ETH rig not working suddenly

[rasmukri]

I'm at my wit's end here I can't figure out WTF is wrong with my mining rig.
Today it just crashed randomly and I can't get it to work again.
Did DDU and installed 20.8.1
Running 5.4c and tried 5.5c with no change
Tried changing my OC on afterburner to stock speed or lower no change
I got 6x 570 8gb, biostar TB350-BTC, AMD ryzen 3 1200, and 8gb 2400 ram. the computer turns on and runs just fine and I can use it with no issues until it try to mine then it goes to shit.
Usually within seconds of the last GPU DAG generated thing coming up and all 6 cards reading their individual speeds screen goes blank and I gotta reboot.
Tried changing the config file to -gt 0 instead of 15 no change. actually it got worse when I did that as soon as the auto tune popped up it crashed in seconds before the DAG was even done I think. I don't but it def crashes faster when the -gt is set to anything besides 15 which is where it's been at for like a year and a half with no issue.
I've also cleared the CMOS and redid the bios settings, I also tried loading the bios defaults. I also just updated the bios to the newest one with no change, I was only 1 bios very behind the current so not much of an update I suppose. Although after the bios update it mines for a second then right before it crashed I noticed GPU4 went from 35c to 513c then it crashed but with a BSOD instead of a black screen, bsod said vga failure. Normally mining temp is like 65-70c but since it is just starting that's why it's 35c. How can it runaway spike like that?
So I did 1 card at a time and ran for 5 min and nothing happened. Then I ran 4 nothing then 5 nothing all 6 are running for 9 min as I type this and all is well running at stock everything no dials have been adjusted in afterburner. I'll keep it running and monitor till I go to bed. But I still don't what's wrong. This seems like a I fixed it and I didn't do anything situation and it's pissing me off.

[socks435]

Why not switch to another OS instead if you only have one rig you can use hiveos it's free for 1 rig but if you are using multiple rig go for simpleminer paid 2$ a month per rig.

Have you tried to use other miner? T-rex or phoenix?

[rasmukri]

Why not switch to another OS instead if you only have one rig you can use hiveos it's free for 1 rig but if you are using multiple rig go for simpleminer paid 2$ a month per rig.

Have you tried to use other miner? T-rex or phoenix?

I am currently using phoenix miner and I don't see the reason to change mining software. It is still working at stock speeds but i just can't figure out what was causing the crashes in the first place.
If it still works in the morning I'll try to move it back to 1850 on the memory and see what happens.

[buraz11]

Usually when i saw 511 degree celsius temp the riser was dead, maybe one of the risers was a bit off

[siaomond]

I am currently using phoenix miner and I don't see the reason to change mining software. It is still working at stock speeds but i just can't figure out what was causing the crashes in the first place.
If it still works in the morning I'll try to move it back to 1850 on the memory and see what happens.

well sir, back to basic.
unplug all your GPU
test mining with 1 GPU at the riser with other 5 empty.
after clocking and its not dead, add in another GPU to the next riser.

Mining requires alot of patience, esp on the troubleshooting part.

cheers.

[Bitbtc8]

Relax mate, something is wrong with one of the risers, just a guess though, remove all the graphics card and start testing one by one, make sure they hash before jumping to the other, if everything hash then the graphic cards are all fine, now remains the risers, use a single graphic card to test the risers one after the other

[safar1980]

behind the current so not much of an update I suppose. Although after the bios update it mines for a second then right before it crashed I noticed GPU4 went from 35c to 513c then it crashed but with a BSOD instead of a black screen, bsod said vga failure. Normally mining temp is like 65-70c but since it is just starting that's why it's 35c. How can it runaway spike like that?

Remove this video card from the farm and its raiser for a while.
Test this video card on a separate computer if you have the ability.
Temperature sensors can give errors, this is sometimes found on video cards.
Check the robotic ability of the farm from 5 video cards within 24-48 hours

[sxemini]

Usually when i saw 511 degree celsius temp the riser was dead, maybe one of the risers was a bit off

Yep this indicates a dead riser,  dead riser usb cable, dead riser pcie adapter in the slot or dead riser electricity cable.

[rasmukri]

Well shit, ok it's been going fine all night at stock settings.
But I will crank up the oc and monitor the temp and see if I can decect which one is crashing. Doing one riser at a time sounds like a pain. Although yeah I know one at a time is always how to diagnose stuff. I never would've thought a riser or something went to shit that wasn't even on my radar as a possibility.
I've had them fail in the past but it was obvious as the power cable melted or something showed signs of burning or melting

[rasmukri]

New update.
I tried one card at a time and went through all 6 cards then all 6 riders can't get the errors to happen again.
However a new problem has presented itself
I think it's caused by windows or radeon downloaded the new driver because it crashed and I did a DDU and put 20.8.1 amd it works fine again. How do I stop it from doing that shit?

https://ibb.co/QKYnYP8

https://ibb.co/QKYnYP8

[Mine On]

Did you modify your settings for virtual memory? I usually run 16 to 35gigs of virtual memory. I'm running newest AMD drivers with the newest version of Phoenix with no issues. Modded Bios in GPU's? Patch drivers?

[rasmukri]

Did you modify your settings for virtual memory? I usually run 16 to 35gigs of virtual memory. I'm running newest AMD drivers with the newest version of Phoenix with no issues. Modded Bios in GPU's? Patch drivers?

I have the VM set to 30-45gb since I have 6 cards I need at least 29gb according to phoenix miner when it opens up.
I have the stock bios on the cards. Bios flashing on a card without dual bios is not something I'm gonna do.
It only goes to shit when the radeon driver updates. How do I stop it from auto updating?

[miner29]

Did you modify your settings for virtual memory? I usually run 16 to 35gigs of virtual memory. I'm running newest AMD drivers with the newest version of Phoenix with no issues. Modded Bios in GPU's? Patch drivers?

I have the VM set to 30-45gb since I have 6 cards I need at least 29gb according to phoenix miner when it opens up.
I have the stock bios on the cards. Bios flashing on a card without dual bios is not something I'm gonna do.
It only goes to shit when the radeon driver updates. How do I stop it from auto updating?

Switch to Linux or a Linux based mining OS.

Solves it.  No VM needed...4gb ram and 8-16gb thumbdrive is all that is needed for the mining OS...no other storage needed.

[Wotan Wipeout]

Give simplemining a try. Its free for 1 Month.
I have nearly all of my rigs there.

[Mine On]

Did you modify your settings for virtual memory? I usually run 16 to 35gigs of virtual memory. I'm running newest AMD drivers with the newest version of Phoenix with no issues. Modded Bios in GPU's? Patch drivers?

I have the VM set to 30-45gb since I have 6 cards I need at least 29gb according to phoenix miner when it opens up.
I have the stock bios on the cards. Bios flashing on a card without dual bios is not something I'm gonna do.
It only goes to shit when the radeon driver updates. How do I stop it from auto updating?

I am using version 20.12.01. Which is newest driver. Running 6 RX580 GPU's. You can shut off "Check for updates" in Radeon sotware.

[rigomortis]

try connecting 1 gpu at a time to find out which is the bad riser.