Bitcoin Forum

Hey guys!

Hoping some of you can help me out here. I'm having an issue with a rig. It was running solid for weeks on end. Then I updated to the new AMD Mining drivers, and it seemed stable for another week or so (don't believe it is the drivers). Now, I'm getting random reboots. System just shuts down.

Wondering if it could be a faulty PCI-E adapter or failing card. If I just let the system reboot, 1 card is not detected. If I turn off power completely, and bring it back up, it is detected.

I can mine for an hour, sometimes 4-5, but it always happens.

Any ideas?

the mining software itself is rebooting?
what are you using?

No, the whole PC is rebooting. Like the power goes off, then comes back on.

What miners are running when the system hangs or reboots? Have you checked the logs? Have you checked the Windows error logs?

I know it's hard to catch a pc rebooting sometimes.. Walk away, miss the error message..ha been there.
But sounds like an OS issue to me. Mining software normally wouldn't reboot your whole system..

Try and catch that error message and see whats causing it to reboot. Trying to see if the older drivers are stable could rule out any other issue..

One of my 6 x 1080 ti rigs use to reboot my software consistently due to a partially faulty riser..
"Gpu 0,1,2,3,4 stuck.. reboot attempted"



Worst case.. change drivers, or clean install of the OS.

Mining software or specifically a GPU crash can absolutely cause the system to reboot. In fact it's one of the most common problems in Windows. A GPU can crash from too much overclocking or defective hardware like risers, causing it to hang when communicating with the miner and the miner will reload in an attempt to resolve the problem. In which case the miner will log which GPU caused the hang and restart the miner, or close depending on how it's configured. If it's a hard crash that can't be resolved by the miner restarting, the miner can crash while restarting and cause the system to reboot or freeze.

You also don't need to be there to read an error message displayed before an OS crash. That's what the Windows error logs are for.

I misunderstood his issue and thought he was saying it was completely shutting his system down. Not just rebooting the system. nvOC had restarted several of my rigs when I started mining so i'm aware. A few friends recently had rigs completely shut off and couldn't get a hold of an error message as smOs has no activity log.. Unfortunately i'm adept with GPU crashes, hanging GPU's, bad risers, and troubleshooting the whole nine yards.

And yeah windows error logs are convenient, I use linux based rigs and never had the need to use whatever log to monitor past error messages, monitor plug in or smOs remote screen capture.

Try swap out your adapter/riser for newer one one at a time, are you doing any oc'ing?

Check the wires on your PSU... Maybe it can not handle the "stress"..

I burned down my 1000W gold+ psu with dual mining ETH+DCR 2 days ago... The yellow wires became black... And the plastic on MOBO (20+4 pin) were melted together.

I guess the probable reason is that it goes to sleep, you could turn off the sleep if you use Windows.
Another reason may be the energy consumption for your rig is too high, you should use more advanced PSU.

If you use 6 cards rig，the PSU should be more than 1250W, and the stability of PSU is also important for the rigs,
Don‘t buy too cheap PSU, the watt in it is not real. I have that experience, all my cheap PSU are no longer use now.

Yup, use modular power and a high efficiency power supply.

I would look into either a power or heat issue. It may be best to take your rig apart, clean out the cards and other components to remove any built up dust and reassemble it. I occasionally have a rig do this too that has been running for months beforehand, and a good cleaning and re-seating of the components and all connectors usually does the trick.

Also check if you have any SATA adapters in use and if you do no more than two per chain, and these should only be used to power risers. I prefer to avoid using the SATA power adapters completely and use PCIe connectors or if I "have to" use the 4 pin Molex.

If after doing the above the issue remains, then start swapping out risers as they do occasionally go bad, usually because of power problems.

You should give us more information on your mining rig, such as PSU, GPU, mobo and all hardware specs as detailed as possible. And OS version + mining software. There are too many reasons to cause a reboot.

Hi  @themonkii, May I know what is your specs of your rig like how many GPUs installed, your OS, and etc. So that We can troubleshoot your rig's issue. :)  ;)

Sorry for the lack of updates on this thread guys, been busy as anything. In the end, I went back to old drivers. The new mining BETA drivers from AMD were causing the issue. I realized that once I tried to apply them to a second rig, and it did the same thing.

So all know, both rigs ran solid for 2-3 weeks straight, no issues prior. LOTS of power 2 x 1000w EVGA Gold modular PSU's, and if it had been a riser, I would have run into the issue prior.

In both rigs a fresh version of Win 10 was installed, and all settings set. I run 4 rigs guys, so this isn't my first trip to the rodeo. I just thought maybe someone else was experiencing issues with these drivers.

What I didn't do, was try running the cards without OC/Voltage adjustments. I just kept what I had been using on the old drivers. This might be the only culprit that I can think of.

I'm going to wait till AMD releases their next version of these drivers before I attempt it again though

Battling with the same issue, need to understand which OS is best to use for mining.

In youtube someone recommended Win 8 and some Win 10.

On Win10 I am failing, every few hours.

I noticed that problems with rigs appear when either the operating system or drivers are updated. I installed new drivers several times, there was no increase in profit. Disabled all updates and rig is running stably.

From my work experience two are the most common reasons for continuous reboots, low quality power supply and Gpu risers. If you have for example an Onda with 8 Pcie x16 onboard slots , I have never had any issues with them as no risers were needed. Also corrupted or old graphic driver can make reboots happen.

Things I would change, risers to the latest version with all the different connections, also ddu and reinstall graphic driver if Psu is Ok.

try msos
my rig is rock solid after weeks of investigation on windows 10 random reboot