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

[Rig4p]

See this post: https://bitcointalk.org/index.php?topic=1854250.msg29350878#msg29350878

Hope it helps.

[taltom]

Been a while since I've popped my head in here...been running rather uneventfully with nvOC and my auto-switcher, watching coin pile up.

The auto-switcher writes its activity to a file, algo-log.  Yesterday, I got around to knocking together a simple tool that looks at the algo-log file and tells you what you've been mining recently:

Code:

#!/usr/bin/env python3

from datetime import datetime, timedelta
import sys
import operator

end=datetime.now().replace(microsecond=0)
try:
  start=end-timedelta(days=int(sys.argv[1]))
except:
  start=end-timedelta(days=1)

l={}

with (open("algo-log", "r")) as f:
  for i in f:
    curr=datetime.strptime(i[0:19], "%Y-%m-%d %H:%M:%S")
    curr_algo=i[28:].split(" ")[0]
    if (curr>start):
      try:
        l[curr_algo]+=curr-last
      except:
        l[curr_algo]=curr-last
    last=curr
    last_algo=curr_algo
try:
  l[last_algo]+=end-last
except:
  l[last_algo]=end-last

for i in sorted(l.items(), key=operator.itemgetter(1), reverse=True):
  print(i[0]+" {0:0.3f}".format(i[1].days*24+i[1].seconds/3600))

By default, it counts up which coins have been mined over the past day and prints a sorted list.  Here's what I've been mining:

Code:

feathercoin 6.901
monacoin 5.013
digibyte-skein 4.043
vertcoin 4.014
zencash 4.010
bitcoin-gold 1.508

Follow it with a number to get that many days' worth of analysis.  Here's my past week (./top-algos.py 7):

Code:

feathercoin 48.667
monacoin 22.975
vertcoin 17.542
zencash 16.712
digibyte-skein 16.211
groestlcoin 12.701
zclassic 8.029
bitcoin-gold 7.022
ethereum 4.514
monero 4.020
zcash 3.511
maxcoin 3.511
ethereum-classic 3.009

That looks quite handy to have.  I hate to ask, but how do I go about running that?  I'm not familiar with python at all.

Thanks,
Terry

[papampi]

Been a while since I've popped my head in here...been running rather uneventfully with nvOC and my auto-switcher, watching coin pile up.

The auto-switcher writes its activity to a file, algo-log.  Yesterday, I got around to knocking together a simple tool that looks at the algo-log file and tells you what you've been mining recently:

Code:

#!/usr/bin/env python3

from datetime import datetime, timedelta
import sys
import operator

end=datetime.now().replace(microsecond=0)
try:
  start=end-timedelta(days=int(sys.argv[1]))
except:
  start=end-timedelta(days=1)

l={}

with (open("algo-log", "r")) as f:
  for i in f:
    curr=datetime.strptime(i[0:19], "%Y-%m-%d %H:%M:%S")
    curr_algo=i[28:].split(" ")[0]
    if (curr>start):
      try:
        l[curr_algo]+=curr-last
      except:
        l[curr_algo]=curr-last
    last=curr
    last_algo=curr_algo
try:
  l[last_algo]+=end-last
except:
  l[last_algo]=end-last

for i in sorted(l.items(), key=operator.itemgetter(1), reverse=True):
  print(i[0]+" {0:0.3f}".format(i[1].days*24+i[1].seconds/3600))

By default, it counts up which coins have been mined over the past day and prints a sorted list.  Here's what I've been mining:

Code:

feathercoin 6.901
monacoin 5.013
digibyte-skein 4.043
vertcoin 4.014
zencash 4.010
bitcoin-gold 1.508

Follow it with a number to get that many days' worth of analysis.  Here's my past week (./top-algos.py 7):

Code:

feathercoin 48.667
monacoin 22.975
vertcoin 17.542
zencash 16.712
digibyte-skein 16.211
groestlcoin 12.701
zclassic 8.029
bitcoin-gold 7.022
ethereum 4.514
monero 4.020
zcash 3.511
maxcoin 3.511
ethereum-classic 3.009

That looks quite handy to have.  I hate to ask, but how do I go about running that?  I'm not familiar with python at all.

Thanks,
Terry

Save it to a file (top-algos.py) then chmod it

Code:

chmod a+x top-algos.py

Then as salfter said run it followed by day numbers:

Code:

./top-algos.py 7

or dont chmod and run :

Code:

python3 top-algos.py 7

[infowire]

So if i let my miner run 24/7 my disks get full.
I thought it was the logs but i made a cron job to delete them.

Nothing unusual on attached pictures but your 75GB partition is fully used even though it doesn't show when you use df.

The only explanation is that your rig was "hacked" and rootkit installed to hide folders and files (hence the 75GB usage) and probably used as torrent sandbox or streaming box
I wrote hacked in quotes because no real hacking is required when you don't change the password and leave the default one which is publicly available.

What I suggest is to reimage your SSD, unplug the network cable, boot up from freshly imaged SSD, as soon as gnome terminal starts close it to prevent nvoc scripts startup and watchdog restarting. Then press F12 to start guake and change password with:

Code:

passwd

then change root password as well:

Code:

sudo su
passwd

Plug your Ethernet cable and reboot.
Make note of your m1 and root passwords as you might need them sometime.
If you have multiple rigs do this for all of them as the "hacker" has your IP and will try to regain access again.

But why it deletes files on reboot ? If it was 'hacked'

[leenoox]

So if i let my miner run 24/7 my disks get full.
I thought it was the logs but i made a cron job to delete them.

Nothing unusual on attached pictures but your 75GB partition is fully used even though it doesn't show when you use df.

The only explanation is that your rig was "hacked" and rootkit installed to hide folders and files (hence the 75GB usage) and probably used as torrent sandbox or streaming box
I wrote hacked in quotes because no real hacking is required when you don't change the password and leave the default one which is publicly available.

What I suggest is to reimage your SSD, unplug the network cable, boot up from freshly imaged SSD, as soon as gnome terminal starts close it to prevent nvoc scripts startup and watchdog restarting. Then press F12 to start guake and change password with:

Code:

passwd

then change root password as well:

Code:

sudo su
passwd

Plug your Ethernet cable and reboot.
Make note of your m1 and root passwords as you might need them sometime.
If you have multiple rigs do this for all of them as the "hacker" has your IP and will try to regain access again.

But why it deletes files on reboot ? If it was 'hacked'

You didn't mention anything about deleting files on reboot. Anyway, I am not sure what's going on there, some of your screenshots show the root partition as 8.6GB and some as 74GB. Either way, both of them are full, no free space. Even if it's not "hacked" it seems something is corrupted there. I can understand 8.6GB partition (which is kinda small) being full but on other screenshots it shows 74GB being full which is unlikely at normal operation. Reimaging should help.

[WaveFront]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

[papampi]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

Flash with latest nvOC v19-2.0 Shrinked Image


Info:

Code:

Image size: 11 GB
Download ZIP size: 3.8 GB
Auto Expand to full size on first boot
Fixed copy 1bash from fat partition to SSD/HDD on first boot
SHA256: ACCA9787169E6E722ED74FB02DBE1C44E7760CEB537CB79D4C61C7FFF8AD3057

Or :

Code:

cd /home/m1/Downloads
wget -N https://raw.githubusercontent.com/papampi/nvOC_by_fullzero_Community_Release/19-2.1/expand_rootfs.sh
sudo /home/m1/Downloads/expand_rootfs.sh

[WaveFront]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

Flash with latest nvOC v19-2.0 Shrinked Image


Info:

Code:

Image size: 11 GB
Download ZIP size: 3.8 GB
Auto Expand to full size on first boot
Fixed copy 1bash from fat partition to SSD/HDD on first boot
SHA256: ACCA9787169E6E722ED74FB02DBE1C44E7760CEB537CB79D4C61C7FFF8AD3057

Or :

Code:

cd /home/m1/Downloads
wget -N https://raw.githubusercontent.com/papampi/nvOC_by_fullzero_Community_Release/19-2.1/expand_rootfs.sh
sudo /home/m1/Downloads/expand_rootfs.sh

Thanks Papampi. I will do this.

[WaveFront]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

Flash with latest nvOC v19-2.0 Shrinked Image


Info:

Code:

Image size: 11 GB
Download ZIP size: 3.8 GB
Auto Expand to full size on first boot
Fixed copy 1bash from fat partition to SSD/HDD on first boot
SHA256: ACCA9787169E6E722ED74FB02DBE1C44E7760CEB537CB79D4C61C7FFF8AD3057

Or :

Code:

cd /home/m1/Downloads
wget -N https://raw.githubusercontent.com/papampi/nvOC_by_fullzero_Community_Release/19-2.1/expand_rootfs.sh
sudo /home/m1/Downloads/expand_rootfs.sh

Hello,
I flash the image, the rig boots into Ubuntu and everything goes fine until the next reboot when Ubuntu cannot load and gets into rescue mode:

Code:

error: unknown filesystem
Entering rescue mode
grub rescue >

Checking the volume, it show shows that somehow it didn't expand the filesystem

[papampi]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

Flash with latest nvOC v19-2.0 Shrinked Image


Info:

Code:

Image size: 11 GB
Download ZIP size: 3.8 GB
Auto Expand to full size on first boot
Fixed copy 1bash from fat partition to SSD/HDD on first boot
SHA256: ACCA9787169E6E722ED74FB02DBE1C44E7760CEB537CB79D4C61C7FFF8AD3057

Or :

Code:

cd /home/m1/Downloads
wget -N https://raw.githubusercontent.com/papampi/nvOC_by_fullzero_Community_Release/19-2.1/expand_rootfs.sh
sudo /home/m1/Downloads/expand_rootfs.sh

Hello,
I flash the image, the rig boots into Ubuntu and everything goes fine until the next reboot when Ubuntu cannot load and gets into rescue mode:

Code:

error: unknown filesystem
Entering rescue mode
grub rescue >

Checking the volume, it show shows that somehow it didn't expand the filesystem

I imaged many SSD and USB's with it and none get to problems.
Did it expanded the partition on first boot and rebooted after 5 seconds?

[WaveFront]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

Flash with latest nvOC v19-2.0 Shrinked Image


Info:

Code:

Image size: 11 GB
Download ZIP size: 3.8 GB
Auto Expand to full size on first boot
Fixed copy 1bash from fat partition to SSD/HDD on first boot
SHA256: ACCA9787169E6E722ED74FB02DBE1C44E7760CEB537CB79D4C61C7FFF8AD3057

Or :

Code:

cd /home/m1/Downloads
wget -N https://raw.githubusercontent.com/papampi/nvOC_by_fullzero_Community_Release/19-2.1/expand_rootfs.sh
sudo /home/m1/Downloads/expand_rootfs.sh

Hello,
I flash the image, the rig boots into Ubuntu and everything goes fine until the next reboot when Ubuntu cannot load and gets into rescue mode:

Code:

error: unknown filesystem
Entering rescue mode
grub rescue >

Checking the volume, it show shows that somehow it didn't expand the filesystem

I imaged many SSD and USB's with it and none get to problems.
Did it expanded the partition on first boot and rebooted after 5 seconds?

Hi Papampi,
At first reboot it goes straight into rescue mode.

If I image the standard nvOC image (not the Shrinkied Image) and if I follow the "SSD-USB resize guide for nvOC" document, I run into a problem when I execute the command

Code:

gksudo gedit /etc/fstab

as I get the error:

Code:

m1@m1-desktop:~$ gksudo gedit /etc/fstab

(process:20027): Gtk-WARNING **: Locale not supported by C library.
	Using the fallback 'C' locale.

(gksudo:20027): Gtk-WARNING **: cannot open display: 

The system works pretty fine without extending the partition, so I guess I can leave everything like it is

[infowire]

sudo: ./expand_rootfs.sh: command not found
i downloaded and i am in the directory.

[hallzero]

Hi guys,

as a newbie to linux i have to thank all developers for their work. Thank you guys and please don´t stop, if you did, you would leave people like me hanging in the cold.

A problem i encountered with the new community release version 2.0 is the following.

If i try to connect any of my rigs (doesn´t matter whether nvidia or amd cards) to any etn pool (doesn´t matter as well which pool it is) nvOC uses KTccminer as the miner but once i connect the pool always sets the difficulty at 19, exactly at 19 and this repeats no matter what rig i connect to what etn pool. I had all rigs running under 19. 1.4 and everything was good, now i can´t connect to any of the etn pools.

Any suggestions?

[damNmad]

Hi guys,

as a newbie to linux i have to thank all developers for their work. Thank you guys and please don´t stop, if you did, you would leave people like me hanging in the cold.

A problem i encountered with the new community release version 2.0 is the following.

If i try to connect any of my rigs (doesn´t matter whether nvidia or amd cards) to any etn pool (doesn´t matter as well which pool it is) nvOC uses KTccminer as the miner but once i connect the pool always sets the difficulty at 19, exactly at 19 and this repeats no matter what rig i connect to what etn pool. I had all rigs running under 19. 1.4 and everything was good, now i can´t connect to any of the etn pools.

Any suggestions?

It uses 'KTccminer-cryptonight'

Can you paste result of this command (type in guake)

ps aux | grep miner

Paste the result over here.

Can you also paste the ETN coin details you have in 1bash?

[WaveFront]

Hello,
Probably the question has been asked zillions of time, but, is there a way to infer the memory manufacturer of a GPU card from Linux/Ubuntu?
I know I can do this using GPU-Z on Windows, but is it really impossible on Linux?

[papampi]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

Flash with latest nvOC v19-2.0 Shrinked Image


Info:

Code:

Image size: 11 GB
Download ZIP size: 3.8 GB
Auto Expand to full size on first boot
Fixed copy 1bash from fat partition to SSD/HDD on first boot
SHA256: ACCA9787169E6E722ED74FB02DBE1C44E7760CEB537CB79D4C61C7FFF8AD3057

Or :

Code:

cd /home/m1/Downloads
wget -N https://raw.githubusercontent.com/papampi/nvOC_by_fullzero_Community_Release/19-2.1/expand_rootfs.sh
sudo /home/m1/Downloads/expand_rootfs.sh

Hello,
I flash the image, the rig boots into Ubuntu and everything goes fine until the next reboot when Ubuntu cannot load and gets into rescue mode:

Code:

error: unknown filesystem
Entering rescue mode
grub rescue >

Checking the volume, it show shows that somehow it didn't expand the filesystem

I imaged many SSD and USB's with it and none get to problems.
Did it expanded the partition on first boot and rebooted after 5 seconds?

Hi Papampi,
At first reboot it goes straight into rescue mode.

If I image the standard nvOC image (not the Shrinkied Image) and if I follow the "SSD-USB resize guide for nvOC" document, I run into a problem when I execute the command

Code:

gksudo gedit /etc/fstab

as I get the error:

Code:

m1@m1-desktop:~$ gksudo gedit /etc/fstab

(process:20027): Gtk-WARNING **: Locale not supported by C library.
	Using the fallback 'C' locale.

(gksudo:20027): Gtk-WARNING **: cannot open display: 

The system works pretty fine without extending the partition, so I guess I can leave everything like it is

If on first boot it goes to rescue mode then there should be a problem with downloaded image, have you checked it's SHA-256?

You ran "gksudo gedit" from local machine or some remote like TeamViewer or vnc?
What happens if you run the expand_rootfs.sh script on normal image?

[papampi]

Hello,
Probably the question has been asked zillions of time, but, is there a way to infer the memory manufacturer of a GPU card from Linux/Ubuntu?
I know I can do this using GPU-Z on Windows, but is it really impossible on Linux?

Don't think there is way to do so under Linux yet.

[hallzero]

Hi guys,

as a newbie to linux i have to thank all developers for their work. Thank you guys and please don´t stop, if you did, you would leave people like me hanging in the cold.

A problem i encountered with the new community release version 2.0 is the following.

If i try to connect any of my rigs (doesn´t matter whether nvidia or amd cards) to any etn pool (doesn´t matter as well which pool it is) nvOC uses KTccminer as the miner but once i connect the pool always sets the difficulty at 19, exactly at 19 and this repeats no matter what rig i connect to what etn pool. I had all rigs running under 19. 1.4 and everything was good, now i can´t connect to any of the etn pools.

Any suggestions?

It uses 'KTccminer-cryptonight'

Can you paste result of this command (type in guake)

ps aux | grep miner

Paste the result over here.



Can you also paste the ETN coin details you have in 1bash?

Thanks for trying to help this is the result of your command

m1@m1-desktop:~$ ps aux | grep miner
m1       10259  288  0.6 564212 49368 pts/17   Sl+  22:37   2:15 /home/m1/cpuOPT/cpuminer -a cryptonight -o stratum+tcp://pool.etnminers.com:3333 -u etnkEv3h8DxMLSWMBYTmMLfKNgnaQf6k7hGbkEbK6hbeGku69dh8Lm7iihpV4dg3BRXcJFAKfGUaNac isz4bmS6E8HRgEA2cJZ.1600.19_2_ASGT405XE3 -p x -t 3
m1       10553  0.0  0.0  27188  2896 ?        Ss   22:37   0:00 SCREEN -dmSL miner /home/m1/KTccminer-cryptonight/ccminer -o stratum+tcp://pool.etnminers.com:7777 -u etnkEv3h8DxMLSWMBYTmMLfKNgnaQf6k7hGbkEbK6hbeGku69dh8Lm7iihpV4dg3BRXcJFAKfGUaNac isz4bmS6E8HRgEA2cJZ.19_2_ASGT405XE3 -p x
m1       10554 12.5  5.1 64984620 414744 pts/24 Ssl+ 22:37   0:04 /home/m1/KTccminer-cryptonight/ccminer -o stratum+tcp://pool.etnminers.com:7777 -u etnkEv3h8DxMLSWMBYTmMLfKNgnaQf6k7hGbkEbK6hbeGku69dh8Lm7iihpV4dg3BRXcJFAKfGUaNac isz4bmS6E8HRgEA2cJZ.19_2_ASGT405XE3 -p x
m1       10702  0.0  0.0  14224   944 pts/16   R+   22:38   0:00 grep --color=auto miner


I also tried to connect to etn.easyhash.io:3631 with the same result, diff gets set at exactly 19 and stays there. I can however set the diff manually by adding it to my etn address as "...address.20000" This works but then the diff stays at that at all times. Once again everything worked fine in 19_1.4

As for the details of my 1bash settings for the etn coin. I have my etn address in there as shown above as well as the port. I use port 3333 for the cpu threads and 7777 for the gpu´s in the rig i am using right now to get back online. There are 4 1050 ti`s in there.

When i look at at the outcome above it adds the diff 1600 for the cpu thread (which i set manually by adding it to 1bash) and it puts my custom worker name right behind the etn address. Seeing this now i just converted back to HOST instead of CUSTOM worker name but that doesn´t help a lot. In that case the diff gets set to 100, stays there and the miner never starts mining. The cpu works but only because i set it manually to a 1600 diff. KTccminer-cryptonight never starts mining even though it shows accepted shares.

Unfortunately i don´t know enough about Linux to get anywhere close to resolving this.

[bobguy22]

Sorry If this is posted somewhere on this fourm or the website but is there a way to setup profit switching and wifi?

[WaveFront]

Hello :-)
After I flash a hard disk with the nvOC image, is there an easy way to expand the filesystem to occupy the full capacity of the hard drive?

Flash with latest nvOC v19-2.0 Shrinked Image


Info:

Code:

Image size: 11 GB
Download ZIP size: 3.8 GB
Auto Expand to full size on first boot
Fixed copy 1bash from fat partition to SSD/HDD on first boot
SHA256: ACCA9787169E6E722ED74FB02DBE1C44E7760CEB537CB79D4C61C7FFF8AD3057

Or :

Code:

cd /home/m1/Downloads
wget -N https://raw.githubusercontent.com/papampi/nvOC_by_fullzero_Community_Release/19-2.1/expand_rootfs.sh
sudo /home/m1/Downloads/expand_rootfs.sh

Hello,
I flash the image, the rig boots into Ubuntu and everything goes fine until the next reboot when Ubuntu cannot load and gets into rescue mode:

Code:

error: unknown filesystem
Entering rescue mode
grub rescue >

Checking the volume, it show shows that somehow it didn't expand the filesystem

I imaged many SSD and USB's with it and none get to problems.
Did it expanded the partition on first boot and rebooted after 5 seconds?

Hi Papampi,
At first reboot it goes straight into rescue mode.

If I image the standard nvOC image (not the Shrinkied Image) and if I follow the "SSD-USB resize guide for nvOC" document, I run into a problem when I execute the command

Code:

gksudo gedit /etc/fstab

as I get the error:

Code:

m1@m1-desktop:~$ gksudo gedit /etc/fstab

(process:20027): Gtk-WARNING **: Locale not supported by C library.
	Using the fallback 'C' locale.

(gksudo:20027): Gtk-WARNING **: cannot open display: 

The system works pretty fine without extending the partition, so I guess I can leave everything like it is

If on first boot it goes to rescue mode then there should be a problem with downloaded image, have you checked it's SHA-256?

You ran "gksudo gedit" from local machine or some remote like TeamViewer or vnc?
What happens if you run the expand_rootfs.sh script on normal image?

Hi Papampi,
I didn't explain myself correctly.
The image SHA-256 is correct.
Ubuntu boot a first time normally displaying the window with the message "resizing partition at next boot..." (cannot remember the exact message). After this it reboots into rescue mode.
I didn't run "gksudo gedit" directly from the local machine, but through an SSH session