Woke up and Miner has 2 GPUs not running

[nathan0658@gmail.com]

I hate posting this on here (instead of searching as I usually do) but I am stumped and need a central place to collect the possible ideas at this point.

Woke up this morning and my Mining Rig (5xRX570 4gb) cards was only running 3 cards.

I restarted the miner and 2 of the cards are still not running. I did have to uninstall the drivers on the remaining cards using the DDU tool and reinstall the block chain drivers and patch them before they started working and but I cannot get the other 2 GPUs to turn on or work.

I have moved around on PCI slots, changed risers, plugged in different PSU (PCIe) cable to see if that was an issue but nothing yet.

if anyone had any fresh idea that I can try tonight when I am off work please respond.

I do plan on trying this also:

1. unplugging all GPUs and plugging only 1 GPU into a slot that is working currently to see if I get a response.
2. I am going to try to use a Riser from one of the GPUs that is currently working also to see if it could still be a bad riser.


I cannot think of anything else that would be causing this issue. It was working great for 1-2 weeks hashing at 29.5 for eth on all 5 cards.

 

[P00P135]

Only thing you can do is test each card individually one by one to see if they work and narrow down the problem.

[nathan0658@gmail.com]

Do you think it could be a SATA power issue?

I am going to try to do 1 card on the PSU so see if I can trouble shoot more. My PSU is 1000w and I am using only 650watts max.

[joshuajones02]

Do you think it could be a SATA power issue?

I am going to try to do 1 card on the PSU so see if I can trouble shoot more. My PSU is 1000w and I am using only 650watts max.

Your best bet is to narrow the problem down and rule if its an issue with the GPU's, riser cards, or motherboard. I would start by connecting the gpu's 1 by 1, well let your machine run and then see which ones aren't heating up and then start with testing those. I spent a week trying to diagnose a similar problem and it ended up that the PCIe slots on the mobo went bad, had to RMA it and finally getting it back this weekend (its been a total of 1 month) decided to just by a better board and use the RMA to diagnose problems in the future

[philipma1957]

Do you think it could be a SATA power issue?

I am going to try to do 1 card on the PSU so see if I can trouble shoot more. My PSU is 1000w and I am using only 650watts max.

Your best bet is to narrow the problem down and rule if its an issue with the GPU's, riser cards, or motherboard. I would start by connecting the gpu's 1 by 1, well let your machine run and then see which ones aren't heating up and then start with testing those. I spent a week trying to diagnose a similar problem and it ended up that the PCIe slots on the mobo went bad, had to RMA it and finally getting it back this weekend (its been a total of 1 month) decided to just by a better board and use the RMA to diagnose problems in the future

spare mobos are the way to go  as they are cheap  plenty of 99 dollar to 129 dollar mobos around.

@ op here is a good mobo for 5 or 6 cards

https://www.amazon.com/GIGABYTE-GA-Z270P-D3-LGA1151-Crossfire-Motherboard/dp/B01N2V0AHI/ref=sr_1_2?


free same day shipping if in the usa  and a prime member.

[bigjee]

I hate posting this on here (instead of searching as I usually do) but I am stumped and need a central place to collect the possible ideas at this point.

Woke up this morning and my Mining Rig (5xRX570 4gb) cards was only running 3 cards.

I restarted the miner and 2 of the cards are still not running. I did have to uninstall the drivers on the remaining cards using the DDU tool and reinstall the block chain drivers and patch them before they started working and but I cannot get the other 2 GPUs to turn on or work.

I have moved around on PCI slots, changed risers, plugged in different PSU (PCIe) cable to see if that was an issue but nothing yet.

if anyone had any fresh idea that I can try tonight when I am off work please respond.

I do plan on trying this also:

1. unplugging all GPUs and plugging only 1 GPU into a slot that is working currently to see if I get a response.
2. I am going to try to use a Riser from one of the GPUs that is currently working also to see if it could still be a bad riser.


I cannot think of anything else that would be causing this issue. It was working great for 1-2 weeks hashing at 29.5 for eth on all 5 cards.

 

Lets start with the simple solutions.
After your steps 1 and 2.
3. Reset any bios mods to your card and run at factory default clock and memory speeds.
Check each card to identify which one is the issue. It seems to me that the card is unstable on current core/mem clocks.
4. If steps 1-3 fail check your power supply with each card installed at a time to see if it runs smoothly.
Crappy PSUs have been known to cause unstable systems. Since you are using only 650w try switching out your PSU to a spare one if you have one on hand.

Report back what you find.

[nathan0658@gmail.com]

Thanks everyone for the replies. I will test when I get off work and post my results.

Prior to this happening, I was mining 29.5 on each card for approx. 147 total on ETH. I am new to mining and I know there is a learning curve on all new things.

I TRY to teach myself like I said but I do appreciate all the people who gave possible fixes.

If this does not correct the issue, I will be returning the 2 GPUs for an exchange and if they do not have the same model in stock I am going to request to pay for the difference and get the 580 8gb cards. fingers crossed I can troubleshoot though.

[nathan0658@gmail.com]

Here is an update on what I have figured out so far. Mining rig only recognizes 2 cards right now.

I did have my GPU risers  plugged into Sata cords instead of to the PSU with a PCIE. So not enough wattage I think after the cord possibly got weak. Wasn't ever hot though.

I had 1 extra sata cords (daisy changed 2-3 total manufacturer) laying around to use for now to see if I could still get all my cards to at least power up and I did so I can at least continue to mine until I get another PSU this weekend to power the risers and the other psu to power the cards. I am just out of PCIE slots on my PSU.

I have seen "pcie splitters" and this would solve my problem also and be heaped. I have a 1000w Raidman PSU and using about 600-650 dual mining 5 cards.

Are these safe to use? What about pcie cord extenders.

Thanks for the feed back on these possible options.

[nathan0658@gmail.com]

Everything is working now!!

I bought a 6th card even though i still couldnt get the 5th kne showing because kf black friday deal haha.

I also bought all new risers and did  a DDU in safe mode and did a fresh install of the block chain drivers.

This has resolved my issue. Now i wish i would habe gotten new risers to bwgin with to see if i had trouble with it.

This should be a good example for new miners like myself that if something goes wrong, check the small things first. Risers, cables, etc...

I will buy a few more to have as spares for the future. Now i just need to flash the gpu5 amd gpu6 amd ill be hashing at 177mh/s with these RX570 4gb.

Thanks to all for the assistance and patience for my dumb questions.