[ mining os ] nvoc - page 222. | Bitcointalksearch.org

leenoox

full member

Activity: 200

Merit: 101

Quote from: Temporel on September 20, 2017, 06:46:19 AM

Quote from: leenoox on September 19, 2017, 11:17:06 PM

Quote from: Temporel on September 19, 2017, 08:55:15 PM

Quote from: leenoox on September 19, 2017, 08:17:33 PM

Quote from: Temporel on September 19, 2017, 07:40:39 PM

Quote from: Bigf00t on September 19, 2017, 07:21:59 PM

Set it to LOCAL in 1bash

found it, thank you.

So I made the change and reboot the system and for some reason, the GPUx id of my cards changed. I have 2 1080Ti mixed with the GTX-1070s so I had to manually changed the powerlimit for each cards and now one of the 1070 is set to 225 and a 1080Ti to 135watts.

Weird...

press F12 to open guake terminal, then press the button at the lower right corner to open new tab, then type

Code:

gedit .bashrc

add the following at the bottom:

Code:

export CUDA_DEVICE_ORDER=PCI_BUS_ID

save and restart.
The solution was posted by fullzero with credit given to car1999 about 10-11 pages back

Thank you !
I use the remote output now so I did it using nano in the terminal instead.

EDIT: the ids are still wrong but I think my settings are fine, its just mixed up for some reason in the output.

try this instead, it will load it as permanent global variable for login and non-login, interactive and non-interactive shells:

Code:

sudo echo 'export CUDA_DEVICE_ORDER=PCI_BUS_ID' | sudo tee -a /etc/profile.d/cuda-device-order.sh

you need to logout and login again and restart the mining process for it to take effect (or just reboot).

Please post results.

it works, thank you very much !

Do you think this fix will be include in v0020 ?

You are welcome, I'm glad I could help. Fullzero stated that he will include the fix in the next release but it seems that the old solution (adding to .bashrc) does not work in all cases. As confirmed by you and few others it seems that the latest solution (/etc/profile.d) works well, I hope he will include this in version 20 of nvOC.

In the mean time everyone that experience similar problem, especially people that have rigs with mixed cards should execute that code in shell to add the variable and fix gpu order to prevent applying wrong OC.

scryptr

legendary

Activity: 1797

Merit: 1028

Quote from: Razvan17 on September 20, 2017, 03:38:10 AM

Quote from: nexenta on September 16, 2017, 12:01:18 AM

Guys, I am having issues unzipping the NVOS (nvOC_v0019.zip) on both MAC and PC. The file size shows 6,436,759,477 bytes (6.44 GB on disk), but when I am trying to unzip the size of the image is getting huge (over 250GB!). I downloaded the google drive version of NVOS zip from this forum. Has anyone encountered this issue and can give some advice? Thank you in advance!

Download and install 7-Zip in Windows and then unzip it.

I had the same issue.

7-ZIP WILL PROPERLY HANDLE THE ARCHIVE--

I also had the same issue. The file is not corrupt, but the built-in Windows extractor could not extract the archive. --scryptr

Temporel

full member

Activity: 224

Merit: 100

Quote from: leenoox on September 19, 2017, 11:17:06 PM

Quote from: Temporel on September 19, 2017, 08:55:15 PM

Quote from: leenoox on September 19, 2017, 08:17:33 PM

Quote from: Temporel on September 19, 2017, 07:40:39 PM

Quote from: Bigf00t on September 19, 2017, 07:21:59 PM

Set it to LOCAL in 1bash

found it, thank you.

So I made the change and reboot the system and for some reason, the GPUx id of my cards changed. I have 2 1080Ti mixed with the GTX-1070s so I had to manually changed the powerlimit for each cards and now one of the 1070 is set to 225 and a 1080Ti to 135watts.

Weird...

press F12 to open guake terminal, then press the button at the lower right corner to open new tab, then type

Code:

gedit .bashrc

add the following at the bottom:

Code:

export CUDA_DEVICE_ORDER=PCI_BUS_ID

save and restart.
The solution was posted by fullzero with credit given to car1999 about 10-11 pages back

Thank you !
I use the remote output now so I did it using nano in the terminal instead.

EDIT: the ids are still wrong but I think my settings are fine, its just mixed up for some reason in the output.

try this instead, it will load it as permanent global variable for login and non-login, interactive and non-interactive shells:

Code:

sudo echo 'export CUDA_DEVICE_ORDER=PCI_BUS_ID' | sudo tee -a /etc/profile.d/cuda-device-order.sh

you need to logout and login again and restart the mining process for it to take effect (or just reboot).

Please post results.

it works, thank you very much !

Do you think this fix will be include in v0020 ?

Razvan17

newbie

Activity: 18

Merit: 0

Quote from: nexenta on September 16, 2017, 12:01:18 AM

Guys, I am having issues unzipping the NVOS (nvOC_v0019.zip) on both MAC and PC. The file size shows 6,436,759,477 bytes (6.44 GB on disk), but when I am trying to unzip the size of the image is getting huge (over 250GB!). I downloaded the google drive version of NVOS zip from this forum. Has anyone encountered this issue and can give some advice? Thank you in advance!

Download and install 7-Zip in Windows and then unzip it.

I had the same issue.

papampi

full member

Activity: 686

Merit: 140

Linux FOREVER! Resistance is futile!!!

Quote from: leenoox on September 19, 2017, 11:17:06 PM

try this instead, it will load it as permanent global variable for login and non-login, interactive and non-interactive shells:

Code:

sudo echo 'export CUDA_DEVICE_ORDER=PCI_BUS_ID' | sudo tee -a /etc/profile.d/cuda-device-order.sh

you need to logout and login again and restart the mining process for it to take effect (or just reboot).

Please post results.

Nice workaround @leenoox
Thanks.

papampi

full member

Activity: 686

Merit: 140

Linux FOREVER! Resistance is futile!!!

Quote from: Razvan17 on September 20, 2017, 02:20:10 AM

Hi guys, Im new here on this forum.

First of all I want to thank Mr. Fullzero for this brilliant nvOC.

I just started to use it on the following hardware:

Motherboard - Gigabyte - GA-M61SME-S2
CPU - AMD Athlon 2100+ - 2 cores - 65w
USB 3.0 memory stick SanDisk 32Gb
GPU0 - Assus Gtx 1050Ti 4Gb Expedition
GPU1 - Assus Gtx 1060 6Gb Expedition
Risers - v006

I am currently mining NiceEth with nvOC_v0019 and the results are:
Speed is between 36.7 - 37.8Mh
Power consumption 217w for the entire system on the wall.

OC settings Gtx 1050Ti 4Gb:
GPU clock - 100
Memory clock - 1000 (Micron memory)
Power target - 70
Temp target - 67
The fans are automatic (Manual_Fan="NO") and they run in 49
Read temp - 66-67
Power used - 63-64

OC settings for Gtx 1060 6Gb:
GPU clock - 100
Memory clock - 1000 (Micron memory)
Power target - 100
Temp target - 67
The fans are automatic (Manual_Fan="NO") and they run in 43
Read temp - 66-67
Power used - 98-100

Thx again.

Nice.
Welcome to the family.

TAKCuCT

newbie

Activity: 25

Merit: 0

Quote from: leenoox on September 19, 2017, 11:17:06 PM

try this instead, it will load it as permanent global variable for login and non-login, interactive and non-interactive shells:

Code:

sudo echo 'export CUDA_DEVICE_ORDER=PCI_BUS_ID' | sudo tee -a /etc/profile.d/cuda-device-order.sh

you need to logout and login again and restart the mining process for it to take effect (or just reboot).

Please post results.

Worked! Thanks!

Razvan17

newbie

Activity: 18

Merit: 0

Hi guys, Im new here on this forum.

First of all I want to thank Mr. Fullzero for this brilliant nvOC.

I just started to use it on the following hardware:

Motherboard - Gigabyte - GA-M61SME-S2
CPU - AMD Athlon 2100+ - 2 cores - 65w
USB 3.0 memory stick SanDisk 32Gb
GPU0 - Assus Gtx 1050Ti 4Gb Expedition
GPU1 - Assus Gtx 1060 6Gb Expedition
Risers - v006

I am currently mining NiceEth with nvOC_v0019 and the results are:
Speed is between 36.7 - 37.8Mh
Power consumption 217w for the entire system on the wall.

OC settings Gtx 1050Ti 4Gb:
GPU clock - 100
Memory clock - 1000 (Micron memory)
Power target - 70
Temp target - 67
The fans are automatic (Manual_Fan="NO") and they run in 49
Read temp - 66-67
Power used - 63-64

OC settings for Gtx 1060 6Gb:
GPU clock - 100
Memory clock - 1000 (Micron memory)
Power target - 100
Temp target - 67
The fans are automatic (Manual_Fan="NO") and they run in 43
Read temp - 66-67
Power used - 98-100

Thx again.

papampi

full member

Activity: 686

Merit: 140

Linux FOREVER! Resistance is futile!!!

Quote from: CryptAtomeTrader44 on September 19, 2017, 05:04:28 PM

Quote from: ?? on ??

Quote from: papampi on September 17, 2017, 12:24:52 AM

I could compile KTccminer on nv0019, but not on my other ubuntu
gave me some errors
just download the source
and run the commands with no args

Code:

./autogen.sh
./configure.sh
./build.sh

it should be ok

remember to:

cd to the KTccminer directory first:

cd /home/m1/KTccminer

Klaust aswered under papami's Github ticket.
He said and/or write :

"I have a 1070 too, and under Windows it works.
Did you use the build.sh script to compile it?"

In my case, i just downloaded tar salfter's KTccminer and uncompressed in /home/m1.
But i saw the mph_switching_profit stop mining when Monero bécame most profitable to mine. The rig don't mine nothing before to toggle to anoter coin algorithm.

unfortunately if I have the skills to compile or mining, I do not have those to know how to code. I try to read your scripts, which I sometimes understand, but not always everything and not all the time ...

Thank you very much to you fullzero for this work and the great progress of nvOC.

I'm sure everything can still make great strides. Why not with a web interface for example?

you should open 3main and remove -a cryptonight from KTccminer-cryptonight ccminer
edit the line :

Code:

"bin": "/home/m1/KTccminer-cryptonight/ccminer -a cryptonight -o stratum+tcp://{HOST}:{PORT} -u {NAME}.{MINER} -p x",

to:

Code:

"bin": "/home/m1/KTccminer-cryptonight/ccminer -o stratum+tcp://{HOST}:{PORT} -u {NAME}.{MINER} -p x",

leenoox

full member

Activity: 200

Merit: 101

Quote from: Temporel on September 19, 2017, 08:55:15 PM

Quote from: leenoox on September 19, 2017, 08:17:33 PM

Quote from: Temporel on September 19, 2017, 07:40:39 PM

Quote from: Bigf00t on September 19, 2017, 07:21:59 PM

Set it to LOCAL in 1bash

found it, thank you.

So I made the change and reboot the system and for some reason, the GPUx id of my cards changed. I have 2 1080Ti mixed with the GTX-1070s so I had to manually changed the powerlimit for each cards and now one of the 1070 is set to 225 and a 1080Ti to 135watts.

Weird...

press F12 to open guake terminal, then press the button at the lower right corner to open new tab, then type

Code:

gedit .bashrc

add the following at the bottom:

Code:

export CUDA_DEVICE_ORDER=PCI_BUS_ID

save and restart.
The solution was posted by fullzero with credit given to car1999 about 10-11 pages back

Thank you !
I use the remote output now so I did it using nano in the terminal instead.

EDIT: the ids are still wrong but I think my settings are fine, its just mixed up for some reason in the output.

try this instead, it will load it as permanent global variable for login and non-login, interactive and non-interactive shells:

Code:

sudo echo 'export CUDA_DEVICE_ORDER=PCI_BUS_ID' | sudo tee -a /etc/profile.d/cuda-device-order.sh

you need to logout and login again and restart the mining process for it to take effect (or just reboot).

Please post results.

Temporel

full member

Activity: 224

Merit: 100

Quote from: leenoox on September 19, 2017, 08:17:33 PM

Quote from: Temporel on September 19, 2017, 07:40:39 PM

Quote from: Bigf00t on September 19, 2017, 07:21:59 PM

Set it to LOCAL in 1bash

found it, thank you.

So I made the change and reboot the system and for some reason, the GPUx id of my cards changed. I have 2 1080Ti mixed with the GTX-1070s so I had to manually changed the powerlimit for each cards and now one of the 1070 is set to 225 and a 1080Ti to 135watts.

Weird...

press F12 to open guake terminal, then press the button at the lower right corner to open new tab, then type

Code:

gedit .bashrc

add the following at the bottom:

Code:

export CUDA_DEVICE_ORDER=PCI_BUS_ID

save and restart.
The solution was posted by fullzero with credit given to car1999 about 10-11 pages back

Thank you !
I use the remote output now so I did it using nano in the terminal instead.

EDIT: the ids are still wrong but I think my settings are fine, its just mixed up for some reason in the output.

leenoox

full member

Activity: 200

Merit: 101

Quote from: Temporel on September 19, 2017, 07:40:39 PM

Quote from: Bigf00t on September 19, 2017, 07:21:59 PM

Set it to LOCAL in 1bash

found it, thank you.

So I made the change and reboot the system and for some reason, the GPUx id of my cards changed. I have 2 1080Ti mixed with the GTX-1070s so I had to manually changed the powerlimit for each cards and now one of the 1070 is set to 225 and a 1080Ti to 135watts.

Weird...

press F12 to open guake terminal, then press the button at the lower right corner to open new tab, then type

Code:

gedit .bashrc

add the following at the bottom:

Code:

export CUDA_DEVICE_ORDER=PCI_BUS_ID

save and restart.
The solution was posted by fullzero with credit given to car1999 about 10-11 pages back

Temporel

full member

Activity: 224

Merit: 100

Quote from: Bigf00t on September 19, 2017, 07:21:59 PM

Set it to LOCAL in 1bash

found it, thank you.

So I made the change and reboot the system and for some reason, the GPUx id of my cards changed. I have 2 1080Ti mixed with the GTX-1070s so I had to manually changed the powerlimit for each cards and now one of the 1070 is set to 225 and a 1080Ti to 135watts.

Weird...

Bigf00t

newbie

Activity: 60

Merit: 0

Set it to LOCAL in 1bash

Temporel

full member

Activity: 224

Merit: 100

finally up and running with 11 cards, all I have to say is well done fullzero, this works really well.

I have a question, how do I get the ewbf output ? I have the terminal window open with all the info and those last lines:

LAUNCHING: Mining Process
process in screen miner; attach with: screen -r miner

but no ewbf infos like I have in Windows.

Thank you.

CryptAtomeTrader44

full member

Activity: 340

Merit: 103

It is easier to break an atom than partialities AE

Quote from: ?? on ??

Quote from: papampi on September 17, 2017, 12:24:52 AM

I could compile KTccminer on nv0019, but not on my other ubuntu
gave me some errors
just download the source
and run the commands with no args

Code:

./autogen.sh 
./configure.sh
./build.sh

it should be ok

remember to:

cd to the KTccminer directory first:

cd /home/m1/KTccminer

Klaust aswered under papami's Github ticket.
He said and/or write :

"I have a 1070 too, and under Windows it works.
Did you use the build.sh script to compile it?"

In my case, i just downloaded tar salfter's KTccminer and uncompressed in /home/m1.
But i saw the mph_switching_profit stop mining when Monero bécame most profitable to mine. The rig don't mine nothing before to toggle to anoter coin algorithm.

unfortunately if I have the skills to compile or mining, I do not have those to know how to code. I try to read your scripts, which I sometimes understand, but not always everything and not all the time ...

Thank you very much to you fullzero for this work and the great progress of nvOC.

I'm sure everything can still make great strides. Why not with a web interface for example?

CryptAtomeTrader44

full member

Activity: 340

Merit: 103

It is easier to break an atom than partialities AE

I read this post: subject [nicehash vs zpool vs miningpoolhub - And the winner is ...]
https://bitcointalksearch.org/topic/nicehashcom-vs-zpoolca-vs-miningpoolhubcom-and-the-winner-is-2067256
The author displays a table with comparisons betwen these three pool and concludes :
"Even with the talk about zpool taking 20% or what not, for me, it still comes out on top, so that just proves that zpool is more profitable for multi-algo switching the BTC wallet. "

Would it be possible to make a switching profit script for multipool Zpool multi-algorithm in the future ?

More complex: Would it be possible to make a script that would switch between minning profit scripts and all coins they permit to mine ?

For example, switching between the pools of each multipool based on the most cost-effective cryptocurrencie, depending on the pools activated.
From the onion miner on zpool, then from ETH on minningpooolhub 10 minutes later, or zcash (equihash) on nicehash the next 10 minutes depending on their level of profitability ?

CryptAtomeTrader44

full member

Activity: 340

Merit: 103

It is easier to break an atom than partialities AE

I have not managed to connect via the local network in ssh with putty (from a W7 x64) on my rig in nvOC019. No error message. Putty emits a sound of error and closes immediately!
Did I forget something in addition to REMOTE and SSH = YES?
I ping it with no pb but ssh seems to be unactivated :-( How can i verify if SSH server is launched ?

Do you have any idea why my putty (or rig) immediately refuses the ssh connection frommy W7x64 PC ?

azsh07

newbie

Activity: 10

Merit: 0

hi,

sorry for noob question. First I would like to thank you to fullzero for your effort.

My question is, how can i log in via ssh to my headless rig (12x P106) since I tried to using putty but it says connection is refused? I also reformat usb key and result are still same. I tried ping rig address but destination host unreachable. However in my router detect the ip address of this rig.

Can I connect monitor using mainboard DVI (Asrock H110 dont have VGA) for check status locally?

Many thanks. Please forgive my poor language.

VoskCoin

sr. member

Activity: 1414

Merit: 487

YouTube.com/VoskCoin

Quote from: VoskCoin on September 19, 2017, 08:25:00 AM

So I had a three card riser-less build (well actually one was on a riser since the AORUS are huge, no issues with this setup)

1080 ti / 1080 ti / 1070 mini

I moved to a 6 card riser build and now I have this issue . . any ideas?
(1070 mini 1070 mini 1080 TI 1080 TI 1070 mini 1070 mini

software or hardware and then what?

This should be a 3k average sols rig

Added more details below ////

2x 1080 TI + 1070 earning the same as 2x 1080 TI 4x 1070

(1.8k sol rig avg)

3k sol rig, avg 2.6k due to error, but its as if all 3 of the 1070s on the second PSU (connected add2psu PSU1 to mobo/cpu/all risers +pata to Add2psu / PSU2 mobo to add2psu+3x 1070 mini

nvOC_19 on both . .

@crazydane On average I get the best daily returns from suprnova/luck/zhash roughly maybe in that order? if my miner finds a block on luck it obviously comes out ahead. I'm using luck because it's easy to look at basic stats individually

Also in other news . . that's actually good news. . my bitmain antminer d3 has arrived and is hashing away 2 gh/s higher than advertised from 15.5 to 17.5 iirc

When is antOC_19 coming out fullzero? xD

Topic: [ mining os ] nvoc - page 222. (Read 418546 times)