MP 6,1 Mac Pro 6.1 2013 shuts down then in a hour or 2 it can be turned on again?

[MacZoltan]

Hi,

I have a 2013 mac pro 12 core D700 128GB memory unit which acts really strange.

It starts up and runs for a while just fine, geekbench runs Heaven runs browsing etc...
Then it shuts down and from that point when you press the power button the lights not turn on nor the mac pro.
If you let it sit for an hour or 2 then it runs agan like nothing happened, all the apps open up I had running before and no crash log either at least no pop up stating there was an error or crash at all.

I reinstalled the OS and the same happened, clean install Mojave runs ok, Heaven Geekbench does too then shuts down some point and I have to give it an hour or 2 before turnin it on again.

 

[MacZoltan]

 

[krakman]

The only way to find out is take it to an apple store and have them to provide a repair quote, it will state what part need replacing

 

[rpmurray]

Maybe it's overheating and shutting down to protect itself. Have you opened it up and checked if it needs cleaning? You can install an app like iStat Menus or TG Pro to monitor temperatures and see if that's what's happening. Also check the console logs to see what's happening at the time it shuts down.

 

[MacZoltan]

The only way to find out is take it to an apple store and have them to provide a repair quote, it will state what part need replacing

is the quote free?

 

[MacZoltan]

Maybe it's overheating and shutting down to protect itself. Have you opened it up and checked if it needs cleaning? You can install an app like iStat Menus or TG Pro to monitor temperatures and see if that's what's happening. Also check the console logs to see what's happening at the time it shuts down.

console logs like?

 

[rpmurray]

Open the Console application in Applications, check system.log and any of the other logs to see if they give any clues.

 

[arche3]

Thermal issues imo. Prob related to graphics cards. Or powers supply.

 

[MacZoltan]

console logs like?

cleaned, temp monitor shows no extreme temps and which log you refer exactly?
There are many like user logs system logs, i searched in them for crash sections but no luck so far to figure it out if it is CPU/GPU or memory.

Replaced the memory, did not help, replaced the logic board , also did not help.

 

[mikas]

That sounds irritating to troubleshoot. Maybe try terminal;
"log show --predicate 'eventMessage contains "Previous shutdown cause"' --last 24h"
or last 48 hours or any more appropriate value.
And if anything comes out of that, then try to interpret the number somehow to proceed.

Previous Shutdown Causes Explained - George Garside

Updated for macOS 11 Big Sur, decode your Mac's previous shutdown cause code number for macOS, explaining what the number means and how to fix the issue.

georgegarside.com

How to determine Mac shutdown causes

This guide will show you how to determine why exactly your Mac shut down or restarted using macOS's built-in logging.

www.ifixit.com

 

[MacZoltan]

That sounds irritating to troubleshoot. Maybe try terminal;
"log show --predicate 'eventMessage contains "Previous shutdown cause"' --last 24h"
or last 48 hours or any more appropriate value.
And if anything comes out of that, then try to interpret the number somehow to proceed.

Previous Shutdown Causes Explained - George Garside

Updated for macOS 11 Big Sur, decode your Mac's previous shutdown cause code number for macOS, explaining what the number means and how to fix the issue.

georgegarside.com

How to determine Mac shutdown causes

This guide will show you how to determine why exactly your Mac shut down or restarted using macOS's built-in logging.

www.ifixit.com

updated to Big Surr no issues, but as soon as i try to run something intensive, this time geekbench it shuts down.
I dont even have to wait long just unplug the power cord and press the power button a couple of times and it restarts.

Used the command you gave me and got the result as pictured.
-2?

 

[mikas]

Shutdown Codes

CIT (UK) - help desk and customer service portal

www.centralit-helpdesk.co.uk

A positive code indicates a software or HID initiated shutdown, while a negative code (in the range of -1 to -127) indicates hardware-related Intel SMC-initiated shutdown.

Could not find -2 from anywhere tha available lists, but unfortunately it points to hardware error:
Maybe GPU, CPU, RAM. Did you check the ECC errors? Maybe bent pins at the CPU socket after CPU swap, or a loose/tight screw after procedure.

Did you run CPU or GPU tests in GeekBench? Maybe try tasking GPU only, if it shuts down quicker than with CPU test, it might be the GPU warming up.

I can't remember the spesific cause for early GPU failures in 6,1, might have been a weak solder somewhere, or some other component frying/drying in there, so I guess we don't know exatly how/why the fault will trigger itself.

I'm not sure, but maybe you can get away from a faulty GPU with an eGPU, but then again, maybe not. Someone must have tried that already.

Good Luck.

 

[MacZoltan]

Shutdown Codes

CIT (UK) - help desk and customer service portal

www.centralit-helpdesk.co.uk

A positive code indicates a software or HID initiated shutdown, while a negative code (in the range of -1 to -127) indicates hardware-related Intel SMC-initiated shutdown.

Could not find -2 from anywhere tha available lists, but unfortunately it points to hardware error:
Maybe GPU, CPU, RAM. Did you check the ECC errors? Maybe bent pins at the CPU socket after CPU swap, or a loose/tight screw after procedure.

Did you run CPU or GPU tests in GeekBench? Maybe try tasking GPU only, if it shuts down quicker than with CPU test, it might be the GPU warming up.

I can't remember the spesific cause for early GPU failures in 6,1, might have been a weak solder somewhere, or some other component frying/drying in there, so I guess we don't know exatly how/why the fault will trigger itself.

I'm not sure, but maybe you can get away from a faulty GPU with an eGPU, but then again, maybe not. Someone must have tried that already.

Good Luck.

all checked, i think this is the infamous D700 issue Apple made a free repair till 2018 in a recall program.
The memory i replaced and the CPU so i would guess the D700s are messing up

 

[mikas]

Unfortunately that is the best guess (like Arche3 predicts earlier too).

 

[MacZoltan]

So I took it to a Genious bar as they provide free of charge diagnostics compared to third party repair shops.
And if the D700 is failing they can fix it for £500 which is still cheaper then the aftermarket price of a dual card replacement.
But they found nothing...
More importantly now it works just fine, the only change they did is reseted to factory defaults and installed Catalina.
Very strange as before clean instal of Mojave crashed, Big surr crashed, this catalin a works. (Test i used is davinchi resolve benchamrk/heaven and geekbench 3.3 registered running paralell all maxed settings)
Either they did something they can do and we cant in terms of resetting (i did PRAM reset before)
or I dont know

 

[mikas]

That's kind of strange, but anyway let's give that genious a ?.

 

[th0masp]

If you just got it back wait a little longer before claiming it to be fixed for good. I had those same symptoms on mine and every time I got it back from repairs it was good for a week or so at least, then bang! and back to where it started in ever shorter intervals.

I'm not sure what they did to mine that finally fixed it but last time it was in the shop the power supply was considered the most likely issue. 9 months stable now btw.

 

[theSeb]

Sounds like they reset SMC or NVRAM or both.

 

[complaxneoda]

my Mac Pro 6,1 show some similar problems. when I push the cpu 100% and increase the fan speed to 1900rpm (100%), my screen will turn black and cannot wakeup anymore, but I still hear the fan noise. Hard reboot can makes it alive. this setting worked well before I upgrade to 128gb RAM. do you try your test in 64 or lower ram amount situation?