Games randomly making PC crash (Linux Mint)

[undiagnosed but pretty sure]

I recently got Linux Mint 17.3 on my backup PC since I didnt want to spend 165 euro just to get Windows on my backup PC. I installed Steam, had a few problems launching it in the beginning but after snooping around a bit I found out how to fix that problem and Steam works fine now. But I have one problem. Whenever I play games (usually Team Fortress 2 and Counter Strike: Global Offensive) at some random point they make the computer crash. They either make the screen have some red/blue dots and then restart or just stop posting and forcing me to cut the power of my computer to restart it. It usually happens around 10-20 minutes into the game, but the most it's gone without crashing while I'm ingame is 1 hour. If anyone needs the specs of my PC, here's a pastebin of them http://pastebin.com/HdKAFSkB

[Zyro]

This sounds like some hardware problem. Maybe it's overheating, did you check that? Dust in the fans?

[undiagnosed but pretty sure]

引用自 Zyro：

This sounds like some hardware problem. Maybe it's overheating, did you check that? Dust in the fans?

I was gonna check now but it's late and I have a busy day tomorrow. Will check tomorrow after I'm done with everything and update. Thanks for the response.

[Zyro]

I once lost a GPU to the dust...
http://ein-eike.de/2013/09/29/holt-den-sauger-raus/
... and it was showing coloured boxes in its agony...

[undiagnosed but pretty sure]

引用自 TDMT Nikolomara：

引用自 Zyro：

This sounds like some hardware problem. Maybe it's overheating, did you check that? Dust in the fans?

I was gonna check now but it's late and I have a busy day tomorrow. Will check tomorrow after I'm done with everything and update. Thanks for the response.

Alright, I cleaned all of the dust and applied some new thermal paste and everything was going good! Until I crashed again. It took longer than usual but that might have just been luck. So the problem isn't dust nor fans not working correctly.

[Zyro]

I really would take a look at the temperatures. There should be plenty of programs out there that do this. Check temperatures, play half an hour, check temperatures again, ...
Of course, it might have other reasons. Memory check? Ah, btw, check game data?

[undiagnosed but pretty sure]

引用自 Zyro：

I really would take a look at the temperatures. There should be plenty of programs out there that do this. Check temperatures, play half an hour, check temperatures again, ...
Of course, it might have other reasons. Memory check? Ah, btw, check game data?

I just did the temperature thing and... whoah. So, my CPU has 3 cores. One is at 37 idle, 49 load. Two is cruising at 35 at all times (I have single core rendering in-game). Three is... 127 idle, 127 load. I google'd a bit but I didn't find anything about this.

[Zyro]

127 degrees Celsius?!?
That a) would surely be the cause for crashing (this might damage your computer!) and
b) be very strange, as even single core load should be moved from core to core.

[Zyro]

Another possiblity would be that sensors or what you are using to measure temperatures doesn't know about your CPU. Or maybe Linux kernel missbehaving on this strange 3 core setup?

[undiagnosed but pretty sure]

引用自 Zyro：

Another possiblity would be that sensors or what you are using to meadure heat doesn't know about your CPU. Or maybe Linux kernel missbehaving on this strange 3 core setup?

Probably Linux missbehaving because I didn't have this issue when this PC was running Windows (I had Windows but the SSD that it was on died and I didn't want to buy another copy of it since in 2 weeks I'm gonna build a new PC so I downloaded Ubuntu 16.04 but it didn't want to work. Then I installed Mint and it worked fine except for those occasional crashes)

[undiagnosed but pretty sure]

Alright, I started looking at the log files and I found this which might be useful. I'm kinda new to PCs so I don't really know what this means

May 30 21:02:51 Nik-pc dbus[755]: [system] Successfully activated service 'org.freedesktop.nm_dispatcher'
May 30 21:02:53 Nik-pc udisksd[1855]: Error probing device: Error sending ATA command IDENTIFY DEVICE to /dev/sdb: Unexpected sense data returned:#0120000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................#0120010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................#012 (g-io-error-quark, 0)
May 30 21:02:53 Nik-pc kernel: [ 44.509100] ata4: SATA link down (SStatus 0 SControl 310)
May 30 21:02:53 Nik-pc kernel: [ 44.509121] ata4.00: disabled
May 30 21:02:53 Nik-pc kernel: [ 44.509158] ata4: EH complete
May 30 21:02:53 Nik-pc kernel: [ 44.509192] ata4.00: detaching (SCSI 3:0:0:0)
May 30 21:02:53 Nik-pc kernel: [ 44.518322] sd 3:0:0:0: [sdb] Synchronizing SCSI cache
May 30 21:02:53 Nik-pc kernel: [ 44.518423] sd 3:0:0:0: [sdb] Synchronize Cache(10) failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK
May 30 21:02:53 Nik-pc kernel: [ 44.518429] sd 3:0:0:0: [sdb] Stopping disk
May 30 21:02:53 Nik-pc kernel: [ 44.518473] sd 3:0:0:0: [sdb] Start/Stop Unit failed: Result: hostbyte=DID_BAD_TARGET driverbyte=DRIVER_OK

[Zyro]

"IDENTIFY DEVICE to /dev/sdb"
... this is trouble with some hard drive.
Any drive you're missing, you cannot access?
Shouldn't make your CPU overheating, though.
I tried to google for problems with triple cores heat, but didn't find anything special...

[undiagnosed but pretty sure]

引用自 Zyro：

"IDENTIFY DEVICE to /dev/sdb"
... this is trouble with some hard drive.
Any drive you're missing, you cannot access?
Shouldn't make your CPU overheating, though.
I tried to google for problems with triple cores heat, but didn't find anything special...

I do have my 1TB drive which is not recognized by Linux. I will disconnect it and see what happens

[Zyro]

What temperature do your cores have after some minutes (or hours) of idling?
TBH I find it more likely that the measurement is wrong than that the kernel is misbehaving that badly.

[undiagnosed but pretty sure]

引用自 Zyro：

What temperature do your cores have after some minutes (or hours) of idling?
TBH I find it more likely that the measurement is wrong than that the kernel is misbehaving that badly.

The temps remain the same when idling, when gaming #1 Fluctuates between 46 and 49 which I think is normal, #2 stays the same as it is not doing anything and #3 stays at 127.

Just checked the temps after removing the drive but #3 is still at 127, but #1 has gone down from an average of 47.5 while gaming to 44.

[Zyro]

It's very unlikely that the kernel is failing that hard, it's unrealistic that a single core has got a temp that high, and it is conspicuous that it's sitting at 127 (the highest number stored in one byte if you allow for negatives, called Int8.Max). I bet against a heat problem.

You should check other stuff, like game data or memory.