[OS] nvOC easy-to-use Linux Nvidia Mining

[Chieftec]

My search nvOC dont get results, sorry.
What i can get this command (~/nvOC report)

My nvos write - xorg problem detected, and reboot again and again

Don't quote me on this but I think xorg problem refers to you booting using your onboard video rather than the graphics card.. you should use PEG not IGD

I connect the monitor to the video card, not the motherboard, but the problem remains

Did you connect monitor to integrated at first boot?
If yes I think easiest way is to re-image, disable IGD, connect monitor to main PCIe slot and boot.

Thanx, i make other way: write image nvOC 19 1.4 and this problem solved.

Mining proccess starts, but if i write "sudo su", system give me message "unable to resolve host 19_1_4"

What is it?

[arif_habibjos]

So i have used the new nvoc 14, and its great Smiley

I have a Zotec Geforce GTX 1070 amp card.... and i want to know

   1. how do i switch the led lights off?
   2. How can i overclock the card in Ubuntu?

   many thanks for your help in advance

[MassDestruction]

Hey everyone. I'm having an issue. My miner tends to reset on me, and when it does it seems like the operating system gets into a bad state, and can't be restarted again unless I rewrite the storage device with a fresh copy of nvoc. Here is information on my miner

PSU: Corsair AX860
Motherboard: Biostar TB250 Pro (has 12x PCIE slots)
Video Cards: 5 GTX 1060s and 1 1050 TI
Storage: SanDisk 120 GB connected via StarTech USB 3.0 to 2.5 Sata HDD/SSD cable
Processor: i5 6600k with Hyper 212 cooler(to be replaced soon with G3900 with stock cooler)
Ram: 16GB
Current Power Draw: 690 watts

Here are shots of the miner in case anyone is interested.
https://imgur.com/Pkes5Su
https://imgur.com/S49LlKl


I'm currently mining ETH, and I'm using the default clocks in the oneBash. Which is 100 for the core, and 100 for the memory.

During my debugging process while trying to figure out what was going on I set the bios to use integrated graphics, and plugged my monitor into the integrated graphics port on the mother board. I was able to pick this up on screen.
https://imgur.com/61g5FYP

Usually when i do this the miner will boot loop(A normal quirk for the TB250). So I'm guessing this is related to the problem I have, but I'm not sure how to move on from here

That's really strange ...

You should not connect your monitor to IGD, disable IGD in bios and connect monitor to the graphic card on main PCIe 16x.
Since you have 5x1060 and 1x1050, connect your 1050 to that slot so it  become GPU0 for easier settings in individual power and oc.
When using mix cards make sure to set the correct values for each card with individual power limit, and oc

You can find your GPUs  by running this if you are on v0019-2.0

Code:

./nvOC gpumap

Why connect SSD with USB converter cable? connect it directly with sata cable and see how it goes.
How you connect riser powers?  golden rule is 2 risers for each cable.

Yes I normally connect it to the 1050 TI in the PCIe 16x slot. I only started using IGD to try and debug things. It's currently set back to normal, and I'm still having the same issues.

I use the StarTech USB 3.0 to 2.5 Sata HDD/SSD cable because the author of nvoc recommended it.

For the risers I'm using no more than 2 sata for one

Also I assume the oc values represent a MHZ offset right?

[papampi]

My search nvOC dont get results, sorry.
What i can get this command (~/nvOC report)

My nvos write - xorg problem detected, and reboot again and again

Don't quote me on this but I think xorg problem refers to you booting using your onboard video rather than the graphics card.. you should use PEG not IGD

I connect the monitor to the video card, not the motherboard, but the problem remains

Did you connect monitor to integrated at first boot?
If yes I think easiest way is to re-image, disable IGD, connect monitor to main PCIe slot and boot.

Thanx, i make other way: write image nvOC 19 1.4 and this problem solved.

Mining proccess starts, but if i write "sudo su", system give me message "unable to resolve host 19_1_4"

What is it?

Thats a known error in v0019-1.4
You can edit  hostname and hosts in /etc and set the correct values and reboot
Or you can update to v0019-2.0 with update script and let it fix it and update to v0019-2.0

Code:

cd /home/m1/Downloads
wget https://www.dropbox.com/s/cyo86xoqotplges/nvOC-19-2-update
bash /home/m1/Downloads/nvOC-19-2-update

[papampi]

Hey everyone. I'm having an issue. My miner tends to reset on me, and when it does it seems like the operating system gets into a bad state, and can't be restarted again unless I rewrite the storage device with a fresh copy of nvoc. Here is information on my miner

PSU: Corsair AX860
Motherboard: Biostar TB250 Pro (has 12x PCIE slots)
Video Cards: 5 GTX 1060s and 1 1050 TI
Storage: SanDisk 120 GB connected via StarTech USB 3.0 to 2.5 Sata HDD/SSD cable
Processor: i5 6600k with Hyper 212 cooler(to be replaced soon with G3900 with stock cooler)
Ram: 16GB
Current Power Draw: 690 watts

Here are shots of the miner in case anyone is interested.
https://imgur.com/Pkes5Su
https://imgur.com/S49LlKl


I'm currently mining ETH, and I'm using the default clocks in the oneBash. Which is 100 for the core, and 100 for the memory.

During my debugging process while trying to figure out what was going on I set the bios to use integrated graphics, and plugged my monitor into the integrated graphics port on the mother board. I was able to pick this up on screen.
https://imgur.com/61g5FYP

Usually when i do this the miner will boot loop(A normal quirk for the TB250). So I'm guessing this is related to the problem I have, but I'm not sure how to move on from here

That's really strange ...

You should not connect your monitor to IGD, disable IGD in bios and connect monitor to the graphic card on main PCIe 16x.
Since you have 5x1060 and 1x1050, connect your 1050 to that slot so it  become GPU0 for easier settings in individual power and oc.
When using mix cards make sure to set the correct values for each card with individual power limit, and oc

You can find your GPUs  by running this if you are on v0019-2.0

Code:

./nvOC gpumap

Why connect SSD with USB converter cable? connect it directly with sata cable and see how it goes.
How you connect riser powers?  golden rule is 2 risers for each cable.

Yes I normally connect it to the 1050 TI in the PCIe 16x slot. I only started using IGD to try and debug things. It's currently set back to normal, and I'm still having the same issues.

I use the StarTech USB 3.0 to 2.5 Sata HDD/SSD cable because the author of nvoc recommended it.

For the risers I'm using no more than 2 sata for one

Also I assume the oc values represent a MHZ offset right?

Thats recommended for easy swapping and test run, better to direct connect which is much faster.
OC values are Mhz as you said
Power limit is actual value in WATT

[papampi]

So i have used the new nvoc 14, and its great Smiley

I have a Zotec Geforce GTX 1070 amp card.... and i want to know

   1. how do i switch the led lights off?
   2. How can i overclock the card in Ubuntu?

   many thanks for your help in advance

GPU LED control is available in v0019-2.0
have a look at 2 posts before to see how to update your v0019-1.4 to v0019-2.0

All settings available in 1bash

LED settings(V19-2.0):

Code:

CONTROL_GPU_LED="YES"       # YES or NO # Control GPU LED light. NOTE: It only works on some models

LED_BRIGHTNESS=100          # 0 - 100 # Set the LED brightness intensity. 0 = OFF, 100 = full brightness

Overclock settings (available in all versions):

Code:

POWERLIMIT="YES"            # YES or NO # Choose YES to set Global Power Limit for all GPU's

POWERLIMIT_WATTS=75         # Global Power Limit in watts for all GPU's. If enabled, INDIVIDUAL settings will override this. 

__CORE_OVERCLOCK=100        # Global Core overclock for all GPU's if not using INDIVIDUAL settings

MEMORY_OVERCLOCK=100        # Global Memory overclock for all GPU's if not using INDIVIDUAL settings

[MassDestruction]

I went through my restart log and saw this. Perhaps this is the cause of my instability, and why the NVOC wont boot.

Mon Jan 15 05:00:29 EST 2018 - Lost GPU so restarting system. Found GPU's:

Failed to initialize NVML: Driver/library version mismatch

Can anyone provide any insight as what could potentially be happening?
Are there any other more insightful logs I could pull up?

[papampi]

I went through my restart log and saw this. Perhaps this is the cause of my instability, and why the NVOC wont boot.

Mon Jan 15 05:00:29 EST 2018 - Lost GPU so restarting system. Found GPU's:

Failed to initialize NVML: Driver/library version mismatch

Can anyone provide any insight as what could potentially be happening?
Are there any other more insightful logs I could pull up?

It seems nvidia driver had a corrupted update, what was your nvOC version.

As soon as rig started, close gnome-terminal and run :

Code:

sudo dpkg --configure -a

It should fix the corrupted update.
I recommend update to latest

[Chieftec]

My search nvOC dont get results, sorry.
What i can get this command (~/nvOC report)

My nvos write - xorg problem detected, and reboot again and again

Don't quote me on this but I think xorg problem refers to you booting using your onboard video rather than the graphics card.. you should use PEG not IGD

I connect the monitor to the video card, not the motherboard, but the problem remains

Did you connect monitor to integrated at first boot?
If yes I think easiest way is to re-image, disable IGD, connect monitor to main PCIe slot and boot.

Great! Thanx a lot!
Thanx, i make other way: write image nvOC 19 1.4 and this problem solved.

Mining proccess starts, but if i write "sudo su", system give me message "unable to resolve host 19_1_4"

What is it?

Thats a known error in v0019-1.4
You can edit  hostname and hosts in /etc and set the correct values and reboot
Or you can update to v0019-2.0 with update script and let it fix it and update to v0019-2.0

Code:

cd /home/m1/Downloads
wget https://www.dropbox.com/s/cyo86xoqotplges/nvOC-19-2-update
bash /home/m1/Downloads/nvOC-19-2-update

[Avarets]

There is error in 3main file (seems to be present long time).
If you are using P106-100 cards only then it is not possible to use individual overclock settings (which is my case since one card probably has Hynix memory).
You should modify 3main file:

Code:

if [ $P106_100_FULL_HEADLESS_MODE == "YES" ]
  then
    if grep -q "XORG_UPDATED" /home/m1/xorg_flag;
    then
      sleep 2
      UPDATE="NO"
    fi
    if [ $UPDATE == "YES" ]
    then
      sudo nvidia-xconfig --enable-all-gpus --cool-bits=12
      cd /home/m1
      echo XORG_UPDATED > '/home/m1/xorg_flag'
      sleep 4
      echo "XORG UPDATED"
      echo ''
      echo "Rebooting in 5"
      echo ''
      echo "disconnect monitor if connected"
      sleep 5
      sudo reboot
    fi
    sleep 4
    ${NVD} -a GPUGraphicsClockOffset[2]=${__CORE_OVERCLOCK}
    ${NVD} -a GPUMemoryTransferRateOffset[2]=${MEMORY_OVERCLOCK}
fi

Instead of

Code:

    ${NVD} -a GPUGraphicsClockOffset[2]=${__CORE_OVERCLOCK}
    ${NVD} -a GPUMemoryTransferRateOffset[2]=${MEMORY_OVERCLOCK}

You can use

Code:

    gpu=0
    while [ $gpu -lt $GPUS ]
    do
      for j in $TI
      do
        CORE=${__CORE_OVERCLOCK[${gpu}]}
        MEM=${MEMORY_OVERCLOCK[${gpu}]}
        ${NVD} -a [gpu:$gpu]/GPUGraphicsClockOffset[${j}]=$CORE
        ${NVD} -a [gpu:$gpu]/GPUMemoryTransferRateOffset[${j}]=$MEM
      done
      gpu=$(($gpu+1))

One more, tried to use v0019-2.0. But claymore consistently crashed. Had to revert to v0019 (not 1.4). Maybe some driver issue.

[MassDestruction]

I went through my restart log and saw this. Perhaps this is the cause of my instability, and why the NVOC wont boot.

Mon Jan 15 05:00:29 EST 2018 - Lost GPU so restarting system. Found GPU's:

Failed to initialize NVML: Driver/library version mismatch

Can anyone provide any insight as what could potentially be happening?
Are there any other more insightful logs I could pull up?

It seems nvidia driver had a corrupted update, what was your nvOC version.

As soon as rig started, close gnome-terminal and run :

Code:

sudo dpkg --configure -a

It should fix the corrupted update.
I recommend update to latest

I cant do that because the rig refuses to boot. I took the harddrive out and moved it to another machine to get those logs.

Im using version 19. The last stable update from the author.

[latechguy]

Can someone educate me as to what makes a motherboard "supported" or not?

I don't know squat about mining generally, but in building basic linux servers over the years I've never really been concerned with motherboard support.

I ask because I was considering playing around with nvOC on an old HP linux server with a single 1050ti I had laying around.  Wanted to get familiar with nvOC working with nvidia cards before I buy and build out a real rig.

However, I saw something about supported motherboards and figured I would ask before I wasted my time.

Thanks!

[ivdezine]

Sorry in advance if this was already asked and answered, but can this OS be used to mine bitcoin gold using 1080ti GPUs?

Yes it can. I do off and on do a run on BTG. I use Claymores miner and it works great. Good steady hash rate.  I do use 1060's tho but the '80s should be faster is all.

thay

Thanks so much for the response and confirming, Thaelin!

[dbolivar]

Just writing to say THANKS to everyone involved in this project, from the initial versions to the new 2.0-Community Release. I've been using nvOC for some months, and recently updated the base image to the 2.0-CR version. Everything super smooth and stable, with cool new features. The update scripts for the base image and new miners work like a charm. Keep up the good work, guys! It's an excellent case of how the open-source community can achieve great things.

[MassDestruction]

Hey everyone. I'm having an issue. My miner tends to reset on me, and when it does it seems like the operating system gets into a bad state, and can't be restarted again unless I rewrite the storage device with a fresh copy of nvoc. Here is information on my miner

PSU: Corsair AX860
Motherboard: Biostar TB250 Pro (has 12x PCIE slots)
Video Cards: 5 GTX 1060s and 1 1050 TI
Storage: SanDisk 120 GB connected via StarTech USB 3.0 to 2.5 Sata HDD/SSD cable
Processor: i5 6600k with Hyper 212 cooler(to be replaced soon with G3900 with stock cooler)
Ram: 16GB
Current Power Draw: 690 watts

Here are shots of the miner in case anyone is interested.
https://imgur.com/Pkes5Su
https://imgur.com/S49LlKl


I'm currently mining ETH, and I'm using the default clocks in the oneBash. Which is 100 for the core, and 100 for the memory.

During my debugging process while trying to figure out what was going on I set the bios to use integrated graphics, and plugged my monitor into the integrated graphics port on the mother board. I was able to pick this up on screen.
https://imgur.com/61g5FYP

Usually when i do this the miner will boot loop(A normal quirk for the TB250). So I'm guessing this is related to the problem I have, but I'm not sure how to move on from here

That's really strange ...

You should not connect your monitor to IGD, disable IGD in bios and connect monitor to the graphic card on main PCIe 16x.
Since you have 5x1060 and 1x1050, connect your 1050 to that slot so it  become GPU0 for easier settings in individual power and oc.
When using mix cards make sure to set the correct values for each card with individual power limit, and oc

You can find your GPUs  by running this if you are on v0019-2.0

Code:

./nvOC gpumap

Why connect SSD with USB converter cable? connect it directly with sata cable and see how it goes.
How you connect riser powers?  golden rule is 2 risers for each cable.

Yes I normally connect it to the 1050 TI in the PCIe 16x slot. I only started using IGD to try and debug things. It's currently set back to normal, and I'm still having the same issues.

I use the StarTech USB 3.0 to 2.5 Sata HDD/SSD cable because the author of nvoc recommended it.

For the risers I'm using no more than 2 sata for one

Also I assume the oc values represent a MHZ offset right?

Ok I think I'm starting to narrow down my problem? So whenever this message pops up in claymore(link to an image of the message below) the rig hard crashes and reboots. Then it goes into that unrecoverable state(where the drivers get messed up?). There is a link to the image below.
Any insight would be greatly appreciated!

https://imgur.com/JLh1QiT

It doesn't look like an error. I'm not entirely sure what this pool switches message means yet though, and why it's displaying my video cards again.
Also note that I removed the majorly of my cards, and left only two in there. 1 1050 Ti and 1 1060

[FNT]

Hi!

@papampi: auto OC: ok, understand. I don't want to push the cards like hell, but I think about an option
to optimize cards hashrate/lifetime/Temp/power usage, etc...

anybody mining GBX? I tried KTminer (8.17) and SP, TP, etc... but the only one which works is "KXccminer-skunk"?

Thanks and bye, FNT

[papampi]

ccminer-2.2.4- tpruvot

Download:

Code:

cd /home/m1/Downloads/miners
wget https://www.dropbox.com/s/c3uitjc2zcy0yw5/ccminer-2.2.4-tpruvot.tar.gz
tar -xzvf ccminer-2.2.4-tpruvot.tar.gz

Check on your rig :

Code:

/home/m1/Downloads/miners/ccminer-2.2.4-tpruvot/ccminer -a ALGO -o SERVER_ADDRESS:PORT   -u USERNAME.WORKERNAME -p x

Copy to miner folder:

Code:

cp /home/m1/Downloads/miners/ccminer-2.2.4-tpruvot/* /home/m1/TPccminer/

[NameTaken]

Code:

wget -O- https://www.dropbox.com/s/c3uitjc2zcy0yw5/ccminer-2.2.4-tpruvot.tar.gz | tar -xzC /home/m1/TPccminer/ --strip 1  ccminer-2.2.4-tpruvot/ccminer

[carvincam]

Respectfully request to add BTG! (Bitcoin gold)

Thank you for this amazing Ubuntu build!

[papampi]

Code:

wget -O- https://www.dropbox.com/s/c3uitjc2zcy0yw5/ccminer-2.2.4-tpruvot.tar.gz | tar -xzC /home/m1/TPccminer/ --strip 1  ccminer-2.2.4-tpruvot/ccminer

lol ... I fix the update script with your suggestion.
Didn't want to replace the users current miner before they test it


Now that you bring it up, any way not to double show echo lines with "set -x"