New PC at Xmas, random freezing

[Warmo161]

Hi guys, since Xmas after building my new PC, ive had one big problem, at any random time, my computer just freezes, as in, mouse not moving, no CTRL ALT DEL, or anything, so I have to hold the power button and turn it on again

PC Specs:

Intel 6700k CPU
Corsair Vengence 8GB
Nvidia 660TI 2GB
Evga Modular PSU
Thermaltake Lv10 GT Case
2 SSDs, and 3 HDDs
Coolermaster H 212 EVO
Asus Xonar DSX Sound Card
Windows 10 Pro
MSI Krait Z170x Motherboard


Over the past week, I have been doing some tests in a attempt to fix my PC:

I have ran the PC without the GPU (Reverting to Intergrated)
Used 3 types of GPU drivers
Running Memtest against the RAM
Disabling ECO mode on the PSU
Running without the sound card
Rewiring the SATA cables for the HDDs -

• in the case, the 5 HDD's are connected to power by one SATA power cable, so I redid the wiring so each drive now has their own power

All of these have failed, my next plan was to run the PC without any of the drives apart from the OS, if that fails then I am planning to remove the CPU from the MB and see if there is any bent pins. Oh and try Windows 7 (Or a Linux Live CD)


I have no idea what to do, can any of you guys help?

EDIT: Forgot to add the motherboard

 

[Warmo161]

Have you checked temperatures with HW Monitor or similar?
Andi.

Yep, everything seems ok, I've ran furmark and prime 95, the PC can handle load just fine

Tried just running with the boot / OS disk attached?

Certainly has the same symptoms as a disk on the way out.

I'm going to try that later today, I have ran a quick SMART test on all the drives and its not giving me anything bad

What's in the Event Log? Any clues?

Nope, I check it everytime it restarts, apart from some messages from VMWare, nothing wierd showing up

EDIT 3: The only thing I get is "Event 41, Kernel-Power" error each time


Tried a BIOS update?[/QUOTE]

Yep, everything is up to date, I found a post on reddit talking about some problems with the Z170 chipset, nevertheless, Im running the latest BIOS (A5) and I can tell no difference between that and the Version I used what it shipped with (A0)


EDIT: Forgot to add what motherboard I have (Z170A MSI Krait)

EDIT 2: This is the SMART log from my oldest drive I have in the PC ( Seagate 1tb Sata 2 drive) via HDTune

HD Tune: ST31000528AS Health

ID Current Worst ThresholdData Status
(01) Raw Read Error Rate 108 99 6 19158060 Ok
(03) Spin Up Time 94 94 0 0 Ok
(04) Start/Stop Count 88 88 20 13014 Ok
(05) Reallocated Sector Count 100 100 36 0 Ok
(07) Seek Error Rate 86 60 30 426085149 Ok
(09) Power On Hours Count 86 86 0 12381 Ok
(0A) Spin Retry Count 100 100 97 0 Ok
(0C) Power Cycle Count 96 96 20 4347 Ok
(B7) (unknown attribute) 100 100 0 0 Ok
(B8) (unknown attribute) 100 100 99 0 Ok
(BB) (unknown attribute) 100 100 0 0 Ok
(BC) (unknown attribute) 100 99 0 20 Ok
(BD) (unknown attribute) 100 100 0 0 Ok
(BE) Airflow Temperature 76 52 45 404095000 Ok
(C2) Temperature 24 48 0 24 Ok
(C3) Hardware ECC Recovered 30 22 0 19158060 Ok
(C5) Current Pending Sector 100 100 0 0 Ok
(C6) Offline Uncorrectable 100 100 0 0 Ok
(C7) Ultra DMA CRC Error Count 200 200 0 9 Ok
(F0) Head Flying Hours 100 253 0 25070 Ok
(F1) (unknown attribute) 100 253 0 159032258 Ok
(F2) (unknown attribute) 100 253 0 -1430909628 Ok

Power On Time : 12381
Health Status : Ok

 

[Warmo161]

There are a few things you can try, firstly right click on start and click Command prompt (Admin) type in sfc /scannow and hit enter , allow it to run System File Checker, it should find and fix and core file integrity issues.

Then you can try run in safe mode, if none of your symptoms exist in safe mode then most likely it is a driver conflict of some kind.

Have you overclocked? could be unstable if so run at stock speeds to determine if it is that.

Please check your event viewer for any errors given and post back with your findings, to do so simply press Windows+R to open the Run dialog, enter eventvwr (or eventvwr.msc) and hit OK.

Is there a difference between the sfc command and going to Properties/Tools -Error checking when right-clicking the drive in my computer? As i've allready done that one with no errors coming back

Nothing is overclocked, apart from the memory which was running on a XMP profile, although after testing it and finding the computer crashing with and without the RAM OC running, the RAM isnt the issue in my build

EDIT: I'm going to try safe mode later, after running my PC without my oldest HDD, if it crashes again, i'll move on to testing safe mode (Although its annoying that the freezing is random so I could be in safe mode for hours)

 

[Warmo161]

In my experience with machines I have built / had and troubleshooting friends, 99% of times if an OC is not involved, lockups are caused by GPU, either PSU not doing it's job or the GPU either failing or GPU memory working beyond it's limits.

Are you saying that even with integrated you still got identical lock ups?

Make sure if your PSU is multiple rail that you have it laid out correctly. That is to say, not everything daisy-chained off of the same rail. So make sure you have enough amps on that to cover everything it's plugged into. If you have a PSU issue, when put the GPU in it may only magnify the issue, especially if it doesn't have it's own two feeds with enough amps to run.

Have you tried OCCT power test and checked to see what the system voltages are doing?

If you haven't mind, run OCCT at your own risk, I've set fire to overclocked systems motheboards with that application lol.

Yep, running on integrated, I still got the PC crashing, so luckily its not my graphics card??


I have my PSU from my past machine (A not moduar XFX PSU that sometimes clicks ), I could test my machine on that. Its good I have a list of other things to test if I have another crash. (Just got back home, will continue to use pc without the Seagate HDD as posted earlier)

I've ran OCCT and with the GPU and Processor on full, it seems ok...

EDIT: if its any help, im more than happy to make a diagram or somthing to explain how my PSU is wired up, I just havent had one before so I have no idea if im doing anything right or wrong

 

[Warmo161]

Would do no harm to test the other supply(that click u hear is likely just a relay which is perfectly normal btw).

Out of curiosity, is your version of windows 10 an upgrade from a previous version of windows eg 7 8 or 8.1? I ask only because I had a lot of bugs until I clean installed win 10 after upgrading myself.

If you are still crashing in safe mode it's likely hardware related but if its fine then its a driver conflict most likely.

These kind of things can prove frustrating but on the upside you will learn a lot from it when finding the resolution.

I have a suspicion that the psu could be the culprit, just my gut feeling. Test your older psu and report your finding's fingers crossed for you

Hmm, I was thinking it was the PSU too, well, i'll keep looking and will be updating what happens as much as I can

Also, when I set up the PC, it was fully a clean install (Apart from the key ofcourse)

 

[Warmo161]

Ok, another report, It crashed again (So the old HDD isnt the problem) so I have changed to my old PSU, all I have to do is wait...

 

[Warmo161]

and... its not the PSU, I will try safe mode tomorrow, if that fails, then its the motherboard??

 

[Warmo161]

I run windows 10 pro with a xonar stx and no issues, I would be more inclined to agree with mugwuffin's theory.

Edit: Not sure if relevant to op's situation but Kitguru released an article about "skylakes freezing bug"

I saw a post on the hardware subreddit a while ago commenting on such a bug, as soon as it comes out for my board i'll give it a go (Just checked and theres another update (I have A5 and aperently A6 was relesed just before 2016)


Tbh, I thought I fixed my problem the other day, I thought it must be the HDD's so I changed the setttings in the BIOS to make them hot-swappable, and then it crashed again 5 minutes ago, after nearly 2 days of no problems.
I think at the moment, I'll keep waiting until MSI makes another bios update before I do any testing

I've had this problem before and it was the SSD I was booting from. No SMART errors or anything to warn of a problem, it just stopped functioning at random unpredictable moments which in turn locked up the system entirely.

Try taking your main SSD out as the next step and install windows onto another drive perhaps?

Hmm, interesting, I dont easily have a spare drive, so I was wondering, would a linux live Cd work?

 

[Warmo161]

Ok, its been a few days, and I havent had any problems, I thought it was because I disabled 'link power management' in the intel rapid storage program on my machine, and it has made no difference. It crashed on me a second ago, so now im running on safe mode

Regarding the last post, making the drives hot swappable made no difference. I also ran my PC with no drives for a day, adding another drive the next, and had no problems

Also, I have tried to look for my spare HDD i was going to use to boot another windows install, but I cant find it, how ironic.

 

[Warmo161]

Wow, you really are having some issues nailing this one, sorry it's been such a pain. Nothing more annoying than having your build throw a wobbly and no matter what you check it all seems fine!!

I would have thought OCCT would have shown the weak link, that thing nails machines hard.

I guess my only last thought would be a setting in the bios as you have touched on. Perhaps the storage controller config?

Is it set to AHCI? Have you changed any config on the storage system since installing windows before it started crashing?

I haven't touched a thing related to SATA settings, I did go through it and you can only select ACHI or RAID (If I could, I would have changed it to IDE) To be fair, the computer was crashing nearly every day after setting everything up, its only now I'm trying to fix it (I don't want it to randomly crash when doing something important)

I reformatted windows last night, and it crashed again this morning. I now think its my new SSD, so just now, I've installed Windows on my OLD SSD and now, only time will tell if my idea is right...

 

[Warmo161]

EDIT:

After 3-4 days with no problems on the old SSD, I decided to process a refund for the new Samsung SSD



And then... It crashes again. Is there anything I havent tried? i'm all out of ideas




Another EDIT:

Ok guys, I need some quick help, if I am to return it. I opened my PC up and removed my CPU from the Mobo, in the process, I found a bent pin in the socket, is this the source of my problems? (Added screenshots below) its at the bottom left, sorry for the ugly photos, I have a good phone, honest!

https://imgur.com/a/Itb2E