Windows 7 BSOD after waking from sleep and shutdown

[fillswitch]

Hello!

I've been troubleshooting this issue for months now and it is really frustrating. My issue is every time my computer goes to sleep/is shut down, when I start it back up, the computer works normally for about 7-10mins or so, and then BSODs. Once I reboot from the BSOD, my computer runs just fine until the next sleep-bsod episode

I've done Memtest86+ on my RAM modules to see if they were the culprit.
Module 1: Link Removed
Module 2: Link Removed

I noticed something strange on these tests though. When I would test my memory on slots 1-2, Memtest86+ would read my memory's speeds/timings correctly: 1866 9-10-9-28. But when I tested the same modules in slots 3-4, I would get half of that: DDR3-784 4-5-5-15. I then checked if slots 3-4 read with the lower speeds/timings in CPU-Z, but they did not; they read with the normal 1866 9-10-9-28. I'm not sure if that's an indicator of a problem at all, but it seems unlikely.

Trusting my logic, I decided to see if having both modules in slots 1-2 would alleviate the issue. I put the computer to sleep overnight (usually the scenario that results in a BSOD) and upon waking from sleep, it seemed fine for about 10 minutes (just like before every other BSOD), then it crashed: Link Removed

Here's my build for reference: Link Removed

Any ideas why this is happening?

 

[kemical]

Sounds good Fil. If it helps any I've gone through this process myself and sometimes it's the only way to track down the problem. Best of luck with the troubleshooting.. I'll await your next update.

 

[fillswitch]

I just installed the newest Nvidia driver. I had updated the Marvell driver and disabled the AODriver2.sys. Didn't get a chance to change anything yet and got a BSOD this morning: Link Removed. I will now remove the RAM and reset the settings.

Also, when you asked for the make/model of my PSU, did you find anything wrong with it? Its 750w so it should be able to power most anything.

 

[kemical]

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 3B, {c0000005, fffff8800421011b, fffff88009907c70, 0}

Probably caused by : dxgkrnl.sys ( dxgkrnl!DoublyLinkedList<DMMVIDPNSOURCEMODE>::DoublyLinkedList<DMMVIDPNSOURCEMODE>+df )

Followup: MachineOwner

As mentioned previously this could be down to faulty RAM, drivers or software.

mvs91xx.sys Tue Jul 30 07:52:22 2013: Marvel AHCI driver. 2014 version available on motherboard support page:
http://www.gigabyte.com/products/product-page.aspx?pid=4672#driver

AODDriver2.sys Wed Nov 21 07:44:04 2012: AMD Overdrive; also in EasyTune6 for Gigabyte motherboard Known BSOD issues in Win7 remove to test.

As I mentioned return back to your original set up where no bsod's were occurring and then go on from there. Once your sure your not getting blue screens after using sleep mode then add something back but make sure it's only one thing and then test again making sure your not getting a bsod after a session on sleep mode.

Your PSU wattage sounded fine.

 

[fillswitch]

Woke my computer, and yet again, a new BSOD: Link Removed
This was right after resetting my XMP settings and only using one stick of RAM.

Just to provide some more context, whenever my computer BSODs after waking from sleep, I either get a Windows Error Recovery screen (which I always click "Start Windows Normally") or Enter Safe Mode screen (which I also always start normally). After I make my selection, my computer usually BSODs again (before making it through the "Starting/Resuming Windows" screen). The process then repeats itself anywhere from 1-5 times normally. Then, I'll catch a break and my computer will start up normally.

Another thing to note, sometimes when my computer is asleep and I try waking it up, it will not wake. I usually try waking with my keyboard first, and then with the power button on my tower. When neither of those work, I have to unplug/replug my PSUs power cable from the outlet in the wall, and then use the power button (keyboard won't wake it from sleep). I don't think this is the cause, but it may point to a symptom. Is there possibly something wrong with my outlet? I made sure no other devices use that outlet (I used to have a power strip coming from the wall). Something to note: I have a fan in a close by outlet that I have been running everyday, for hours on end (since it's been a hot past few weeks). Also in a close outlet, I run my monitors and speakers. Not sure if any of this info is helpful, but at this point, you never know.

I'm not quite sure what to do at this point, since we reverted to what was supposed to be the last known working state. I do remember you mentioning switching the ports for my SATA drives? Maybe that will help.

 

[kemical]

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 1000007E, {ffffffffc0000005, fffff8000387fa00, fffff8800352b6b8, fffff8800352af10}

Probably caused by : ntkrnlmp.exe ( nt!HvpDropPagedBins+80 )

Followup: MachineOwner

EXCEPTION_CODE: (NTSTATUS) 0xc0000005 - The instruction at 0x%08lx referenced memory at 0x%08lx. The memory could not be %s.

Hey Fil,
first change the sticks over so your using the other stick that was removed. I'd also check in your manual which SATA ports are the Marvell ones and move your drive to another controller. I've read of other users having to do this because of the Marvell controller although as your only blue screening after using sleep I'm not so sure but try it anyway...
Regarding your outlet, I guess the only way to test that theory is to either use a different outlet or test the outlet itself with a voltage meter.
Can you list what is actually connected to your pc at this time, how many drives, printers, hubs or whatever.
Another thing I noticed with this dump file is the disappearance of certain drivers. The dump previously had your Realtek audio drivers listed and now they are not same goes for the Xbox controller driver that also has disappeared. Did you check the SMART data on the drives like I asked? If not then please check that it's enabled in the bios, download speedfan and use that to check the drive. Speedfan has an actual section which will give you a report on the drives health and is extremely useful in gauging a drives health.

 

[fillswitch]

Alright cool. I don't have the time to set this all up before going to work, but I'll do the following tonight:

• Swap the RAM modules in use
• Move SATA drives from Marvell ports (if applicable)
• Turn on SMART data
• Download Speedfan and check disk health

Here's all of what I have plugged into my computer: Link Removed
Additional items plugged in are:

• HP Deskjet 1050 J410 (USB)
• Xbox 360 Wireless Receiver (USB)
• ASUS VE246 and VE278 monitors (DVI)

I'm not sure why the Realtek and Xbox drivers are not appearing. I do have them installed, from what I know.

 

[kemical]

Thanks for the update Fil, if the bsod still persists it might be worth unplugging the printer to see if that's connected?

 

[fillswitch]

Hey Kemical,

So I checked all over my BIOS settings and I cannot find a way to enable S.M.A.R.T. options. (Here's my Mobo for reference: Link Removed) I then remembered Speccy (Link Removed) has a section for this.

Here is my C: drive: Link Removed
Here is my M: drive: Link Removed

Let me know if this is comparable to checking S.M.A.R.T. options in the BIOS. If not, I could use some help finding where to turn these options on.

I will now swap RAM modules, move SATA drives away from Marvell, and run SpeedFan.

 

[fillswitch]

I've swapped RAM modules and ran SpeedFan (extended test):
Link Removed (no errors)
Link Removed (updated - no errors)

I couldn't find what particular SATA ports are Marvell enabled via the manual (Link Removed) so I haven't changed that. One thing to note though, I mistakenly tested my last config with the sole RAM stick in slot 2 rather than slot 1. I'm sure that didn't help.

So far no BSODs upon my most recent wake. I will note that my computer did not respond to waking via keyboard again, so I reverted to switching the outlet in the wall and started with the power button. The comp resumed where I last left it; been running for approx 30 mins now.

 

[kemical]

Looking at the speedfan values for the Vertex you can see that it's health is declining. If you check the speedfan smart page again you'll notice a button for getting an online report about your drive. These are worth looking at. Also I did a little checking on the Vertex and make sure which firmware you have because there was a version which caused bsod after waking. See this link:
https://www.pcspecialist.co.uk/foru...0-BSOD-caused-by-OCZ-Vertex-3-SSD-Fix-out-now!!

 

[fillswitch]

That is interesting. Unfortunately, I have the latest firmware (see Vertex 3 on this page: Link Removed):
Link Removed

Even with swapping the power outlets, my computer needs to be unplugged and replugged back into the wall before restarting. I'm going to do a little side investigation on that myself.
Edit: I sent in an email to Rosewill's tech support. Hopefully I'll get an answer on this funky unplug/replug situation.

 

[fillswitch]

I also want to note that my last two sleep cycles have been BSOD free; albeit on both instances my computer needed to be unplugged/replugged into the wall outlet (see issue description in previous post). My current config is as follows:

• RAM stick #1 (I've marked them with a sharpie) in Slot #1
• No XMP settings on
• Drives set to AHCI or SATA (not IDE)

I don't know if the unplug/replug has anything to do with Windows resuming successfully, but better to be thorough with the details. My comp is currently asleep and I will be testing if it BSODs when I get home.

 

[kemical]

Sounds good Fil. Try adding stuff back when your ready but as discussed only one at a time and see how you go.. Once your sure it won't bsod then add something else and so on..

 

[fillswitch]

Just an update: I haven't tried adding anything back quite yet. My last wake attempts have been successful, with the caviat that every time I went to wake it, I would have to unplug/replug the power cable and then hit the power button to wake. It would always resume correctly though. Before adding things, I wanted to make sure that the computer wouldn't BSOD after waking correctly (without un/replugging the power). Just today when I got home from work, I was able to wake the computer normally via the keyboard. The computer resumed normally and has been running for an hour now.

I will now add RAM stick #2 to slot #2 and see if I get a BSOD from a handful of wake cycles.

 

[fillswitch]

Wow that was way too fast for my liking. New BSOD: Link Removed

I will try another wake cycle with the current config, but it sounds like my next steps are:

• Using RAM module #1 in slot 2 and RAM module #2 in slot 1
• Using RAM #1 in slot 3 and RAM #2 in slot 4

Let me know if you agree or not.

 

[kemical]

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 1E, {ffffffffc0000005, fffff800034bf309, 0, a}

Probably caused by : memory_corruption

Followup: memory_corruption

Hey Fil,
I was only thinking about your issue this morning and wondering how things were going so I guess now I got my answer.. Testing your RAM further does sound like a good idea and I have a feeling that it's either the RAM or the RAM slots. Do you have any other RAM which you could pop in as a replacement to see how she runs?
Also this unplugging of the power cable and then reconnecting to start the machine up... Do you mean you unplug it from the back of the machine and then reconnect it? If so that doesn't sound right at all. Do you have a different power lead you could try using just in case there is a break inside the one your using? (any reply from Rosewills?)
Lastly have you checked your RAM is compatible with your motherboard? I had a look myself and couldn't see your sticks listed:
Link Removed
It might be worth sending a email to GSkill and enquiring?

 

[fillswitch]

I have some older RAM I should be able to run. I'll try two sticks in Slots 1 and 2 and we'll see what happens.

As far as unplugging the cable, I am unplugging the power cable from the wall outlet. And by power lead, do you mean the cable from the PSU to the wall outlet? I can try that too. One other thing I've noticed is every time the comp won't wake from sleep, when I replug it in, I always see a short spark of electricity in the outlet. I know that this happens in general with outlets, but still something interesting since its consistent.

My reply from Rosewill told me that the PSU might be defective (go figure). They wanted me to test it out first. Here's the email: Link Removed

I just emailed G.Skill and I'll probably drop a line to Gigabyte too to see if there is a compatibility issue.

 

[kemical]

Hmm... I'm unsure what you mean regarding the plug outlet situation. Do you mean that you have a power strip plugged into the wall socket and the machines power lead is plugged into that? The spark is as you say probably normal.
I'll certainly be interested to know what happens when you run the older RAM. I know thisn't probably what you want to hear but I'm hoping that's it's those sticks you've already got installed which are defective if only so we know what component it is (that has the fault).

 

[fillswitch]

So let me re-explain the power situation:
My computer's power cable (what I believe you are calling the Power lead: http://www.newegg.com/Product/Product.aspx?Item=N82E16812228823) is plugged into the computers PSU and plugged directly into a wall outlet (Link Removed). Now about 4 out of 5 times when I try to wake my computer via keyboard, it doesn't wake. Before my BSOD issues it worked via keyboard just fine. So I try to wake my computer via keyboard and it will not wake. I then try using the power button on the front of my tower, and that doesn't work either. The only thing that works is unplugging my power cable/lead from the wall and replugging it back into the wall. After that, I can then wake my computer via the power button on the front of the tower. I hope that makes sense. If not, I can take a picture of how I have it set up. Just let me know.

And I understand about hoping for the old RAM to work too. It would be nice to nail down the issue. I installed two 2gb sticks of OCZ RAM (http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=5210465) into slots 1 and 2 and then let it fall asleep. I did change the sleep time to 15 mins (where previously it was 1hr) to expedite the process. I was able to wake my computer via keyboard this time (not sure if its the 1/5 time that it worked or related to using the older RAM). It has been about 45 mins since I woke the computer and we're still running.

I've changed the sleep time back to 1hr and I'm planning on putting it through another sleep cycle. It may very well be the RAMs compatibility or just bad sticks in general. That would be unfortunate, but hopefully I can figure something out to right this situation.

I'll keep you posted.

 

[fillswitch]

Another update. Another wake cycle and no BSOD with my old RAM. And another wake cycle that I was able to wake the machine via keyboard.

So I could have sworn I saw people online with the same Mobo/RAM combo as I and I hadn't heard of any issues from them. Is it possible that I just need to increase the voltage going to my new RAM via the BIOS? I ask this because of being able to wake the comp via keyboard with the old RAM but not the new RAM. It seemed like the new RAM needed a jump start or something.

I will start looking around the internet to see if anyone with my Mobo/RAM combo was able to run the combo without fail. I might also start compiling an email to G.Skill about replacing my RAM? I have still not heard back from G.Skill about the compatibility of the RAM I bought and the mobo I have.