Question Random reboots MPG X570 gaming plus

[papaschtroumpf]

Hopefully posting in the right place...
So yeah, my PC is getting old (5 years) but I'd rather wait to replace it, or at least hand it down to my kids it here is the rub:

Over the last few days it started rebooting randomly. No bluescreen, just reboot (and automatically reboot is off in Windows)
Happens even when the PC is idle, so not tied to high power usage. It happened the day after Windows 11 applied an update, so I thought it might be related, it keep on reading.

I updated drivers and even bios (yikes!), no dice. I e eventually booted a Linux distro from a USB drive but it rebooted while idle within a hour or two, so I think that exonerate Windows.

Visual inspection, everything appears well seated, filters are good, fans running (system monitor in Windows doesn't show overheating).

Overclocking off in the bios.

Ran memtest and passed. I'm guessing it'll reboot eventually but not due to memory errors.

I would usually suspect the power supply but it doesn't seem related to drawing more power. I may still buy a new supply and swap it, it if it's not it I'm out the price of a supply.

It feels like the MB is the most likely culprit. I didn't see any swollen caps but probably need to take a second look. Any advice on how to test the MB?

What would be an equivalent/compatible MB? Seems like they're used and expensive, what could be an affordable MB to rehouse my Ryzen 5 and DDR4, or should I give up and buy everything new?


One test I haven't done is to remove the graphic card and use only the integrated graphics, so I should do that in case it's the graphics card gone bad.

 

[gdansk]

If you do determine it to be the motherboard there are a lot of new AM4 motherboards available yet including other X570 boards for about $130 here.

 

[Shmee]

Anything odd in event viewer? What PSU and GPU are you using? You could also try a CPU stress test, to see if the CPU is unstable. If it reboots right away, that is a big clue.

 

[papaschtroumpf]

GPU is an RX 490
PSU is a Corsair CX650

The system has been running rock solid for almost 5 years.

I'll have to look at CPU stress test, I never considered the CPU could go bad.

I'll have to run windows again and see if the event viewer has anything specific besides "unexpected reboot"

 

[DAPUNISHER]

You are using a 3400G correct?

 

[papaschtroumpf]

Yesterday I ran 3DMark free and it performed all test without crashing (then rebooted later while doing email...)

I just tried a tool called OCCT personal for CPU stress, and I tested CPU only (it has an option to test CPU+Mem)

The core temp shot up to the mid-90s and stabilized there, CPU freq at 3550, MB temp in the high 40s.
There is a bazillion readings reported by the tool some I may be missing something but it looked stable. No WHEA errors while running the test either...

Been running for 20 minutes so far (a full test is 1h) so wouldn't say it's failed "real quick"

EDIT: OCCT CPU test ran for the full hour without reboot.,

 

[papaschtroumpf]

You are using a 3400G correct?

If you mean the CPU, it's a Ryzen 5 3600

 

[DAPUNISHER]

If you mean the CPU, it's a Ryzen 5 3600

Then there is no iGPU to test with and you will get no video without the vid card.

5yrs as you noted, is definitely long enough for the CPU, board, or PSU to have gotten flaky. I'd start with the PSU as I have seen them cause reboots sitting idle.

 

[papaschtroumpf]

the windows Event viewer shows a bunch Critical events (the only criticals in the list):

Event ID 41 The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

about 22 in the last 7 days, 13 in the last 24h, one in June, one in May

Under errors I see: EventID 18:

A fatal hardware error has occurred.
Reported by component: Processor Core
Error Source: Machine Check Exception
Error Type: Cache Hierarchy Error
Processor APIC ID: 0
The details view of this entry contains further information.

17 in the last 7 days, 10 in the last 24h and times match one of the critical above

And I just noticed the "Details" tab, I couldn't figure out earlier where the details view they were talking about was (facepalm)

-

System

- Provider [ Name] Microsoft-Windows-WHEA-Logger [ Guid] {c26c4f3c-3f66-4e99-8f8a-39405cfed220}

EventID 18

Version 0

Level 2

Task 0

Opcode 0

Keywords 0x8000000000000000

- TimeCreated [ SystemTime] 2024-07-16T00:45:14.5652422Z

EventRecordID 105359

- Correlation [ ActivityID] {1b03029e-ab30-45f4-a880-6b149db0ef12}

- Execution [ ProcessID] 5092 [ ThreadID] 5844

Channel System

Computer Psycho

- Security [ UserID] S-1-5-19

-

EventData

ErrorSource

3

ApicId

0

MCABank

10

MciStat

0xfea020000004010b

MciAddr

0x40000018fc512f0

MciMisc

0xd012000300000000

ErrorType

9

TransactionType

2

Participation

256

RequestType

0

MemorIO

256

MemHierarchyLvl

3

Timeout

256

OperationType

256

Channel

256

Length

1163

RawData

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

 

[papaschtroumpf]

funny 40 minutes of OCCT and not a hiccup, when I was trying to perform a backup earlier and couldn't complete in between reboots. Maybe it needed to be pushed to get cobwebs out

Maybe I'll try to do a bunch of disk activities next to see if it could be the HD (I have a 4TB spinner and an 1Tb EVO)

 

[papaschtroumpf]

If you mean the CPU, it's a Ryzen 5 3600

got it, I saw the MB has an HDMI port but I didn't realize the support would be in the GPU not the MB. I never tried to use it. I may have some other graphic card to try, I'll do some GPU torture to see if it causes crashes but I would have thought 3DMark qualified and it ran fine.

 

[Tech Junky]

Event viewer is useless.

I had a random reboot problem with my Intel setup that turned out to be a crappy driver from Intel. Once that was updated everything went back to normal. However since you tested with Linux and had the same issue it's more likely to be hw than a driver.

Maybe start with a PSU tester rather than having to run everything apart to check everything. $20 and 5 mins gives you the answer. From there my next suspects would be the CPU or mobo. There's always the chance something is just loose and needs to be reseated.

 

[VirtualLarry]

In the BIOS:

Restore optimized values
Set XMP profile 1
Disable Global C-States
Set Current Profile: Normal

 

[In2Photos]

How old is your BIOS? My son had an issue with his b550 board. It worked great for a while, then started crashing during games. We suspected PSU but even a different PSU had the same problem. Updated the BIOS and the problem went away.

 

[papaschtroumpf]

There's always the chance something is just loose and needs to be reseated

yeah probably the next step, with a careful visual inspection of everything under bright light.

Funny I haven't had a reboot since I started my post even though I have been toruring the system with OCCT (memory test OK, although I didn;t let it run for hte full hours but it did 200 cycles. Power test OK so far)

How old is your BIOS? My son had an issue with his b550 board. It worked great for a while, then started crashing during games. We suspected PSU but even a different PSU had the same problem. Updated the BIOS and the problem went away.

updated it yesterday, and chipset driver, etc... (and seen reboot since) although I just used the AMD automatic driver detect (to make sure video drivers were up-to-date) and it seems to have updated the chipset drivers (replaced MSI with AMD one?) as well as installed Adrenalin and updated graphic drivers. We'll see if that helped.

 

[papaschtroumpf]

Finally got a reboot in the middle of not doing much (mucking around with excel)
WHEA only shows the " The system has rebooted without cleanly shutting down first..." log.
Time to give it a rest for today, back to taking everything apart in the morning.
Thanks for the help and suggestions, I'm pretty sure the BIOS was set to defaults after update, but I still need to do that per VirtualLarry's advice.

 

[papaschtroumpf]

I may have found the issue and fell like a moron, but relieved, I connected the computer in a different power cable/surge protector (also different keyboad, mouse and display) and stable for hours so far.
I now suspect my surge protector may have been glitching. Still a working theory but would be a lot cheaper to fix!

 

[Tech Junky]

I now suspect my surge protector may have been glitching.

Anything else plugged into it? Seems a bit suspect that a SP would cause this sort of issue as when they fail they keep working and just lack the protection part of the equation.

Pivot-Plug 12-Outlet Surge Protector, 8 ft. Cord | Belkin US

Plug in your electronics with greater ease and flexibility

www.belkin.com

This is what I've been running for a few years now and it just works. Probably overkill now that I've consolidated the number of devices plugged into it but, it's still peace of mind w/ the $$$$ of the devices plugged in.

-- when you have minimal system stress the system is stable?
-- when you are just idling the system reboots...

first suspect is usually the PSU // Linux test you did had the same issue ruling out a SW issue
next based on costs MOBO
next RAM/GPU/etc.
next CPU

At some point though you have to consider if your time and money are worth chasing the gremlins or just rebuild a new system instead? By the time you're done with swapping parts to nail it down you'll have a new system in terms of being able to return parts but same performance as before vs taking the same funds and getting an upgrade in the process. Sticking with AM4 and porting over parts is an option but the gains wouldn't be significant enough for me to spend on that situation. in my case I would be looking at AM5 to make it worthwhile. At least if you switch now to a 7 series you still have the option to drop in an 8 series CPU when the prices drop later on or even a 9 series in a couple of years.

 

[In2Photos]

I may have found the issue and fell like a moron, but relieved, I connected the computer in a different power cable/surge protector (also different keyboad, mouse and display) and stable for hours so far.
I now suspect my surge protector may have been glitching. Still a working theory but would be a lot cheaper to fix!

Are you not using a UPS? If not you should be!

 

[kschendel]

If it's just a MOV surge protector, yes they can go bad either from age or from spikes wearing them down. I replace mine every 3-4 years whether they need it or not. (although the computers are all on UPS's. Only non-essential equipment is on surge protection.)

If that's not it, I think I'd be looking at the CPU. I've heard (unconfirmed!) rumblings that 3600's can age and die more often than some other models. You can probably find a 5600 for fairly cheap and get a modest speed boost as well.

 

[kschendel]

Trying the integrated graphics without the GPU is a good next step to rule out any issues there.

OP said earlier that the CPU was a 3600, so no integrated graphics unfortunately.