]

Well, two hours later another hang...beginning to think i need an exorcist...

 

[buckyswider]

I wonder if getting an ATI based video card and using DDU to obliterate the NVIDIA install(s) would be a worthwhile endeavor? If it fixes it all the better; if it doesn't, it eliminates the NVIDIA driver from suspicion...