Pages:
Author

Topic: [OS] nvOC easy-to-use Linux Nvidia Mining - page 36. (Read 418253 times)

newbie
Activity: 2
Merit: 0
Hi

I am using main BioStar H81S2, how can I boot nvOC from USB Stick. I tried to using it, but it not show from boot menu.

Disable UEFI, enable CSM or Legacy boot devices. Try different USB ports. If you still having trouble join the Discord chat, it's easier to troubleshoot through live chat (I think the link is in the third post on page 1 of this thread)

Big thank you
full member
Activity: 200
Merit: 101
Hello!

My motherboard does not see ssd with nvoc installed, while with the usb flash I can become on ssd ubuntu,
is there any bash script that would update ubuntu to nvoc fork?

Thank you!

You have to untar (unzip) the image you downloaded.
Depends on the program you use, you might need to unzip it twice (From *.tar.gz to *.tar, then run it again to unzip the *.tar to get *.img). Some programs can directly unzip the tar.gz in one step, some need two steps.
Once you unzipped the downloaded file and got the *.img you can use hddrawcopy to burn the img to your SSD.

Once the image is burned to your SSD, plug it to your PC, remove any other HDD, SSD or USB from your PC, enter BIOS, disable UEFI, enable legacy/CSM and save and restart. It should boot into nvoc.

There is no script to update/turn Ubuntu into nvoc. There's more than just few scripts... nvidia drivers, cuda, miners, removed default packages, installed extra packages, custom OS tweaks... lots of manual tweaks, not easy to do with script.
full member
Activity: 200
Merit: 101
Code:
[2018-03-27 04:21:10] Starting on stratum+tcp://europe.lyra2z-hub.miningpoolhub.com:20581
[2018-03-27 04:21:10] NVML GPU monitoring enabled.
[2018-03-27 04:21:10] 8 miner threads started, using 'lyra2z' algorithm.
[2018-03-27 04:21:13] Stratum difficulty set to 5 (0.01953)
[2018-03-27 04:21:20] GPU #0: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #2: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #1: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #5: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #3: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #4: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #7: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #1: MSI GTX 1080 Ti, 79.74 kH/s
[2018-03-27 04:21:20] GPU #0: MSI GTX 1080 Ti, 79.74 kH/s
[2018-03-27 04:21:20] GPU #6: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #2: MSI GTX 1080, 78.89 kH/s
[2018-03-27 04:21:20] GPU #3: MSI GTX 1080, 76.36 kH/s
[2018-03-27 04:21:20] GPU #5: MSI GTX 1080, 76.31 kH/s
[2018-03-27 04:21:21] GPU #4: MSI GTX 1080, 71.80 kH/s
[2018-03-27 04:21:21] GPU #7: MSI GTX 1080, 71.76 kH/s
[2018-03-27 04:21:21] GPU #6: MSI GTX 1080 Ti, 69.34 kH/s
[2018-03-27 04:21:25] GPU #3: MSI GTX 1080, 2183.67 kH/s
[2018-03-27 04:21:37] GPU #4: MSI GTX 1080, 2198.53 kH/s
[2018-03-27 04:21:37] accepted: 1/1 (diff 0.033), 10.70 MH/s yes!
[2018-03-27 04:21:38] GPU #2: MSI GTX 1080, 2176.62 kH/s
[2018-03-27 04:21:45] GPU #4: MSI GTX 1080, 2195.73 kH/s
[2018-03-27 04:21:45] accepted: 2/2 (diff 0.033), 11.23 MH/s yes!
[2018-03-27 04:21:46] GPU #7: MSI GTX 1080, 2174.82 kH/s
[2018-03-27 04:21:46] accepted: 3/3 (diff 0.033), 11.59 MH/s yes!
[2018-03-27 04:21:56] GPU #0: MSI GTX 1080 Ti, 2923.20 kH/s
[2018-03-27 04:21:56] GPU #0: 2019 MHz 17.20 kH/W 113W 51C FAN 30%
[2018-03-27 04:21:57] accepted: 4/4 (diff 0.033), 12.30 MH/s yes!
[2018-03-27 04:21:58] GPU #6: MSI GTX 1080 Ti, 4806.30 kH/s
[2018-03-27 04:21:58] GPU #6: unspecified launch failure
[2018-03-27 04:21:58] GPU #1: MSI GTX 1080 Ti, 4754.69 kH/s
[2018-03-27 04:21:58] GPU #1: unspecified launch failure
[2018-03-27 04:21:58] GPU #2: MSI GTX 1080, 6329.96 kH/s
[2018-03-27 04:21:58] GPU #2: unspecified launch failure
[2018-03-27 04:21:58] GPU #6: unspecified launch failure
[2018-03-27 04:21:58] GPU #4: MSI GTX 1080, 10.13 MH/s
[2018-03-27 04:21:58] GPU #4: unspecified launch failure
[2018-03-27 04:21:58] GPU #1: unspecified launch failure
[2018-03-27 04:21:58] GPU #5: MSI GTX 1080, 3618.08 kH/s
[2018-03-27 04:21:58] GPU #5: unspecified launch failure
[2018-03-27 04:21:58] GPU #7: MSI GTX 1080, 11.08 MH/s
[2018-03-27 04:21:58] GPU #7: unspecified launch failure
[2018-03-27 04:21:58] GPU #3: MSI GTX 1080, 3992.41 kH/s
[2018-03-27 04:21:58] GPU #3: unspecified launch failure
[2018-03-27 04:21:58] GPU #0: unspecified launch failure
[2018-03-27 04:21:58] GPU #2: unspecified launch failure
[2018-03-27 04:21:58] GPU #6: unspecified launch failure

when i start mining ZCOIN its starts but after gets just \\unspecified launch failure\\, what do i need to do?

Try lowering OC settings, see if that helps.
newbie
Activity: 7
Merit: 0
Does anyone know how I can dual mine Ethereum with verge(xvg). I am currently running nvoc 19.2
newbie
Activity: 9
Merit: 0
Hello!

My motherboard does not see ssd with nvoc installed, while with the usb flash I can become on ssd ubuntu,
is there any bash script that would update ubuntu to nvoc fork?

Thank you!
newbie
Activity: 6
Merit: 0
[2018-03-27 04:21:10] Starting on stratum+tcp://europe.lyra2z-hub.miningpoolhub.com:20581
[2018-03-27 04:21:10] NVML GPU monitoring enabled.
[2018-03-27 04:21:10] 8 miner threads started, using 'lyra2z' algorithm.
[2018-03-27 04:21:13] Stratum difficulty set to 5 (0.01953)
[2018-03-27 04:21:20] GPU #0: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #2: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #1: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #5: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #3: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #4: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #7: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #1: MSI GTX 1080 Ti, 79.74 kH/s
[2018-03-27 04:21:20] GPU #0: MSI GTX 1080 Ti, 79.74 kH/s
[2018-03-27 04:21:20] GPU #6: Intensity set to 17, 131072 cuda threads
[2018-03-27 04:21:20] GPU #2: MSI GTX 1080, 78.89 kH/s
[2018-03-27 04:21:20] GPU #3: MSI GTX 1080, 76.36 kH/s
[2018-03-27 04:21:20] GPU #5: MSI GTX 1080, 76.31 kH/s
[2018-03-27 04:21:21] GPU #4: MSI GTX 1080, 71.80 kH/s
[2018-03-27 04:21:21] GPU #7: MSI GTX 1080, 71.76 kH/s
[2018-03-27 04:21:21] GPU #6: MSI GTX 1080 Ti, 69.34 kH/s
[2018-03-27 04:21:25] GPU #3: MSI GTX 1080, 2183.67 kH/s
[2018-03-27 04:21:37] GPU #4: MSI GTX 1080, 2198.53 kH/s
[2018-03-27 04:21:37] accepted: 1/1 (diff 0.033), 10.70 MH/s yes!
[2018-03-27 04:21:38] GPU #2: MSI GTX 1080, 2176.62 kH/s
[2018-03-27 04:21:45] GPU #4: MSI GTX 1080, 2195.73 kH/s
[2018-03-27 04:21:45] accepted: 2/2 (diff 0.033), 11.23 MH/s yes!
[2018-03-27 04:21:46] GPU #7: MSI GTX 1080, 2174.82 kH/s
[2018-03-27 04:21:46] accepted: 3/3 (diff 0.033), 11.59 MH/s yes!
[2018-03-27 04:21:56] GPU #0: MSI GTX 1080 Ti, 2923.20 kH/s
[2018-03-27 04:21:56] GPU #0: 2019 MHz 17.20 kH/W 113W 51C FAN 30%
[2018-03-27 04:21:57] accepted: 4/4 (diff 0.033), 12.30 MH/s yes!
[2018-03-27 04:21:58] GPU #6: MSI GTX 1080 Ti, 4806.30 kH/s
[2018-03-27 04:21:58] GPU #6: unspecified launch failure
[2018-03-27 04:21:58] GPU #1: MSI GTX 1080 Ti, 4754.69 kH/s
[2018-03-27 04:21:58] GPU #1: unspecified launch failure
[2018-03-27 04:21:58] GPU #2: MSI GTX 1080, 6329.96 kH/s
[2018-03-27 04:21:58] GPU #2: unspecified launch failure
[2018-03-27 04:21:58] GPU #6: unspecified launch failure
[2018-03-27 04:21:58] GPU #4: MSI GTX 1080, 10.13 MH/s
[2018-03-27 04:21:58] GPU #4: unspecified launch failure
[2018-03-27 04:21:58] GPU #1: unspecified launch failure
[2018-03-27 04:21:58] GPU #5: MSI GTX 1080, 3618.08 kH/s
[2018-03-27 04:21:58] GPU #5: unspecified launch failure
[2018-03-27 04:21:58] GPU #7: MSI GTX 1080, 11.08 MH/s
[2018-03-27 04:21:58] GPU #7: unspecified launch failure
[2018-03-27 04:21:58] GPU #3: MSI GTX 1080, 3992.41 kH/s
[2018-03-27 04:21:58] GPU #3: unspecified launch failure
[2018-03-27 04:21:58] GPU #0: unspecified launch failure
[2018-03-27 04:21:58] GPU #2: unspecified launch failure
[2018-03-27 04:21:58] GPU #6: unspecified launch failure



when i start mining ZCOIN its starts but after gets just \\unspecified launch failure\\, what do i need to do?
member
Activity: 224
Merit: 13
Compiled it with cuda9 as well, there is no difference whatsoever on some hundred of 1070, 1070Ti and 1080Ti

YMMV

Good deal. Any chance you have both cuda8 and cuda9 on the same machine at the same time? I suspect it is 2 different machines or one machine that you upgraded. Can you share the process for upgrading to cuda9?

-Stubo
newbie
Activity: 16
Merit: 0
When you have time, could you check/review this new miner at:

https://github.com/brian112358/nevermore-miner/releases

Apparently is 20% faster and more stable than msft

Hey Lee:

I downloaded it for some quick and dirty testing. This miner requires cuda 9 and the Community Release is at 8:
Code:
m1@Testy:~/Downloads/nevermore-linux$ ./ccminer
./ccminer: error while loading shared libraries: libcudart.so.9.0: cannot open shared object file: No such file or directory

Since I have been out of the loop for a while, I don't know where we are with upgrading cuda versions.

Hey Stubo,

Compile it from source, it works a charm with cuda8.

autogen.sh
configure
build.sh


Compiled it with cuda9 as well, there is no difference whatsoever on some hundred of 1070, 1070Ti and 1080Ti

YMMV
member
Activity: 224
Merit: 13
When you have time, could you check/review this new miner at:

https://github.com/brian112358/nevermore-miner/releases

Apparently is 20% faster and more stable than msft

Hey Lee:

I downloaded it for some quick and dirty testing. This miner requires cuda 9 and the Community Release is at 8:
Code:
m1@Testy:~/Downloads/nevermore-linux$ ./ccminer
./ccminer: error while loading shared libraries: libcudart.so.9.0: cannot open shared object file: No such file or directory

Since I have been out of the loop for a while, I don't know where we are with upgrading cuda versions.

Have you tried the top link on that page (not the source tarballs)? It might be precompiled with statically compiled cuda libraries.

As for the next release I think we should go with Cuda 9. Most of the miners support Cuda 9 I believe. And those that don't, seems their devs are living in stone age and probably are not actively developed anymore and don't deserve to be included Wink
Correct me if I'm wrong but there seems to be enough miners for all algos that will play nicely with cuda 9. Any particular (highly used) miner still stuck in stone age?

It compiled up for me and doesn't explode when I exec with no args:
Code:
m1@Testy:~/Downloads/nevermore-miner-0.1-alpha$ ./ccminer
*** nevermore 0.1 for nVidia GPUs by brian112358@github ***
    Built with the nVidia CUDA Toolkit 8.0 64-bits

  Originally based on Christian Buchner and Christian H. project
  Include some kernels from alexis78, djm34, djEzo, tsiv and krnlx.

Dev donation set to 1.0%. Thanks for supporting this project!

[2018-03-26 16:38:13] NVML GPU monitoring enabled.
[2018-03-26 16:38:13] 2 miner threads started, using 'blakecoin' algorithm.
[2018-03-26 16:38:13] HTTP request failed: malformed
[2018-03-26 16:38:13] get_work failed, retry after 30 seconds

Here is what I did:
Code:
# Nevermore v0.1-alpha
cd /home/m1/Downloads
wget -N https://github.com/brian112358/nevermore-miner/archive/v0.1-alpha.tar.gz
tar -zxf v0.1-alpha.tar.gz
cd nevermore-miner-0.1-alpha
./build.sh

I don't use ccminer variants so I will leave the testing to others.

newbie
Activity: 35
Merit: 0
Hello guys.
I was using nvOC_v0019 for a couple of months and then I switched to nvOC v0019-2.0.
I like all the new features of nvOC v0019-2.0 but I have problems with stability.
The nvOC_v0019 lasted around 10 days without freezes but nvOC v0019-2.0 lasts only a day or two and then freezes. I went back to nvOC_v0019 and stability improved.
I use SSH to access the miner.

I have two questions:

1) Can you help me with a command that will copy 1bash from 8MB windows partition to the right place, that it will be used on startup (nvOC_v0019 has a problem that it doesn't copy 1bash on first start with HDD)

2) I suspect that the problem with nvOC v0019-2.0 is in newer drivers. Fullzero also wrote that vBASIC uses an older Nvidia Driver. Can anybody give me some instructions that will uninstall nvidia drivers in nvOC v0019-2.0 and then install older drivers?

I would use the vBASIC, but it really lacks a lot of features ...
Fullzero, can you please tell me which driver version do you use in vBASIC and which version in nvOC_v0019?
Whitch command can I use to get Nvidia driver version?


Sorry to hear that you have problems with 19-2.0. It is generally more stable but it is possible that it might perform worse on certain configurations. If you could provide more details we might be able to troubleshoot and help.

1) I don't remember the structure of v0019. I don't even remember if the small FAT partition was mounted correctly. It's quite an old version and won't be easy to guide you without installing v0019 on a rig just to see the structure. Basicly, you would need to copy 1bash from FAT partition to your /home folder then run it through dos2unix to convert it.

2) Although it is quite easy to upgrade to the latest version of the driver, it is not easy to downgrade since PPA repo keeps the latest driver only. If you can find the .deb package of an older driver you can install it manually with dpkg

You can get the driver version with this:
Code:
nvidia-smi -i 0 -q|grep "Driver Version"|cut -f 2 -d ":"|cut -f 2 -d " "


I think we can mount the small fat partition on v19 SSD installs and copy 1bash with following :

Code:
sudo mkdir -p /media/m1/12D3-A869/
sudo mount /dev/sda1 /media/m1/12D3-A869
cp '/media/m1/12D3-A869/1bash' '/home/m1/1bash'
dos2unix /home/m1/1bash




Thanks papampi. I will try this code next time I go to the rigs.
newbie
Activity: 35
Merit: 0
Hello guys.
I was using nvOC_v0019 for a couple of months and then I switched to nvOC v0019-2.0.
I like all the new features of nvOC v0019-2.0 but I have problems with stability.
The nvOC_v0019 lasted around 10 days without freezes but nvOC v0019-2.0 lasts only a day or two and then freezes. I went back to nvOC_v0019 and stability improved.
I use SSH to access the miner.

I have two questions:

1) Can you help me with a command that will copy 1bash from 8MB windows partition to the right place, that it will be used on startup (nvOC_v0019 has a problem that it doesn't copy 1bash on first start with HDD)

2) I suspect that the problem with nvOC v0019-2.0 is in newer drivers. Fullzero also wrote that vBASIC uses an older Nvidia Driver. Can anybody give me some instructions that will uninstall nvidia drivers in nvOC v0019-2.0 and then install older drivers?

I would use the vBASIC, but it really lacks a lot of features ...
Fullzero, can you please tell me which driver version do you use in vBASIC and which version in nvOC_v0019?
Whitch command can I use to get Nvidia driver version?


Sorry to hear that you have problems with 19-2.0. It is generally more stable but it is possible that it might perform worse on certain configurations. If you could provide more details we might be able to troubleshoot and help.

1) I don't remember the structure of v0019. I don't even remember if the small FAT partition was mounted correctly. It's quite an old version and won't be easy to guide you without installing v0019 on a rig just to see the structure. Basicly, you would need to copy 1bash from FAT partition to your /home folder then run it through dos2unix to convert it.

2) Although it is quite easy to upgrade to the latest version of the driver, it is not easy to downgrade since PPA repo keeps the latest driver only. If you can find the .deb package of an older driver you can install it manually with dpkg

You can get the driver version with this:
Code:
nvidia-smi -i 0 -q|grep "Driver Version"|cut -f 2 -d ":"|cut -f 2 -d " "

Thanks, I got nvOC_v0019 nvidia driver version: 384.59
My configuration is ASRock H110 Pro BTC+ with 13 1080tis (different versions of tis). I will try to change the configuration (move some GPUs to antoher machine) and then try to trubleshoot.

full member
Activity: 200
Merit: 101
When you have time, could you check/review this new miner at:

https://github.com/brian112358/nevermore-miner/releases

Apparently is 20% faster and more stable than msft

Hey Lee:

I downloaded it for some quick and dirty testing. This miner requires cuda 9 and the Community Release is at 8:
Code:
m1@Testy:~/Downloads/nevermore-linux$ ./ccminer
./ccminer: error while loading shared libraries: libcudart.so.9.0: cannot open shared object file: No such file or directory

Since I have been out of the loop for a while, I don't know where we are with upgrading cuda versions.

Have you tried the top link on that page (not the source tarballs)? It might be precompiled with statically compiled cuda libraries.

As for the next release I think we should go with Cuda 9. Most of the miners support Cuda 9 I believe. And those that don't, seems their devs are living in stone age and probably are not actively developed anymore and don't deserve to be included Wink
Correct me if I'm wrong but there seems to be enough miners for all algos that will play nicely with cuda 9. Any particular (highly used) miner still stuck in stone age?
member
Activity: 224
Merit: 13
When you have time, could you check/review this new miner at:

https://github.com/brian112358/nevermore-miner/releases

Apparently is 20% faster and more stable than msft

Hey Lee:

I downloaded it for some quick and dirty testing. This miner requires cuda 9 and the Community Release is at 8:
Code:
m1@Testy:~/Downloads/nevermore-linux$ ./ccminer
./ccminer: error while loading shared libraries: libcudart.so.9.0: cannot open shared object file: No such file or directory

Since I have been out of the loop for a while, I don't know where we are with upgrading cuda versions.
full member
Activity: 200
Merit: 101
XMR/Monero mining using XMR-STAK

The new version of XMR-STAK that supports Monero7 is out:
https://github.com/fireice-uk/xmr-stak/releases

The PoWs They Are a-Changin'

So here is what I did to get it going on the Community Release.
The first step is to DL the source and compile it:

Code:
# Build XMR, from https://github.com/fireice-uk/xmr-stak/blob/master/doc/compile_Linux.md
cd /home/m1/xmr
git clone https://github.com/fireice-uk/xmr-stak.git
#mkdir xmr-stak/build
cd xmr-stak
cmake .
make install

The next thing you want to do is just execute the binary:
Code:
cd bin
./xmr-stak
It will do a one-time prompt for coin, pools, addr.worker, pass and create the necessary configuration files. These new configuration files are NOT compatible with previous versions so this must be done. The next step is to setup nvOC for the memory changes:

Code:
# Configure Memory for hugepages
sudo vi /etc/sysctl.conf
# Add line
vm.nr_hugepages=128
# Set user limits
sudo vi /etc/security/limits.conf
# Add lines
* soft memlock 262144
* hard memlock 262144
# Reboot
sudo shutdown -r now

Note: I did all of the steps above while my rig was mining. The final thing is to modify 0miner to strip out all of the old configuration file generation stuff and get down to just the basics. One could go to the trouble of modifying all of the existing code for this in 0miner to make it compatible with the new version of XMR-STAK but I do not have the time to do it so I just went with a stripped down 0miner to just launch the miner and let the user launch it as above to do the one-time config file generation. That 0miner code is:

Code:
if [ $COIN == "XMR" ]
then
  HCD='/home/m1/xmr/xmr-stak/bin/xmr-stak'
  cd /home/m1/xmr/xmr-stak/bin
  screen -dmSL miner $HCD
fi

Hope this helps.



Great work (as always) Stubo. Thanks for providing detailed guide.

When you have time, could you check/review this new miner at:

https://github.com/brian112358/nevermore-miner/releases

Apparently is 20% faster and more stable than msft
member
Activity: 224
Merit: 13
XMR/Monero mining using XMR-STAK

The new version of XMR-STAK that supports Monero7 is out:
https://github.com/fireice-uk/xmr-stak/releases

The PoWs They Are a-Changin'

So here is what I did to get it going on the Community Release.
The first step is to DL the source and compile it:

Code:
# Build XMR, from https://github.com/fireice-uk/xmr-stak/blob/master/doc/compile_Linux.md
cd /home/m1/xmr
git clone https://github.com/fireice-uk/xmr-stak.git
#mkdir xmr-stak/build
cd xmr-stak
cmake .
make install

The next thing you want to do is just execute the binary:
Code:
cd bin
./xmr-stak
It will do a one-time prompt for coin, pools, addr.worker, pass and create the necessary configuration files. These new configuration files are NOT compatible with previous versions so this must be done. The next step is to setup nvOC for the memory changes:

Code:
# Configure Memory for hugepages
sudo vi /etc/sysctl.conf
# Add line
vm.nr_hugepages=128
# Set user limits
sudo vi /etc/security/limits.conf
# Add lines
* soft memlock 262144
* hard memlock 262144
# Reboot
sudo shutdown -r now

Note: I did all of the steps above while my rig was mining. The final thing is to modify 0miner to strip out all of the old configuration file generation stuff and get down to just the basics. One could go to the trouble of modifying all of the existing code for this in 0miner to make it compatible with the new version of XMR-STAK but I do not have the time to do it so I just went with a stripped down 0miner to just launch the miner and let the user launch it as above to do the one-time config file generation. That 0miner code is:

Code:
if [ $COIN == "XMR" ]
then
  HCD='/home/m1/xmr/xmr-stak/bin/xmr-stak'
  cd /home/m1/xmr/xmr-stak/bin
  screen -dmSL miner $HCD
fi

Hope this helps.

[Edit]
If you plan to clone the installation on one rig to others, you should probably blow away the configuration files and run the miner from the command line on each rig. This step generates configuration files that are specific to each GPU installed. So this will be necessary unless you have multiple rigs that are absolutely identical.

newbie
Activity: 12
Merit: 0
Hi guys,

still being fairly new to Linux i encountered the following problem:

Recently i built a rig with a Ryzen 7 1700x and AMD GPU´s and then installed nvOC 19-2.0 Community.

Did not work at all, always got stuck in a login loop and the miner1 password didn´t work, the rig stayed in the login loop.

So obviously i started searching the web and found several threads with exactly the same problem, not for nvOC but Ubuntu 16.04. And there were some solutions which helped others like reassign the read write permissions, renewing .Xauthority and other solutions but none of these worked on my rig, it stayed in the login loop no matter what i did. I used a freshly imaged nvOC 19-2.0 with no changes whatsoever.
I could ctrl+alt+f1 into a tty and login but i could not get nvOC to start and run a miner.

So i figured i just try a different combination and changed the motherboard and now it gets weired. After changing the mobo the same thing happened i was stuck in the login loop so i started to change everything. I tried several motherboard´s, switched memory sticks, used Nvidia GPU´s as well as AMD GPU´s, put nvOC on an USB stick as well as on a harddisk but nothing worked for me in combination with a Ryzen processor. I also tried a Ryzen 3 and a Ryzen 5. Using a windows setup all these combinations worked just fine but not with nvOC.

The second i put a Ryzen processor into any one rig and install nvOC i always get stuck in a login loop.

So is there anybody out there successfully mining with a Ryzen processor and nvOC?
Was anybody stuck in the login loop on nvOC and successfully resolved this matter and how?

Sorry guy´s i am lost here, need some help i can´t use nvOC with a Ryzen processor.


You are using AMD GPU on nvOC?
You should use rxOC for AMD and cant use mix of Nvidia and AMD on nvOC and rxOC

Let me clear it up. No, i was not mixing AMD and Nvidia gpu´s in one rig, instead i built several rigs either using all AMD or all Nvidia.
In fact, after figuring out that it has something to do with the Ryzen cpu i built a rig with a Ryzen 7 1700x and a single AMD gpu then installed nvOC 19-2.0 and got stuck in the login loop. I rebuilt the rig, this time using a single nvidia card alongside the Ryzen 7 1700x and again was stuck in the login loop. To further explain, the Ryzen 7 doesn´t have a builtin gpu, so in order to get anything on the screen you have to use at least a single external gpu.

That is why i was asking if somebody actually is mining succesfully using nvOC in combination with a Ryzen cpu? I can´t get passed the login loop using any Ryzen cpu and nvOC. And since i don´t know enough about Linux i can´t resolve the login loop by myself.

Don't bother with Ryzen CPU on the current nvoc which is based on Ubuntu 16.04 which runs on linux kernel 4.4. That kernel doesn't support Ryzen.
You could upgrade kernel on ubuntu 16.04 to 4.10 which gives partial support for Ryzen but not stable operation (you will have to Google how to do it) but you would still get crashes. You would need at least kernel 4.11 for better Ryzen support but Ubuntu doesn't provide that kernel for 16.04. You could compile latest 4.15 kernel yourself but that requires more than few clicks or simple guidance on this forum thread... that is way out of scope...

Your best bet is to wait for the next nvoc release which will be based on the upcoming Ubuntu 18.04 with kernel 4.15 (ETA May 2018) and use Windows with your Ryzen CPU's in the meantime.

BTW, you stated that you were also trying AMD Gpu's on nvoc. That will absolutely not work with nvoc. There are no AMD drivers in nvoc and all scripts are specifically made for nvidia cards. If you want you can try rxoc that is specifically made for AMD cards but haven't been maintained for a while... I am not sure if fullzero will continue maintaining that OS or not.

Ah, very interesting, that explains a lot. Before i was using a Ryzen cpu i would not get any AMD rig to run with nvOC, thought i was the problem.......didn´t know that it was built specifically for Nvidia cards, now i know. Also i am not knowledgeable enough to compile a linux kernel by myself so i will stick with windows for the moment.
On the other hand, i like nvOC a lot so i hope that in the future i can start using the Ryzen cpu´s to mine with nvOC.

Thank you leenoox, CryptAtomeTrader44 and papampi for clarifying things.
full member
Activity: 686
Merit: 140
Linux FOREVER! Resistance is futile!!!
Hello guys.
I was using nvOC_v0019 for a couple of months and then I switched to nvOC v0019-2.0.
I like all the new features of nvOC v0019-2.0 but I have problems with stability.
The nvOC_v0019 lasted around 10 days without freezes but nvOC v0019-2.0 lasts only a day or two and then freezes. I went back to nvOC_v0019 and stability improved.
I use SSH to access the miner.

I have two questions:

1) Can you help me with a command that will copy 1bash from 8MB windows partition to the right place, that it will be used on startup (nvOC_v0019 has a problem that it doesn't copy 1bash on first start with HDD)

2) I suspect that the problem with nvOC v0019-2.0 is in newer drivers. Fullzero also wrote that vBASIC uses an older Nvidia Driver. Can anybody give me some instructions that will uninstall nvidia drivers in nvOC v0019-2.0 and then install older drivers?

I would use the vBASIC, but it really lacks a lot of features ...
Fullzero, can you please tell me which driver version do you use in vBASIC and which version in nvOC_v0019?
Whitch command can I use to get Nvidia driver version?


Sorry to hear that you have problems with 19-2.0. It is generally more stable but it is possible that it might perform worse on certain configurations. If you could provide more details we might be able to troubleshoot and help.

1) I don't remember the structure of v0019. I don't even remember if the small FAT partition was mounted correctly. It's quite an old version and won't be easy to guide you without installing v0019 on a rig just to see the structure. Basicly, you would need to copy 1bash from FAT partition to your /home folder then run it through dos2unix to convert it.

2) Although it is quite easy to upgrade to the latest version of the driver, it is not easy to downgrade since PPA repo keeps the latest driver only. If you can find the .deb package of an older driver you can install it manually with dpkg

You can get the driver version with this:
Code:
nvidia-smi -i 0 -q|grep "Driver Version"|cut -f 2 -d ":"|cut -f 2 -d " "


I think we can mount the small fat partition on v19 SSD installs and copy 1bash with following :

Code:
sudo mkdir -p /media/m1/12D3-A869/
sudo mount /dev/sda1 /media/m1/12D3-A869
cp '/media/m1/12D3-A869/1bash' '/home/m1/1bash'
dos2unix /home/m1/1bash


full member
Activity: 200
Merit: 101
Hello guys.
I was using nvOC_v0019 for a couple of months and then I switched to nvOC v0019-2.0.
I like all the new features of nvOC v0019-2.0 but I have problems with stability.
The nvOC_v0019 lasted around 10 days without freezes but nvOC v0019-2.0 lasts only a day or two and then freezes. I went back to nvOC_v0019 and stability improved.
I use SSH to access the miner.

I have two questions:

1) Can you help me with a command that will copy 1bash from 8MB windows partition to the right place, that it will be used on startup (nvOC_v0019 has a problem that it doesn't copy 1bash on first start with HDD)

2) I suspect that the problem with nvOC v0019-2.0 is in newer drivers. Fullzero also wrote that vBASIC uses an older Nvidia Driver. Can anybody give me some instructions that will uninstall nvidia drivers in nvOC v0019-2.0 and then install older drivers?

I would use the vBASIC, but it really lacks a lot of features ...
Fullzero, can you please tell me which driver version do you use in vBASIC and which version in nvOC_v0019?
Whitch command can I use to get Nvidia driver version?


Sorry to hear that you have problems with 19-2.0. It is generally more stable but it is possible that it might perform worse on certain configurations. If you could provide more details we might be able to troubleshoot and help.

1) I don't remember the structure of v0019. I don't even remember if the small FAT partition was mounted correctly. It's quite an old version and won't be easy to guide you without installing v0019 on a rig just to see the structure. Basicly, you would need to copy 1bash from FAT partition to your /home folder then run it through dos2unix to convert it.

2) Although it is quite easy to upgrade to the latest version of the driver, it is not easy to downgrade since PPA repo keeps the latest driver only. If you can find the .deb package of an older driver you can install it manually with dpkg

You can get the driver version with this:
Code:
nvidia-smi -i 0 -q|grep "Driver Version"|cut -f 2 -d ":"|cut -f 2 -d " "
full member
Activity: 200
Merit: 101
Hi guys,

still being fairly new to Linux i encountered the following problem:

Recently i built a rig with a Ryzen 7 1700x and AMD GPU´s and then installed nvOC 19-2.0 Community.

Did not work at all, always got stuck in a login loop and the miner1 password didn´t work, the rig stayed in the login loop.

So obviously i started searching the web and found several threads with exactly the same problem, not for nvOC but Ubuntu 16.04. And there were some solutions which helped others like reassign the read write permissions, renewing .Xauthority and other solutions but none of these worked on my rig, it stayed in the login loop no matter what i did. I used a freshly imaged nvOC 19-2.0 with no changes whatsoever.
I could ctrl+alt+f1 into a tty and login but i could not get nvOC to start and run a miner.

So i figured i just try a different combination and changed the motherboard and now it gets weired. After changing the mobo the same thing happened i was stuck in the login loop so i started to change everything. I tried several motherboard´s, switched memory sticks, used Nvidia GPU´s as well as AMD GPU´s, put nvOC on an USB stick as well as on a harddisk but nothing worked for me in combination with a Ryzen processor. I also tried a Ryzen 3 and a Ryzen 5. Using a windows setup all these combinations worked just fine but not with nvOC.

The second i put a Ryzen processor into any one rig and install nvOC i always get stuck in a login loop.

So is there anybody out there successfully mining with a Ryzen processor and nvOC?
Was anybody stuck in the login loop on nvOC and successfully resolved this matter and how?

Sorry guy´s i am lost here, need some help i can´t use nvOC with a Ryzen processor.


You are using AMD GPU on nvOC?
You should use rxOC for AMD and cant use mix of Nvidia and AMD on nvOC and rxOC

Let me clear it up. No, i was not mixing AMD and Nvidia gpu´s in one rig, instead i built several rigs either using all AMD or all Nvidia.
In fact, after figuring out that it has something to do with the Ryzen cpu i built a rig with a Ryzen 7 1700x and a single AMD gpu then installed nvOC 19-2.0 and got stuck in the login loop. I rebuilt the rig, this time using a single nvidia card alongside the Ryzen 7 1700x and again was stuck in the login loop. To further explain, the Ryzen 7 doesn´t have a builtin gpu, so in order to get anything on the screen you have to use at least a single external gpu.

That is why i was asking if somebody actually is mining succesfully using nvOC in combination with a Ryzen cpu? I can´t get passed the login loop using any Ryzen cpu and nvOC. And since i don´t know enough about Linux i can´t resolve the login loop by myself.

Don't bother with Ryzen CPU on the current nvoc which is based on Ubuntu 16.04 which runs on linux kernel 4.4. That kernel doesn't support Ryzen.
You could upgrade kernel on ubuntu 16.04 to 4.10 which gives partial support for Ryzen but not stable operation (you will have to Google how to do it) but you would still get crashes. You would need at least kernel 4.11 for better Ryzen support but Ubuntu doesn't provide that kernel for 16.04. You could compile latest 4.15 kernel yourself but that requires more than few clicks or simple guidance on this forum thread... that is way out of scope...

Your best bet is to wait for the next nvoc release which will be based on the upcoming Ubuntu 18.04 with kernel 4.15 (ETA May 2018) and use Windows with your Ryzen CPU's in the meantime.

BTW, you stated that you were also trying AMD Gpu's on nvoc. That will absolutely not work with nvoc. There are no AMD drivers in nvoc and all scripts are specifically made for nvidia cards. If you want you can try rxoc that is specifically made for AMD cards but haven't been maintained for a while... I am not sure if fullzero will continue maintaining that OS or not.
full member
Activity: 340
Merit: 103
It is easier to break an atom than partialities AE
Hi guys,

still being fairly new to Linux i encountered the following problem:

Recently i built a rig with a Ryzen 7 1700x and AMD GPU´s and then installed nvOC 19-2.0 Community.

Did not work at all, always got stuck in a login loop and the miner1 password didn´t work, the rig stayed in the login loop.

So obviously i started searching the web and found several threads with exactly the same problem, not for nvOC but Ubuntu 16.04. And there were some solutions which helped others like reassign the read write permissions, renewing .Xauthority and other solutions but none of these worked on my rig, it stayed in the login loop no matter what i did. I used a freshly imaged nvOC 19-2.0 with no changes whatsoever.
I could ctrl+alt+f1 into a tty and login but i could not get nvOC to start and run a miner.

So i figured i just try a different combination and changed the motherboard and now it gets weired. After changing the mobo the same thing happened i was stuck in the login loop so i started to change everything. I tried several motherboard´s, switched memory sticks, used Nvidia GPU´s as well as AMD GPU´s, put nvOC on an USB stick as well as on a harddisk but nothing worked for me in combination with a Ryzen processor. I also tried a Ryzen 3 and a Ryzen 5. Using a windows setup all these combinations worked just fine but not with nvOC.

The second i put a Ryzen processor into any one rig and install nvOC i always get stuck in a login loop.

So is there anybody out there successfully mining with a Ryzen processor and nvOC?
Was anybody stuck in the login loop on nvOC and successfully resolved this matter and how?

Sorry guy´s i am lost here, need some help i can´t use nvOC with a Ryzen processor.


You are using AMD GPU on nvOC?
You should use rxOC for AMD and cant use mix of Nvidia and AMD on nvOC and rxOC

Let me clear it up. No, i was not mixing AMD and Nvidia gpu´s in one rig, instead i built several rigs either using all AMD or all Nvidia.
In fact, after figuring out that it has something to do with the Ryzen cpu i built a rig with a Ryzen 7 1700x and a single AMD gpu then installed nvOC 19-2.0 and got stuck in the login loop. I rebuilt the rig, this time using a single nvidia card alongside the Ryzen 7 1700x and again was stuck in the login loop. To further explain, the Ryzen 7 doesn´t have a builtin gpu, so in order to get anything on the screen you have to use at least a single external gpu.

That is why i was asking if somebody actually is mining succesfully using nvOC in combination with a Ryzen cpu? I can´t get passed the login loop using any Ryzen cpu and nvOC. And since i don´t know enough about Linux i can´t resolve the login loop by myself.

I think you have this login loop because you have a problem with your graphic card and Xorg. X server try to start and loading itself with nvidia driver and probably not compiled with AMD Ryzen CPU. So Xorg crashes and return to login screen.

Login in console 1 (CTRL+ALT+F1) and try to compile again this nvidia driver with your CPU.
I'm not able to give you commands but i think you can find them on internet.

Perhaps my suppositions are not true, but i could try this if i had your problem like you discribe.

Hope it helps you.

Leenoox probably helps you better than me on this.
Pages:
Jump to: