Random reboots + boot loop

Mckrau

Commendable
Feb 11, 2017
7
0
1,510
Hi guys, I'm having an issue with a new rig I've just built. I'm getting random reboots (sometimes after a couple of hours, others after 20+), everything works perfectly, then the system reboots and goes into a boot loop. After six or seven tries (or after letting it rest for a minute) everything loads and back to square one. First thing I thought was overheating, even though the reboots come absolutely at random times (sometimes I'm working on TRADOS, sometimes I'm playing games or the computer is just idling) but nothing goes over 40 degrees even when stress testing with either prime95 or Heavy Load. Every single component is new, but I still run memtest (no errors) and let Heavyload run for some hours. Next thing I thought was PSU issues (also brand new, but who knows, right?). But again, stress testing with Prime95 the values for +15v and +5v that were shown in HWinfo were perfectly fine.

I disconnected every single power cable just in case and reconnected to make sure everything was properly connected. Still getting the random reboots.

From what I understand, there are some issues with Gigabyte's Power Loading BIOS feature so I tried running with it both on and off. Doesn't seem to make a difference.


I'm absolutely at a loss here, and any kind of help would be most welcome, I've been building computers for quite a long time know and this is the first time I run into a problem that I can't at least identify. Next step is unplugging everything but the mobo, ssd and a memory stick and running the computer like that, adding the rest of the components one by one until finding the culprit, but since sometimes the computer runs without a hitch for a day or so it would be really time consuming.

The box and everything in it, again, everything brand new:

i5-6600k
mobo gigabyte GA-Z170-HD3
gigabyte Gforce GTX 1070 G1 Gaming
SSD Kingston SSD Now UV400
Corsair Vengeance LED 2x8GB DDR4 3000mhz
PSU TACENS MP700W 200-240v
Corsair hydro f45
Nox Coolbay sx Red devil
Running windows 10 pro edition and latest drivers.

I'd be grateful for any ideas.
 
Solution
Ok. When you mention boot loop, what happens exactly? Partial OS load? Anything show up on your monitor? I'm leaning on a PSU issue at the moment. Even a driver crash or software bug shouldn't cause your system to take 6 or 7 attempts to get up and running again.
No BSOD, random crashes and windows is not doing any memdumps, I normally use whocrashed and it can't find any crash dumps. nothing goes over 45º at anytime, and the psu voltages always stay way above minimum when I check logs. I never tried Event Viewer, installing it right now. Thanks :)

 


Nothing on Event Viewer, or at least nothing eye popping.
 
Ok. When you mention boot loop, what happens exactly? Partial OS load? Anything show up on your monitor? I'm leaning on a PSU issue at the moment. Even a driver crash or software bug shouldn't cause your system to take 6 or 7 attempts to get up and running again.
 
Solution


Sometimes it takes only 2 or 3 tries, sometimes more, sometimes the crash comes before the bios screen shows up, sometimes a little after (always after the normal BIOS beep but before the windows loading screen comes up). Could it be a CPU issue? Never ran into problems with processors but from what I understand in that instance the crashes are random but do not lead to boot looping...
 
Its possible, but CPU is the least likely hardware item to be faulty. Being that some crashes are before UEFI screen shows up, I find it unlikely to be an OS issue. You may need to try your lengthy original test plan. Also, no overclock on anything? Are you using XMP profile?
 



No overclocking was running, but I just downclocked by setting the cpu multiplier to 80.00 so it's running at 3,000 Mhz right now, memory also dc to 2,133 from 3,000. XMP profile was never active. I'm doing this out of desperation more than anything... If I hit a couple of days of uptime like this I'll resign myself to it been an undetected PSU issue, since I believe downclocking reduces power consumption/temperature? Since I'm positive it's not the temp, well...


 


I finally got the chance to try a different PSU, it seems to have done the trick, 24 h uptime and running... I'll increase uptime incrementally until I'm absolutely confident the issue is solved. Thanks for the help 😀