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

[fullzero]

Also I have noticed a problem when using both PROFIT_SWITCHING and or WATCHDOG together; I will make changes to the next 1bash to fix it.

Would this happen to be the problem where all the executables disappear, you can't log back in after logging out, etc.?  I noticed that earlier this week when I gave v18 a shot...something would happen after it had been running an hour or two to where it wouldn't respond to anything but either SysRq-B at the keyboard or a reset or power cycle.  Also, the GPUs appeared to still be under full load, but NiceHash reported no activity.  I still had v17 on its stick, so dropping back to it was no problem.

There may be some problems using v0018 without the bug fixed versions; The newest 1bash + files on the OP should work as intended with your profit switcher.  Let me know if they don't.

[1714221157]

1714221157

[1714221157]

1714221157

[1714221157]

1714221157

[1714221157]

1714221157

[1714221157]

1714221157

[salfter]

(I first announced this as an edit to an existing post, but it's probably easier to miss that than a new post.)

I've updated the switcher to accept either an integer or an array of integers for the various overclocking settings.  If you use arrays, you can provide custom settings to each card.

https://gitlab.com/salfter/nvoc-nicehash-switcher/commit/7480fc8c04622277fb638b46c5337c183908d855

The default configuration shows examples of both types.

[fullzero]

I'm planning to experiment with Salfter's NiceHash switcher, but by reading the relevant code in 1bash and switch, it seems that the individual OC settings (per card) defined in 1bash won't be used at all - only the ones set per coin, applied to all the cards. Since I have a mix of different cards, each one must take different core and mem OC values. I have an initial idea of the code changes needed, but before doing that, I'd like to check if there's another way to do that already implemented (which I missed), or something already being developed in that sense.

I've not gotten around to it yet, but I have an idea for how to enable per-card, per-algorithm settings.

I've updated the switcher to accept either an integer or an array of integers for the various overclocking settings.  If you use arrays, you can provide custom settings to each card.

https://gitlab.com/salfter/nvoc-nicehash-switcher/commit/7480fc8c04622277fb638b46c5337c183908d855

Can you please apply these changes to the version of switch included in v0018.

[fullzero]

I always check the temperature of GPUs, but nvidia-smi shows too many useless infoes, so I create an alias, it's very helpful.

Code:

alias t='nvidia-smi --query-gpu=index,name,temperature.gpu --format=csv,noheader'

output looks like:

0, GeForce GTX 1050 Ti, 70
1, GeForce GTX 1060 6GB, 72
2, GeForce GTX 1050 Ti, 65
3, P106-100, 79
4, GeForce GTX 1060 6GB, 68
5, P106-100, 77
6, P106-100, 78
7, P106-100, 71
8, P106-100, 73

Very useful; IMO the card number is implicit in the display order, and the temperature has better grouping if moved to the left of the card type:

Code:

alias t='nvidia-smi --query-gpu=temperature.gpu,name --format=csv,noheader'

[fullzero]

Does anyone know if there is a way to set temperature target for card instead of powerlimit? I have couple gtx 1060 cards that won't go below 97W when running in nvOC, which is 80% power limit, same thing is in windows no matter which tool I use, but when I use power target in windows and set it to ie.65C the cards power usage goes down to 80W. Unfortunately I wasn't able to find if there is a way to control temp target in linux. Any help would be greatly appreciated.

If you use v0018 and download the newest 1bash and files / copy them to the home directory replacing the ones that are there / configure the 1bash:

setting:

Code:

Maxximus007_AUTO_TEMPERATURE_CONTROL="YES"

and each individual powerlimit and temperature limit + control settings:

Code:

INDIVIDUAL_POWERLIMIT_0=75

INDIVIDUAL_POWERLIMIT_1=75

INDIVIDUAL_POWERLIMIT_2=75

INDIVIDUAL_POWERLIMIT_3=75

INDIVIDUAL_POWERLIMIT_4=75

INDIVIDUAL_POWERLIMIT_5=75

INDIVIDUAL_POWERLIMIT_6=75

INDIVIDUAL_POWERLIMIT_7=75

INDIVIDUAL_POWERLIMIT_8=75

INDIVIDUAL_POWERLIMIT_9=75

INDIVIDUAL_POWERLIMIT_10=75

INDIVIDUAL_POWERLIMIT_11=75

INDIVIDUAL_POWERLIMIT_12=75

INDIVIDUAL_POWERLIMIT_13=75

# Set the target temperature per card.

# Don't set it too low, if the temperature can't be reached with fan adjustments it will decrease power.

# Recommended ranges: 55 - 70 degrees Celsius

# If actual temp is above the target temp this script will up the fan speed

TARGET_TEMP_0=70

TARGET_TEMP_1=70

TARGET_TEMP_2=70

TARGET_TEMP_3=70

TARGET_TEMP_4=70

TARGET_TEMP_5=70

TARGET_TEMP_6=70

TARGET_TEMP_7=70

TARGET_TEMP_8=70

TARGET_TEMP_9=70

TARGET_TEMP_10=70

TARGET_TEMP_11=70

TARGET_TEMP_12=70

TARGET_TEMP_13=70


# If actual temp is above value set above and fan speed is 100,
# the script will lower the power limit

# SET INDIVIDUAL POWERLIMITS ABOVE


# Adjustment size (FAN in percent, POWER in Watts)
__FAN_ADJUST=5
POWER_ADJUST=5

# Difference in actual temperature allowed before action.
#Works only if current is BELOW target temp
ALLOWED_TEMP_DIFF=3

# Restore original power limit if fan speed is lower than this percentage
RESTORE_POWER_LIMIT=90

# lowest fan speed that will be used
MINIMAL_FAN_SPEED=50

[fullzero]

Hi fullzero, I cannot save modified 1bash on Mac because the usb key is formatted as NTFS, do you have a plan to change from NTFS to FAT32 to support Mac?

I will change back to FAT for the next version.

[_Parallax_]

The following can be handy for ppl using ssh for their rig. Add to your .bashrc file the following lines:

Code:

#Alias for the screens
alias miner='screen -x miner'
alias temp='screen -x temp'
alias wdog='screen -x wdog'

After the command 'exec bash' or logging out and in again you can type miner / temp / wdog to go to the respective screens directly.

It is better to add in .bash_aliases

There are lots of ways to do things in linux; but yes I agree its putting a square in the rectangle.  Is there any functional difference _Parallax_?

Agree, there is no difference.
Developers are advised to make user settings in a separate file.

[fullzero]

Does anyone have a problem with SRR?

Not sure what I'm doing wrong but my rig keeps reseting every 2 minutes.

I see SRR option was added to the bash script, does it replace the script that I'm suppose to download from the Simple mining website?

that code is implemented in v0018:

in 1bash set:

Code:

SRR="YES"

 

to your serial:
          

Code:

SRR_SERIAL="000055"

to the slot connected to:

Code:

__SRR_SLOT="1"

I don't have an SRR so I can't verify this works as intended.  

[gs777]

First off all THANK YOU Fullzero for nvOC v0018!
Simply love it!

I have a question - WhatToMine shows great rev for DGB-Groestl (DGB), but I can not mine DGB with Myriad-Groestl algo
It always trying to mine with Skein algo.
If I'm not mistaken nvOC has this algo for Diamond(DMD)
so, how DGB-Groestl could be mined?

I will add this to the list.

Thank man I will appreciate this!
Can you also add Lbry for claymore dual mining?

U can dual ETH / LBRY, the coin name of LBRY is LBC, so just use DUAL_ETH_LBC.
Fullzero and the other primary free worker on this put a lot of work and have a lot to do. Read before ask ...

My apology.
Didn't know that LBRY and LBC same coin
thx for the answer!

[fullzero]

Why did my hash rate mining ETH on Claymore 9.7 go from 23.5 on Windows 7 down to 18.9 on nvOC?  I have the power set to 94 and the mc set to 825.  Is the latest driver that bad for 1060 cards? I am pretty sure I have the latest driver on Windows.

The OC curve for Linux is different than Windows.  You need to set a higher OC to get the same results in linux.  Also in Windows power limit is in % TDP; in linux it is in watts. 

Try a mc of 1000, then keep bumping it up until you see a soft crash.

If your 1060s have the standard TDP of 125: to set an equalivent power limit you need to use 0.87 * 125 (or whatever the exact TDP of your model is) = 109 watts

[WoolF911]

fullzero, I just installed NvOc and I have a problem. For testing I installed 1 videocard 1080 Ti with Asus Prime Z270-A and connected monitor locally.
Every my attempt I have a error "Utilization is too low: reviving did not work so restarting system in 10 seconds", also EWBF showed 5-59 Sol/s on ZEC. Also user interface is very slow and lagging. In settings I can select only 1 resolution 2560x1440.
I'm use 018v with updated 1bash.
Maybe you have idea how can I fix it?

[fullzero]

fullzero, I'm just installed NvOc and I have a problem. For testing I installed 1 videocard 1080 Ti with Asus Prime Z270-A and connected monitor locally.
Every my attempt I have a error "Utilization is too low: reviving did not work so restarting system in 10 seconds", also EWBF showed 5-59 Sol/s on ZEC and crashed after some time. Also user interface is very slow and lagging. In settings I can select only 1 resolution 2560x1440.
I'm use 018v with updated 1bash.
Maybe you have idea how can I fix it?

What imaging program did you use?

[fullzero]

The following can be handy for ppl using ssh for their rig. Add to your .bashrc file the following lines:

Code:

#Alias for the screens
alias miner='screen -x miner'
alias temp='screen -x temp'
alias wdog='screen -x wdog'

After the command 'exec bash' or logging out and in again you can type miner / temp / wdog to go to the respective screens directly.

It is better to add in .bash_aliases

There are lots of ways to do things in linux; but yes I agree its putting a square in the rectangle.  Is there any functional difference _Parallax_?

Agree, there is no difference.
Developers are advised to make user settings in a separate file.

Makes sense.

I'm looking at your log script now.  I've been jugging forum responses and a bunch of other stuff locally today so, its taken me a while to catch up on responses.

[WoolF911]

fullzero, I'm just installed NvOc and I have a problem. For testing I installed 1 videocard 1080 Ti with Asus Prime Z270-A and connected monitor locally.
Every my attempt I have a error "Utilization is too low: reviving did not work so restarting system in 10 seconds", also EWBF showed 5-59 Sol/s on ZEC and crashed after some time. Also user interface is very slow and lagging. In settings I can select only 1 resolution 2560x1440.
I'm use 018v with updated 1bash.
Maybe you have idea how can I fix it?

What imaging program did you use?

HDDRawCopy

[fullzero]

There is an idea how to more standartize the logs of programs for further processing them in scripts, notifications and monitoring:

I propose to do:

1) Create file  .screenrc in /home/m1 with content:
#
acladd root
multiuser on
logfile flush 1
log on
logfile /tmp/miner.output.tmp
caption always
caption string "%{= RW} $STY | host: %H | For return to terminal press: Ctrl+a d | %m/%d/%y %c:%s | go go to the mooooooon "
#

2) In the 1bash file, replace the line of the
screen -dmS miner $HCD -S $ETH_POOL -O $ETHADDR:x -U
on
/usr/bin/screen -c /home/m1/.screenrc -l -L -dmS miner $HCD -S $ETH_POOL -O $ETHADDR:x -U

For genoil, I use this script:
telegram.sh
#====================
#!/bin/bash

PATH=$PATH:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin

/usr/bin/tee /tmp/totalhash.log < /dev/null
/usr/bin/tail -n 1 /tmp/miner.log | awk '{print $2}' >> /tmp/totalhash.log

TOKEN=YOU_TELEGRAM_BOT_TOKEN
TOTALHASH="$(/bin/cat /tmp/totalhash.log)"
TARGET="$(/bin/cat /home/m1/target.log)"

FAN=$(/usr/bin/nvidia-smi --query-gpu=fan.speed --format=csv | awk '{print $1}' | tr '\n' ','|sed -e 's/,/  /g' -e 's/, $/\n/')
WATT=$(/usr/bin/nvidia-smi --query-gpu=power.draw --format=csv | awk '{print $1}' | tr '\n' ','|sed -e 's/,/  /g' -e 's/, $/\n/')
TEMP=$(/usr/bin/nvidia-smi --query-gpu=temperature.gpu --format=csv | awk '{print $1}' | tr '\n' ','|sed -e 's/,/  /g' -e 's/, $/\n/')
UPTIME=$(/usr/bin/uptime -p)
sleep 2

if [[ $TARGET > $TOTALHASH ]]; then
    wget -O /dev/null "http://crierbot.appspot.com/${TOKEN}/send?message=Warning!!!
host $(cat /etc/hostname) too slow hash
ip: $(ifconfig | grep inet | grep -v inet6 | grep -v 127.0.0.1 | cut -d: -f2 | awk '{printf $1}')
$UPTIME
$FAN
$WATT
$TEMP
Total Hash: $TOTALHASH"
    else
        echo > /dev/null
fi

/usr/bin/tee /tmp/miner.log < /dev/null
/usr/bin/tail -n 10 /tmp/miner.output.tmp | grep MH/s | awk '{print $3, $9}' | tr -d MH/s >> /tmp/miner.log
/usr/bin/tee /tmp/miner.output.tmp < /dev/null
#=====================

I created the /home/m1/target.log file and entered the approximate value of HASH for RIG
In crontab, I added the line that is executed every 3 minutes
*/3 *   * * *    m1      /home/m1/telegram.sh

The script compares the value from the /home/m1/target.log file with the value that
the /tmp/totalhash.log receives from the file and if there is a difference, sends me a notification.

I hope that our community optimizes this script and writes more universal, which wil work with all mining programs.

I am going to merge a lot of the notification scripts from you, lost_post, BaliMiner, and Maxximus007.  I will also make them variable for using telegram or email by user selection. 

A version of this method might be the best way to do this.

[fullzero]

fullzero, I'm just installed NvOc and I have a problem. For testing I installed 1 videocard 1080 Ti with Asus Prime Z270-A and connected monitor locally.
Every my attempt I have a error "Utilization is too low: reviving did not work so restarting system in 10 seconds", also EWBF showed 5-59 Sol/s on ZEC and crashed after some time. Also user interface is very slow and lagging. In settings I can select only 1 resolution 2560x1440.
I'm use 018v with updated 1bash.
Maybe you have idea how can I fix it?

What imaging program did you use?

HDDRawCopy

What is

Code:

COIN="ZEC"

set to?

Also

What kind of USB key or SSD are you using?


Depending on if you have:

Code:

COIN="X"

where X is one of the following:

 DMD  GRS  ZPOOL_LYRA2V2  ZPOOL_BLAKE2S
 ZEC   ZCOIN  HUSH   ZEN  ZCL 
 NICE_ETHASH  ETH    MUSIC  ETC  EXP  DCR  PASC
 MONA  VTC    DGB    SIA  FTC  LBC 
 DUAL_ETC_DCR    DUAL_ETC_PASC    DUAL_ETC_LBC    DUAL_ETC_SC
 DUAL_EXP_DCR    DUAL_EXP_PASC    DUAL_EXP_LBC    DUAL_EXP_SC
 DUAL_ETH_DCR    DUAL_ETH_PASC    DUAL_ETH_LBC    DUAL_ETH_SC
 DUAL_MUSIC_DCR  DUAL_MUSIC_PASC  DUAL_MUSIC_LBC  DUAL_MUSIC_SC


If you do; then

I would try re imaging / adding updated 1bash and files.

[WoolF911]

What is

Code:

COIN="ZEC"

set to?

Also

What kind of USB key or SSD are you using?


Depending on if you have:

Code:

COIN="X"

where X is one of the following:

 DMD  GRS  ZPOOL_LYRA2V2  ZPOOL_BLAKE2S
 ZEC   ZCOIN  HUSH   ZEN  ZCL  
 NICE_ETHASH  ETH    MUSIC  ETC  EXP  DCR  PASC
 MONA  VTC    DGB    SIA  FTC  LBC  
 DUAL_ETC_DCR    DUAL_ETC_PASC    DUAL_ETC_LBC    DUAL_ETC_SC
 DUAL_EXP_DCR    DUAL_EXP_PASC    DUAL_EXP_LBC    DUAL_EXP_SC
 DUAL_ETH_DCR    DUAL_ETH_PASC    DUAL_ETH_LBC    DUAL_ETH_SC
 DUAL_MUSIC_DCR  DUAL_MUSIC_PASC  DUAL_MUSIC_LBC  DUAL_MUSIC_SC


If you do; then

I would try re imaging / adding updated 1bash and files.

Also I have the error "Utilization is too low: restart 1bash" and multiple "Maxximus007_AUTO_TEMPERATURE_CONTROL" tabs. What it's mean? Maybe this is a reason of problem?

Code:

ZEC_WORKER="nv$IP_AS_WORKER"
ZEC_ADDRESS="t1fs7FDSwQPayfp9B95NNw5LKLdqnfuTEcF"
ZEC_POOL="zec-ru1.dwarfpool.com"
ZEC_PORT="3337"

EWBF_VERSION="3_4"  # choose 3_3  or  3_4

I'm trying to change COIN="ZEC" to COIN="ZCL", but I still have 0-50 Sol/s.

I'm using Intenso Ultra Line USB3.0 32GB.

I trying to add 1bash and files again, but result is same.

[TenaciousJ]

I made an updated 1bash which should resolve these powerlimit / remote issues.  With the powerlimits the autotemp was not reinitializing unless explicitly killed or the rig was logged out or rebooted.  This had the effect of not allowing changes to the individual powerlimits until such killing or logout / reboot.

In your picture I can see another problem which is most likely what has been killing your 1bash prematurely.

By removing the unused individual powerlimit variables you have created a situation where later in 1bash those variables are undefined. 

For now; don't delete unused variables. 

I can add logic to check or otherwise avoid this type of problem in the future; but it is simple enough to leave the extra variables for now.

That seems to have resolved the issue.. thanks.

[fullzero]

What is

Code:

COIN="ZEC"

set to?

Also

What kind of USB key or SSD are you using?


Depending on if you have:

Code:

COIN="X"

where X is one of the following:

 DMD  GRS  ZPOOL_LYRA2V2  ZPOOL_BLAKE2S
 ZEC   ZCOIN  HUSH   ZEN  ZCL  
 NICE_ETHASH  ETH    MUSIC  ETC  EXP  DCR  PASC
 MONA  VTC    DGB    SIA  FTC  LBC  
 DUAL_ETC_DCR    DUAL_ETC_PASC    DUAL_ETC_LBC    DUAL_ETC_SC
 DUAL_EXP_DCR    DUAL_EXP_PASC    DUAL_EXP_LBC    DUAL_EXP_SC
 DUAL_ETH_DCR    DUAL_ETH_PASC    DUAL_ETH_LBC    DUAL_ETH_SC
 DUAL_MUSIC_DCR  DUAL_MUSIC_PASC  DUAL_MUSIC_LBC  DUAL_MUSIC_SC


If you do; then

I would try re imaging / adding updated 1bash and files.

Also I have the error "Utilization is too low: restart 1bash" and multiple "Maxximus007_AUTO_TEMPERATURE_CONTROL" tabs. What it's mean? Maybe this is a reason of problem?

Code:

ZEC_WORKER="nv$IP_AS_WORKER"
ZEC_ADDRESS="t1fs7FDSwQPayfp9B95NNw5LKLdqnfuTEcF"
ZEC_POOL="zec-ru1.dwarfpool.com"
ZEC_PORT="3337"

EWBF_VERSION="3_4"  # choose 3_3  or  3_4

I'm trying to change COIN="ZEC" to COIN="ZCL", but I still have 0-50 Sol/s.

I'm using Intenso Ultra Line USB3.0 32GB.

I trying to add 1bash and files again, but result is same.

I would change

Code:

ZEC_PORT="3337"

to

Code:

ZEC_PORT="3333"

But your ZEC config looks like it should work with DwarfPool.

It is also worth trying removing the workername:

Code:

ZEC_WORKER="nv$IP_AS_WORKER"

to

Code:

ZEC_WORKER=""

With:

Also I have the error "Utilization is too low: restart 1bash" and multiple "Maxximus007_AUTO_TEMPERATURE_CONTROL" tabs. What it's mean? Maybe this is a reason of problem?

Due to some changes I made in the last 1bash; currently when in LOCAL mode: each restart of 1bash will kill the existing Maxximus007_AUTO_TEMPERATURE_CONTROL process, but not kill the tab containing it; then a new Maxximus007_AUTO_TEMPERATURE_CONTROL tab will be created with a new process inside.

The problem is likely that you are not connecting to the pool for some reason; when this happens the watchdog will detect low utilization and restart 1bash.  

You can see if you can reach the pool by opening a new guake tab and entering the following cmd:

Code:

ping zec-ru1.dwarfpool.com

I would try to use a different pool and see if you can connect.  For now while troubleshooting; I would set :

Code:

IAmNotAJeep_and_Maxximus007_WATCHDOG="NO"

[WoolF911]

I would change

Code:

ZEC_PORT="3337"

to

Code:

ZEC_PORT="3333"

But your ZEC config looks like it should work with DwarfPool.

It is also worth trying removing the workername:

Code:

ZEC_WORKER="nv$IP_AS_WORKER"

to

Code:

ZEC_WORKER=""

With:

Also I have the error "Utilization is too low: restart 1bash" and multiple "Maxximus007_AUTO_TEMPERATURE_CONTROL" tabs. What it's mean? Maybe this is a reason of problem?

Due to some changes I made in the last 1bash; currently when in LOCAL mode: each restart of 1bash will kill the existing Maxximus007_AUTO_TEMPERATURE_CONTROL process, but not kill the tab containing it; then a new Maxximus007_AUTO_TEMPERATURE_CONTROL tab will be created with a new process inside.

The problem is likely that you are not connecting to the pool for some reason; when this happens the watchdog will detect low utilization and restart 1bash.  

You can see if you can reach the pool by opening a new guake tab and entering the following cmd:

Code:

ping zec-ru1.dwarfpool.com

I would try to use a different pool and see if you can connect.  For now while troubleshooting; I would set :

Code:

IAmNotAJeep_and_Maxximus007_WATCHDOG="NO"

fullzero, thank you for your reply. I'm sure that is no problem with connection to the pool, also ping is ok. I think it's a some problem with powerlimit, because when I disabled watchdog and trying to change some parameters, include powerlimit, my hashrate is growing to 200 Sol/s, also when I changed some parameters I seen my hashrate grow to 400 Sol/s, but I don't remember parameters what I changed. So at this moment I have hashrate 200 Sol/s with powerlimit 200, when I trying to change it to 250 I have lower hashrate. Cc 150, mc 900 as you wrote in the first message here. Also console showed power consumption lower than I set powerlimit. Any idea?