[ mining os ] nvoc

[Stubo]

while we are on the programming topic... I really can't stand the misuse of logs to pass information/variables between the scripts. not only that the logs slow down everything, they are killing the usb sticks, causing all kinds of problems, hangs, freezes, etc. and i see more and more people are recomanding usage of screenlog

how about using named pipes (fifo) to pass info between the scripts. nothing is saved/logged by using a named pipe and it is much faster. leave the logs for what they were really ment to: to record problems

example: we want to get the fan speed and temps from the temp control so we can send it trough telegram. the easy solution is to write all those values from the temp control script into a log, then read from the log and send telegram. the problem is that in order to satisfy the telegram, the temp script has to write line by line that info for each gpu into a log, 24/7 so that the info is there when telegram needs it.

if we rewrite the temp control to keep sending all that info to a named pipe, the info will be there when telegram needs it without writing to a log over and over, 24/7, and write to a log only critical info/errors that would be needed for troubleshooting

i started rewriting the temp control and i intend to incorporate the named pipes but that will break compatibility with telegrams, web stats, and whatever else is fetching the info from logs. i just can't find enough free time to rewrite all the scripts in nvoc

if interested in the development and optimization of nvoc, please google "bash named pipe" and research then share your thoughts, maybe we can split the workload if you are interested

My 2cents:

While the use of named pipes should certainly reduce disk IO and save folks USB sticks, the burden would be shifted to memory. While not a problem for me since all of my rigs happen to have at least 16G of RAM, I know of several users of nvOC that go minimalist with both CPU and memory. I think the current memory usage is around 2G for nvOC so these folks build rigs with only 4G of RAM. So, when you consider the price of RAM versus the price of a small SSD, disk is cheaper. I can buy a $33 60G SSD but 16G of RAM will cost me $130.

[dartec]

Update dstm zm miner easy way :

V19-1.3 update zm 5_1 to 5_4 :

Download v19_1_3_update_zm_5_1_to_5_4
then run :

Code:

bash v19_1_3_update_zm_5_1_to_5_4

V19-1.4 update zm 5_3 to 5_4 :

Download v19_1_4_update_zm_5_3_to_5_4
then run :

Code:

bash v19_1_4_update_zm_5_3_to_5_4

Script will download zm 5.4, move and rename it to its folder, then change all 3main miner lines and it will copy your default 3main to ~/Downloads/3main-default.

Thank you for this @papampi

I think there is something wrong with the .gz file ...
According to Linux:  "zm_0.5.4.tar.gz: HTML document, ASCII text, with very long lines"
So Linux is not seeing it as a tarball, but as a HTML document ..

My 2c....

[Stubo]

DSTM ZCash/Equihash Miner

For those of you running this miner and noticed an outage last night, please see this:

https://bitcointalk.org/index.php?topic=2021765.msg24279876#msg24279876

Apparently this miner uses the EU flypool to collect Dev fees so flypool being down caused this miner to stop mining last night. Be on the lookout for a new version with a fix for this single point of failure.

[papampi]

Update dstm zm miner easy way :

V19-1.3 update zm 5_1 to 5_4 :

Download v19_1_3_update_zm_5_1_to_5_4
then run :

Code:

bash v19_1_3_update_zm_5_1_to_5_4

V19-1.4 update zm 5_3 to 5_4 :

Download v19_1_4_update_zm_5_3_to_5_4
then run :

Code:

bash v19_1_4_update_zm_5_3_to_5_4

Script will download zm 5.4, move and rename it to its folder, then change all 3main miner lines and it will copy your default 3main to ~/Downloads/3main-default.

https://imgur.com/a/OJAlM

I get this error. Running code with putty/ssh

It seems like for some reason we cant download the file from google drive by wget
If any one has any idea how let me know please
I used this code to get the file:

Code:

wget --no-check-certificate 'https://drive.google.com/file/d/12MmjEm-qIpdK7zc22jatQWD9WmAFIOLy' -O /home/m1/Downloads/zm_0.5.4.tar.gz

So...
For now made some changes so the script do the job when you download it yourself

Download zm_0.5.4.tar.gz from HERE
Put it in Downloads folder , then run the new script again

V19-1.3 update zm 5_1 to 5_4 :

Download v19_1_3_update_zm_5_1_to_5_4
then run :

Code:

bash v19_1_3_update_zm_5_1_to_5_4

V19-1.4 update zm 5_3 to 5_4 :

Download v19_1_4_update_zm_5_3_to_5_4
then run :

Code:

bash v19_1_4_update_zm_5_3_to_5_4

[Rumo]

Because of the problems this morning with all the pools i got an idea.

Is it possible to have a backup pool for every coin?. If pool 1 is not working switch to pool 2. If pool 2 is not working switch to pool 1 from the second coin in the wtm list and so on. It would save me 300 telegrams per hour ;-).

regards

Matze

[Stubo]

Because of the problems this morning with all the pools i got an idea.

Is it possible to have a backup pool for every coin?. If pool 1 is not working switch to pool 2. If pool 2 is not working switch to pool 1 from the second coin in the wtm list and so on. It would save me 300 telegrams per hour ;-).

regards

Matze

This morning's fix for me was to not switch pools but rather to switch miners since DSTM's miner was affected by flypool. That kind of nutty behavior is an outlier even more so than a pool being down. That being said, anything is possible with nvOC if enough folks want it. Otherwise, it just adds more complexity and configuration to a system that already confounds some users.

[kk003]

while we are on the programming topic... I really can't stand the misuse of logs to pass information/variables between the scripts. not only that the logs slow down everything, they are killing the usb sticks, causing all kinds of problems, hangs, freezes, etc. and i see more and more people are recomanding usage of screenlog

how about using named pipes (fifo) to pass info between the scripts. nothing is saved/logged by using a named pipe and it is much faster. leave the logs for what they were really ment to: to record problems

example: we want to get the fan speed and temps from the temp control so we can send it trough telegram. the easy solution is to write all those values from the temp control script into a log, then read from the log and send telegram. the problem is that in order to satisfy the telegram, the temp script has to write line by line that info for each gpu into a log, 24/7 so that the info is there when telegram needs it.

if we rewrite the temp control to keep sending all that info to a named pipe, the info will be there when telegram needs it without writing to a log over and over, 24/7, and write to a log only critical info/errors that would be needed for troubleshooting

i started rewriting the temp control and i intend to incorporate the named pipes but that will break compatibility with telegrams, web stats, and whatever else is fetching the info from logs. i just can't find enough free time to rewrite all the scripts in nvoc

if interested in the development and optimization of nvoc, please google "bash named pipe" and research then share your thoughts, maybe we can split the workload if you are interested

My 2cents:

While the use of named pipes should certainly reduce disk IO and save folks USB sticks, the burden would be shifted to memory. While not a problem for me since all of my rigs happen to have at least 16G of RAM, I know of several users of nvOC that go minimalist with both CPU and memory. I think the current memory usage is around 2G for nvOC so these folks build rigs with only 4G of RAM. So, when you consider the price of RAM versus the price of a small SSD, disk is cheaper. I can buy a $33 60G SSD but 16G of RAM will cost me $130.

Agreed. I like to add that (as someone said a few days a go); we spend thousands $ in a rig but go minimal on data storage units.
About log files as I already said before there should not be problem to log as long as at least a HDD is being used.

Here is an example code (from kk003_telegram) to eliminate the "-L" flag when running scripts from screen, so may be some scripts that don't need a full log file can benefit of it.
Screen has other ways to do this like hardcopy.

Code:

SLEEP_TIME_MODIFICATION_FILE=12         # Hope 12 segs is enough for screen to write enough new data and save the file

# Start log file for a few seconds and collect miner's statics
if [[ -f ~/kk003_telegram_data/kk003_screenlog.log ]]; then
    > ~/kk003_telegram_data/kk003_screenlog.log                                 # Clear my log file
fi

# Need to check here if miner exists before any "screen -dr miner"

screen -dr miner -X logfile ~/kk003_telegram_data/kk003_screenlog.log  # Don't want to overwrite the user's screen log file (screenlog.0 usually)
screen -dr miner -X log                                                                         # Start login miner

echo "Collecting statics from $ETHMINER_or_GENOIL_or_CLAYMORE for $SLEEP_TIME_MODIFICATION_FILE seconds..."

sleep $SLEEP_TIME_MODIFICATION_FILE                                              # Sleep while colleting data from miner
screen -dr miner -X log                                                                         # Stop login miner

[crazydane]

Is anyone dual mining ZEC and XMR?  I see in 1bash where it is possible to enable mining XMR on the CPU.

I see this in 1bash:

Code:

# Mine XMR with CPU
plusCPU="NO"		        # YES NO

threadCOUNT="1"                 # Number of threads for plusCPU: varies per CPU

and

Code:

# if plusCPU is "YES" replace with your XMR info
XMR_WORKER=$WORKERNAME
XMR_ADDRESS="replace_with_your_XMR_address"
XMR_POOL="xmr-us-east1.nanopool.org:14444"

Both of the above are pretty straight forward.

But then we have this:

Code:

   if [ $COIN == "XMR" -o $COIN == "KRB" ]
   then
      POWERLIMIT_WATTS=62
      __CORE_OVERCLOCK=155
      MEMORY_OVERCLOCK=-300
   fi

Can I assume it will get ignored when I'm CPU mining XMR?

And last but not least, assuming I have a decent CPU to begin with, will dual mining ZEC (on GPU) and XMR on CPU, impact GPU mining performance and/or stability?

[Stubo]

Is anyone dual mining ZEC and XMR?  I see in 1bash where it is possible to enable mining XMR on the CPU.

I see this in 1bash:

Code:

# Mine XMR with CPU
plusCPU="NO"		        # YES NO

threadCOUNT="1"                 # Number of threads for plusCPU: varies per CPU

and

Code:

# if plusCPU is "YES" replace with your XMR info
XMR_WORKER=$WORKERNAME
XMR_ADDRESS="replace_with_your_XMR_address"
XMR_POOL="xmr-us-east1.nanopool.org:14444"

Both of the above are pretty straight forward.

But then we have this:

Code:

   if [ $COIN == "XMR" -o $COIN == "KRB" ]
   then
      POWERLIMIT_WATTS=62
      __CORE_OVERCLOCK=155
      MEMORY_OVERCLOCK=-300
   fi

Can I assume it will get ignored when I'm CPU mining XMR?

And last but not least, assuming I have a decent CPU to begin with, will dual mining ZEC (on GPU) and XMR on CPU, impact GPU mining performance and/or stability?

If you read 1bash, the last part is only executed if damNmad_ALGO_SPECIFIC_OC=YES, default is NO. So you should be good by just setting plusCPU=YES and the XMR pool details.

Hope this helps.

[Rumo]

Because i run a few rigs and all of them are different i moved out the OC settings from 1bash.

Save the code https://drive.google.com/open?id=1knH6GqiPDLKoURkmKGoXDtHJ4Agxi-nj as 1OC in your home folder.  

To use it place

source ~/1OC      #Overclock settings

under  

COIN=""

in 1bash and delete the OC stuff in 1bash.

Hope it is helpful.

regards

Matze

[Rumo]

I added the nicehash stuff to the wtm autoswitch coins if someone is interested.

1. ADD
NICE_SKUNKHASH;NICE_EQUIHASH;NICE_ETHASH;NICE_NEOSCRYPT;NICE_X11GOST;NICE_CRYPTONIGHT;NICE_LYRA2REV2
to the WTM coins in 1bash.

2. in WTM_AUTO_SWITCH Change
# filter WTM coins by user selection only
for i in reversed(data):
    if i["tag"] not in includedCoins:
        data.remove(i)

to

# filter WTM coins by user selection only
for i in reversed(data):
    if i["tag"] == "NICEHASH":
       i["tag"] = "NICE_" + i["algorithm"].upper()

for i in reversed(data):
    if i["tag"] not in includedCoins:
       data.remove(i)

3. in PAPAMPI_WTM change the TOP_COIN= line to TOP_COIN=$(cat WTM_top_coin | grep : | sed 's/:[0-9]*//')

4. ADD

NICEHASH_BTC_ADDRESS="Your nicehash BTC address here"

and

#NICE_LYRA2REV2
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS=$NICEHASH_BTC_ADDRESS
NICE_LYRA2REV2_POOL="stratum+tcp://lyra2rev2.eu.nicehash.com:3347"
NICE_LYRA2REV2_INTENSITY="21"

#NICE_NEOSCRYPT
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_NEOSCRYPT_POOL="stratum+tcp://neoscrypt.eu.nicehash.com:3341"
NICE_NEOSCRYPT_INTENSITY="21"

#NICE_CRYPTONIGHT
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_CRYPTONIGHT_POOL="stratum+tcp://cryptonight.eu.nicehash.com:3355"
NICE_CRYPTONIGHT_INTENSITY="21"

#NICE_X11GOST
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_X11GOST_POOL="stratum+tcp://x11gost.eu.nicehash.com:3359"
NICE_X11GOST_INTENSITY="21"

#NICE_SKUNKHASH
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_SKUNKHASH_POOL="stratum+tcp://skunk.eu.nicehash.com:3362"
NICE_SKUNKHASH_INTENSITY="21"

to your coins in 1bash

5. ADD

########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################

if [ $COIN == "DUAL_NICEHASH" ]
then

HCD=/home/m1/eth/$CLAYMORE_VERSION/ethdcrminer64

ETHASH_ADDR="$DUAL_BTC_ADDRESS.$DUAL_WORKER"
DCR_ADDR="$DUAL_BTC_ADDRESS.$DUAL_WORKER"

screen -dmSL miner $HCD -epool $DUAL_ETHASH_POOL -ewal $ETHASH_ADDR -epsw x -allcoins 1 -allpools 1 -dpool $DUAL_DCR_POOL -dwal $DCR_ADDR -dpsw x $DUAL_EXTENSION_ARGUMENTS

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_NEOSCRYPT" ]
then
HCD='/home/m1/SPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSLL miner $HCD -a neoscrypt -o $NICE_NEOSCRYPT_POOL -u $ADDR -p x -i $PXC_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_LYRA2REV2" ]
then
HCD='/home/m1/SPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSLL miner $HCD -a lyra2v2 -o $NICE_LYRA2REV2_POOL -u $ADDR -p x -i $NICE_LYRA2REV2_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_X11GOST" ]
then
HCD='/home/m1/SPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSLL miner $HCD -a sib -o $NICE_X11GOST_POOL -u $ADDR -p x -i $NICE_X11GOST_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_SKUNKHASH" ]
then
HCD='/home/m1/TPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSLL miner $HCD -a skunk -o $NICE_SKUNKHASH_POOL -u $ADDR -p x -i $NICE_SKUNKHASH_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_CRYPTONIGHT" ]
then
HCD='/home/m1/TPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSLL miner $HCD -a cryptonight -o $NICE_CRYPTONIGHT_POOL -u $ADDR -p x -i $NICE_CRYPTONIGHT_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################

to 3main

Hope this helps

regards

Matze

[leenoox]

while we are on the programming topic... I really can't stand the misuse of logs to pass information/variables between the scripts. not only that the logs slow down everything, they are killing the usb sticks, causing all kinds of problems, hangs, freezes, etc. and i see more and more people are recomanding usage of screenlog

how about using named pipes (fifo) to pass info between the scripts. nothing is saved/logged by using a named pipe and it is much faster. leave the logs for what they were really ment to: to record problems

example: we want to get the fan speed and temps from the temp control so we can send it trough telegram. the easy solution is to write all those values from the temp control script into a log, then read from the log and send telegram. the problem is that in order to satisfy the telegram, the temp script has to write line by line that info for each gpu into a log, 24/7 so that the info is there when telegram needs it.

if we rewrite the temp control to keep sending all that info to a named pipe, the info will be there when telegram needs it without writing to a log over and over, 24/7, and write to a log only critical info/errors that would be needed for troubleshooting

i started rewriting the temp control and i intend to incorporate the named pipes but that will break compatibility with telegrams, web stats, and whatever else is fetching the info from logs. i just can't find enough free time to rewrite all the scripts in nvoc

if interested in the development and optimization of nvoc, please google "bash named pipe" and research then share your thoughts, maybe we can split the workload if you are interested

My 2cents:

While the use of named pipes should certainly reduce disk IO and save folks USB sticks, the burden would be shifted to memory. While not a problem for me since all of my rigs happen to have at least 16G of RAM, I know of several users of nvOC that go minimalist with both CPU and memory. I think the current memory usage is around 2G for nvOC so these folks build rigs with only 4G of RAM. So, when you consider the price of RAM versus the price of a small SSD, disk is cheaper. I can buy a $33 60G SSD but 16G of RAM will cost me $130.

The bauty of named pipes... read here a brief explanation:

https://askubuntu.com/questions/449132/why-use-a-named-pipe-instead-of-a-file

unlike ramdisk, it doesn't consume memory nor it uses disk I/O

[leenoox]

Thanks a lot mate.
It was always a question for me what those loops are for...
Tested the miner start script without the loops and all is good.

I think as you said its some legacy code from old nvOC and can be removed.

And as I said before that wtm-miner is just a copy of 3main miner starts so wtm auto switch dont restart 3main which will take so long and just switch miner.

I rolled a version of 3main without the loops this morning and tested it with poor results. It would appear that the miner runs but the watchdog kills it because of low utilization after the 70 second grace period. So, what I think is going on here is that we need 3main to continue to run not because of the mining screen but because of the OC and PL settings. When 3main ends, those go away, and the miner starts performing poorly. What I did to fix it was to add a single infinite loop at the end of 3main to make sure it never exits. So, at least we can clean up 3main substantially by removing the 30+ instances of that loop. My version of 3main is left with only 3 - two in the SALFTER sections (because I haven't bothered to figure out what is going on there) and the one I added back in at the end.

Can you try enabling persistence mode

Code:

nvidia-smi -pm 1

remove the loop at the end and try if this fixes the utilization after 3main exits

[Stubo]

The bauty of named pipes... read here a brief explanation:

https://askubuntu.com/questions/449132/why-use-a-named-pipe-instead-of-a-file

unlike ramdisk, it doesn't consume memory nor it uses disk I/O

Of course it uses memory. It says right there that the kernel handles it which means kernel memory. So the kernel grows in size to accommodate the pipe contents.

[sergixc]

Hi, what about full support of Asus b250 mining expert.

What does it mean "partly supported"?

Is it a hard limit of 13 standard GPU + 6 P106? or will be possible to use 19 regular GPU?

My mobo does not even post bios messages when Iam inserting 14th regular GPU

Could anybody comment on this?

Thank you in advance!

Have you attached 3rd PSU?
One PSU = 7 GPU
Two PSU's = 13 GPU
Three PSU's = 19 GPU

Partial support means that not all GPU's can be overclocked, first 16 GPU can be overclocked, last 3 GPU will work on stock settings. See the post from two pages back regarding TB250

https://bitcointalk.org/index.php?topic=1854250.msg23785755#msg23785755

I have attached just 1 PSU (3000watt) and attached 13 regula GPUs - they are fine, but when I am attaching 14th GPU - mobo does not even post any messages and does not start at all.
And I am talking about *REGULAR* GPUs, not P106!
Did anybody start this mobo with 14 and more *REGULAR* GPUs ? NOT P106

Thank you

[papampi]

What I meant by separating the miner start codes from 3main, is what I'm already doing in my latest WTM.
3main handles wdog, temp, power limit, ... and the miner start up,
It does not need to have miner start codes inside, but can just fire up the script that have all the startups inside.

In my latest switch, 3main starts the first mining process and sits back, then wtm switcher checks for better coin , stop mining process and start new one by just executing the file I put all mining startups inside without any loops (callled it wtm-miner)

3main now has over 3500 lines, which over 2500 lines are miner startups, with over 80 loops for miners.

[Stubo]

Thanks a lot mate.
It was always a question for me what those loops are for...
Tested the miner start script without the loops and all is good.

I think as you said its some legacy code from old nvOC and can be removed.

And as I said before that wtm-miner is just a copy of 3main miner starts so wtm auto switch dont restart 3main which will take so long and just switch miner.

I rolled a version of 3main without the loops this morning and tested it with poor results. It would appear that the miner runs but the watchdog kills it because of low utilization after the 70 second grace period. So, what I think is going on here is that we need 3main to continue to run not because of the mining screen but because of the OC and PL settings. When 3main ends, those go away, and the miner starts performing poorly. What I did to fix it was to add a single infinite loop at the end of 3main to make sure it never exits. So, at least we can clean up 3main substantially by removing the 30+ instances of that loop. My version of 3main is left with only 3 - two in the SALFTER sections (because I haven't bothered to figure out what is going on there) and the one I added back in at the end.

Can you try enabling persistence mode

Code:

nvidia-smi -pm 1

remove the loop at the end and try if this fixes the utilization after 3main exits

Woah. I get some really odd behavior when I do that. The OS seems to want to hold onto 3main so trying to edit it locally ends up with vi being killed. Also, the temp script keeps getting killed and restarting yet the watchdog continues to run as does the miner. Needless to say, that is not a stable configuration but for different reasons than one would think.

[papampi]

I added the nicehash stuff to the wtm autoswitch coins if someone is interested.

1. ADD to the WTM coins in 1bash.

Code:

NICE_SKUNKHASH;NICE_EQUIHASH;NICE_ETHASH;NICE_NEOSCRYPT;NICE_X11GOST;NICE_CRYPTONIGHT;NICE_LYRA2REV2

2. in WTM_AUTO_SWITCH Change
From:

Code:

# filter WTM coins by user selection only
for i in reversed(data):
    if i["tag"] not in includedCoins:
        data.remove(i)

to:

Code:

# filter WTM coins by user selection only
for i in reversed(data):
    if i["tag"] == "NICEHASH":
       i["tag"] = "NICE_" + i["algorithm"].upper()

for i in reversed(data):
    if i["tag"] not in includedCoins:
       data.remove(i)

3. in PAPAMPI_WTM change the TOP_COIN=
to :

Code:

TOP_COIN=$(cat WTM_top_coin | grep : | sed 's/:[0-9]*//')

4. ADD to your coins in 1bash

Code:

NICEHASH_BTC_ADDRESS="Your nicehash BTC address here"

#NICE_LYRA2REV2
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS=$NICEHASH_BTC_ADDRESS
NICE_LYRA2REV2_POOL="stratum+tcp://lyra2rev2.eu.nicehash.com:3347"
NICE_LYRA2REV2_INTENSITY="21"

#NICE_NEOSCRYPT
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_NEOSCRYPT_POOL="stratum+tcp://neoscrypt.eu.nicehash.com:3341"
NICE_NEOSCRYPT_INTENSITY="21"

#NICE_CRYPTONIGHT
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_CRYPTONIGHT_POOL="stratum+tcp://cryptonight.eu.nicehash.com:3355"
NICE_CRYPTONIGHT_INTENSITY="21"

#NICE_X11GOST
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_X11GOST_POOL="stratum+tcp://x11gost.eu.nicehash.com:3359"
NICE_X11GOST_INTENSITY="21"

#NICE_SKUNKHASH
NICE_WORKER="$WORKERNAME"
NICE_ADDRESS="$NICEHASH_BTC_ADDRESS"
NICE_SKUNKHASH_POOL="stratum+tcp://skunk.eu.nicehash.com:3362"
NICE_SKUNKHASH_INTENSITY="21"

5. ADD to 3main or wtm-miner for latest version

Code:

########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################

if [ $COIN == "DUAL_NICEHASH" ]
then

HCD=/home/m1/eth/$CLAYMORE_VERSION/ethdcrminer64

ETHASH_ADDR="$DUAL_BTC_ADDRESS.$DUAL_WORKER"
DCR_ADDR="$DUAL_BTC_ADDRESS.$DUAL_WORKER"

screen -dmSL miner $HCD -epool $DUAL_ETHASH_POOL -ewal $ETHASH_ADDR -epsw x -allcoins 1 -allpools 1 -dpool $DUAL_DCR_POOL -dwal $DCR_ADDR -dpsw x $DUAL_EXTENSION_ARGUMENTS

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_NEOSCRYPT" ]
then
HCD='/home/m1/SPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSL miner $HCD -a neoscrypt -o $NICE_NEOSCRYPT_POOL -u $ADDR -p x -i $PXC_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_LYRA2REV2" ]
then
HCD='/home/m1/SPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSL miner $HCD -a lyra2v2 -o $NICE_LYRA2REV2_POOL -u $ADDR -p x -i $NICE_LYRA2REV2_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_X11GOST" ]
then
HCD='/home/m1/SPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSL miner $HCD -a sib -o $NICE_X11GOST_POOL -u $ADDR -p x -i $NICE_X11GOST_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_SKUNKHASH" ]
then
HCD='/home/m1/TPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSL miner $HCD -a skunk -o $NICE_SKUNKHASH_POOL -u $ADDR -p x -i $NICE_SKUNKHASH_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

if [ $COIN == "NICE_CRYPTONIGHT" ]
then
HCD='/home/m1/TPccminer/ccminer'
ADDR="$NICE_ADDRESS.$NICE_WORKER"

screen -dmSL miner $HCD -a cryptonight -o $NICE_CRYPTONIGHT_POOL -u $ADDR -p x -i $NICE_CRYPTONIGHT_INTENSITY

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################
########################################################## Nicehash ###############################################

Hope this helps

regards

Matze

Thanks mate
I tried to put your edits in codes to be easier to understand.

But we already have a nicehash auto switch, is this one needed ?

[Alienbert]

Help!

I mined ZEC on a 13 GPU RIG long time without Problems. Suddenly the works stoped!
Now after restart the miner doesnt starts and i get this:

Thu Nov 9 10:16:35 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:16:45 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:16:55 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:17:06 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:17:16 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:17:26 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:18:37 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:19:48 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:20:59 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:21:10 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:22:21 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures

Thu Nov  9 10:26:22 MST 2017 - Starting miner restart script.

error: process ID list syntax error

Usage:
 ps [options]

 Try 'ps --help <simple|list|output|threads|misc|all>'
  or 'ps --help <s|l|o|t|m|a>'
 for additional help text.

For more details see ps(1).
Thu Nov  9 10:26:24 MST 2017 - waiting 70 seconds before going 'on watch'

      GPU_COUNT:  13

GPU UTILIZATION:  0 0 0 0 0 0 0 0 0 0 0 0 0

Thu Nov  9 10:27:35 MST 2017 - GPU under threshold found - GPU UTILIZATION:  0 0 0 0 0 0 0 0 0 0 0 0 0


Now i tried ETH-SIA:
1. the miner start, no mining, restart and then he works, but not good

Thu Nov  9 10:59:02 MST 2017 - GPU under threshold found - GPU UTILIZATION:  89 99 97 100 99 98 98 98 95 98 88 99 100

Thu Nov  9 10:59:02 MST 2017 - GPU under threshold found - GPU UTILIZATION:  89 99 97 100 99 98 98 98 95 98 88 99 100

Thu Nov  9 10:59:02 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures


      GPU_COUNT:  13

GPU UTILIZATION:  96 100 99 94 100 91 95 91 97 95 99 97 99

Thu Nov  9 10:59:55 MST 2017 - 5 by 5: REMEMBER TO THANK IAmNotAJeep and Maxximus007

      GPU_COUNT:  13

GPU UTILIZATION:  94 96 96 99 98 99 100 99 88 100 88 93 99

Thu Nov  9 11:00:40 MST 2017 - GPU under threshold found - GPU UTILIZATION:  94 96 96 99 98 99 100 99 88 100 88 93 99

Thu Nov  9 11:00:40 MST 2017 - GPU under threshold found - GPU UTILIZATION:  94 96 96 99 98 99 100 99 88 100 88 93 99

Thu Nov  9 11:00:40 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures


      GPU_COUNT:  13

What do you think? Is that the PSU? I tried with 6 cards - same message

What can i do 

[papampi]

Help!

I mined ZEC on a 13 GPU RIG long time without Problems. Suddenly the works stoped!
Now after restart the miner doesnt starts and i get this:

Thu Nov 9 10:16:35 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:16:45 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:16:55 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:17:06 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:17:16 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:17:26 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:18:37 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:19:48 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:20:59 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures Thu Nov 9 10:21:10 MST 2017 - Utilization is too low: restart 3main Thu Nov 9 10:22:21 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures

Thu Nov  9 10:26:22 MST 2017 - Starting miner restart script.

error: process ID list syntax error

Usage:
 ps [options]

 Try 'ps --help <simple|list|output|threads|misc|all>'
  or 'ps --help <s|l|o|t|m|a>'
 for additional help text.

For more details see ps(1).
Thu Nov  9 10:26:24 MST 2017 - waiting 70 seconds before going 'on watch'

      GPU_COUNT:  13

GPU UTILIZATION:  0 0 0 0 0 0 0 0 0 0 0 0 0

Thu Nov  9 10:27:35 MST 2017 - GPU under threshold found - GPU UTILIZATION:  0 0 0 0 0 0 0 0 0 0 0 0 0


Now i tried ETH-SIA:
1. the miner start, no mining, restart and then he works, but not good

Thu Nov  9 10:59:02 MST 2017 - GPU under threshold found - GPU UTILIZATION:  89 99 97 100 99 98 98 98 95 98 88 99 100

Thu Nov  9 10:59:02 MST 2017 - GPU under threshold found - GPU UTILIZATION:  89 99 97 100 99 98 98 98 95 98 88 99 100

Thu Nov  9 10:59:02 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures


      GPU_COUNT:  13

GPU UTILIZATION:  96 100 99 94 100 91 95 91 97 95 99 97 99

Thu Nov  9 10:59:55 MST 2017 - 5 by 5: REMEMBER TO THANK IAmNotAJeep and Maxximus007

      GPU_COUNT:  13

GPU UTILIZATION:  94 96 96 99 98 99 100 99 88 100 88 93 99

Thu Nov  9 11:00:40 MST 2017 - GPU under threshold found - GPU UTILIZATION:  94 96 96 99 98 99 100 99 88 100 88 93 99

Thu Nov  9 11:00:40 MST 2017 - GPU under threshold found - GPU UTILIZATION:  94 96 96 99 98 99 100 99 88 100 88 93 99

Thu Nov  9 11:00:40 MST 2017 - Low Utilization Detected: 3main will reinit if there are 6 consecutive failures


      GPU_COUNT:  13

What do you think? Is that the PSU? I tried with 6 cards - same message

What can i do  

What are your cards and what are the OC and power values?
Have you tried zec with dstm ZM miner ?
What is your nvOC version?