Computer restarting when there's "heavy load" on it: hardware fatal error

[carlos1251]

Carlos:
First of all the specs:
CPU - AMD FX-6300
Motherboard - Gigabyte - GA-78LMT-S2 Micro ATX AM3+
GPU - Gigabyte - GeForce GTX 950 2GB Dual WindForce
HDD - Western Digital - Caviar Blue 500GB
RAM - Kingston 2X4 GB DDR3 1333
PSU - Corsair - Builder 500W 80+ Bronze

I have been using this rig for over 4 years, upgrading as the necessity comes. From the original buy there's only the CPU and the motherboard on it right now, everything else is fairly new (2 years max). I have always played games on it and no problem surfaced until recently.
While playing Overwatch the computer restarted, strange but not unheard of, I did had a lot of programs opened while I was playing so I thought nothing of it and moved on. But since then, whenever there's "heavy load" on the pc it'll restart.

I say "heavy load" because it could be a game that I just opened or many tabs on Firefox while discord is open, for an example.

Checking the event viewer I get these two errors before restarting:

Northbridge Fatal error

DCOM Server Error 225

What's your guess?

 

[geofelt]

Your motherboard is a very poor one for a FX-6300.
Perhaps, over time, it has deteriorated.
http://www.tomshardware.com/forum/id-2384030/motherboard-tier-list-am3-chipsets.html

You might have a heat issue.
Take the case covers off and direct a house fan at the innards.

Consider that spending $250 or so will buy you a superior gaming intel 8th gen processor, 8gb of ddr4 and a suitable motherboard.
Such a change will have many upgrades.
I would not spend anything on your current system.

 

[LinuxDevice]

Overclockers will tell you that for a system to be stable at faster clocks, the cooling must be better. Under higher heat a given voltage which was stable under lower heat is no longer stable (faster clocks require higher voltages which in turn cause increased heat). Since you are not overclocking, and assuming hardware is not failing, odds are high that heat has gone up.

The easy step would be to make sure there isn't any dust blocking anything. Dust is a thermal blanket. Make sure fans are working.

The part people often don't think about is that many of the thermal pastes harden and stop working correctly over a few years (I'll never use a thermal paste again which isn't "long life"). After carefully blowing out any dust and perhaps reseating any PCIe cards, replace the thermal compound on the CPU. If it has hardened at all, then you probably found the problem. If it hasn't hardened, then you would have done something good even if it doesn't fix the issue.

FYI, sometimes parts do become more marginal over time, including power supplies.