Instability only on cold boot

[Argi]

Hey,

 

I recently upgraded a bunch of parts of my PC and I'm experiencing weird issues only on first boots of the system.  When I first turn the computer on, I almost always get a blue screen in Windows.  Sometimes right away and sometimes after a bit.  I've seen many error codes; the most recent was CRITICAL_SECTION_CORRUPTION iirc.

 

I've already had to RMA the RAM (one stick was DoA), and the mobo (the ethernet port was faulty) so I think it's less likely to be those two.  I did a pass of memtest and that came back okay, and the new mobo's ethernet seems fine.

 

Any other thoughts?

 

The new parts are i5 4670K, ASUS Z87-Plus, CRUCIAL BLS2KIT4G3D1609DS1 8gb, and a Gigabyte GV-N770OC-2GD GTX 770

[nitins60]

I would do the fresh installation and install the drivers properly so that there won't be an issue with drivers [updated ones].

[Corvid19]

I had a similar problem with a PC at work (instability when cold but ok once the system had warmed up) it turned out to be a faulty batch of RAM, so I would still be suspicious of the memory. Try booting with only one stick installed at a time or completely swap out the memory from a another machine if you can.

[+Mirumir Subscriber¹]

It looks like some kind of a conflict between BIOS and the hardware/OS or BIOS's inability to store information long-term.

 

  Quote

 I've seen many error codes

 

Those codes provide the clues. Try to hit the pause button or take a video or a screen-shot with your phone.

 

  On 18/11/2013 at 05:54, chAos972 said:

Any other thoughts?

 

1) Given the symptoms, the first thing I'd do would be replacing the CMOS battery;

2) I would reseat CPU and RAM, check all the connections;

3) if I had a spare PSU, I'd swap it;

4) If I had any overclocks, I'd disable those and run a BIOS update if one was available;

5) go though BIOS settings;

6) I would also re-apply thermal paste where necessary (video memory for example).

 

Good luck!

[snaphat (Myles Landwehr) Member]

If Win8, it is possible you aren't seeing the issues on non-cold boot because of hybrid-boot. Maybe they'll happen all of the time if hybrid boot is disabled (powercfg -h off)

 

EDIT: if you turn that off and suddenly you start getting BSODs on all boots, I'd hazard a guess that it is probably a software issue. If it simply persist as a cold boot issue, then I'd assume it was more likely to be a hardware issue.

 

In any case, if I were you I'd have tried  fresh install or more likely an alternative OS just to trouble shoot .

[Praetor]

i would check if there are dumps of those BSOD as they provide a good tip for the source of the problem; normally they are stored in the c:\windows\minidump folder. Attach them here so we can take a look at them.

[Argi]

  On 18/11/2013 at 06:46, nitins60 said:

I would do the fresh installation and install the drivers properly so that there won't be an issue with drivers [updated ones].

Before I sent the first motherboard off for RMA, I tried reinstalling windows multiple times and the issue always cropped up.  I haven't since installing the replacement (I didn't bother reinstalling since the parts are the same).

 

  On 18/11/2013 at 07:16, Mr_Self_Destruct said:

I had a similar problem with a PC at work (instability when cold but ok once the system had warmed up) it turned out to be a faulty batch of RAM, so I would still be suspicious of the memory. Try booting with only one stick installed at a time or completely swap out the memory from a another machine if you can.

I'll try that tomorrow with my old RAM.  That should be pretty clear cut.

 

  On 18/11/2013 at 08:30, Praetor said:

i would check if there are dumps of those BSOD as they provide a good tip for the source of the problem; normally they are stored in the c:\windows\minidump folder. Attach them here so we can take a look at them.

 

I looked at 3 minidumps on my work computer and they all showed different forms of memory corruption. The callstack is in kernel code but that doesn't rule out a bad driver.  I'll find out tomorrow morning.

 

3: kd> !analyze -v

*******************************************************************************

*                                                                             *

*                        Bugcheck Analysis                                    *

*                                                                             *

*******************************************************************************

 

SYSTEM_SERVICE_EXCEPTION (3b)

An exception happened while executing a system service routine.

Arguments:

Arg1: 00000000c0000005, Exception code that caused the bugcheck

Arg2: fffff801d2a362a2, Address of the instruction which caused the bugcheck

Arg3: ffffd00024f99b70, Address of the context record for the exception that caused the bugcheck

Arg4: 0000000000000000, zero.

 

Debugging Details:

-----------------

 

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

 

<snip>

 

SYMBOL_NAME:  nt!MmEnumerateAddressSpaceAndReferenceImages+212

FOLLOWUP_NAME:  mmhot

MODULE_NAME: nt

DEBUG_FLR_IMAGE_TIMESTAMP:  5215d156

IMAGE_VERSION:  6.3.9600.16384

STACK_COMMAND:  .cxr 0xffffd00024f99b70 ; kb

IMAGE_NAME:  memory_corruption 

BUCKET_ID_FUNC_OFFSET:  212

FAILURE_BUCKET_ID:  0x3B_nt!MmEnumerateAddressSpaceAndReferenceImages

BUCKET_ID:  0x3B_nt!MmEnumerateAddressSpaceAndReferenceImages

ANALYSIS_SOURCE:  KM

FAILURE_ID_HASH_STRING:  km:0x3b_nt!mmenumerateaddressspaceandreferenceimages

[Praetor]

test with another stick of RAM; this looks like bad RAM to me.

also memtest that computer for the whole night; I've seen tests passing with success after running from an hour or so and then after running overnight (6+ hours) with memtest it accused a failed stick.

[Argi]

I randomly took one of the two 4GB sticks out this morning and low and behold, the computer booted with no issue.  After I get home from work tonight I'll put it back in on its own to confirm the symptoms (I'll make sure it's not the slot, too) but it looks like the replacement ram is also DOA.  

 

Are Newegg fairly understanding in these circumstances?  It's been a while since I ordered cause I've had to keep dealing with faulty parts.

[+PeterUK MVC]

It could be a bad RAM slot.