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

[salfter]

I think I've figured out why my profitability switcher was having problems with some coins.  I noticed it when I tried switching my rig from Genoil back to Claymore (I'm having trouble getting Genoil to run for more than 24 hours without crashing)...the script would run, and I'd end up with two or more instances of the Claymore miner running.

pgrep and pkill, when called with the -f option (to look at the whole command line), need to have "+" escaped for some reason...most likely the pattern is processed as a regular expression and not a simple string.  Any miner command string that included "stratum+tcp" (that's basically all of them except the equihash and Genoil daggerhashimoto miners) thus wasn't searchable or killable.

I've committed a fix that should take care of that.  It also prepends the search string with "^" to match on the miner process itself, not the screen process that spawned it.  (Killing the miner will automatically kill the parent screen process.)

[1714092859]

1714092859

[1714092859]

1714092859

[1714092859]

1714092859

[1714092859]

1714092859

[1714092859]

1714092859

[mensa84]

How many NVIDIA GPUs does nvOC support on 1 mainboard?

[TenaciousJ]

Thanks for links.   

You bet..

If you get a second, i posted a while back about an issue I'm having in 0017 where one card runs at 66% capacity while the others are at 100%.  I'm just wondering if that's something you've seen before, and if there might be a straight forward solution to it.  If not, I'll probably just wait until I swap boards in a few days to mess with it since otherwise it's running solid as a rock with all 6 gpus.

Is the hashrate lower with the GPU running at 66%?

Hey FZ - just wanted to let you know I resolved the issue with my underpowered GPU... turns out having 2 risers on one SATA power line is frowned upon... once I put all the GPUs on their own lines the system is back up at 100%

With 1080 or 1080ti that is best; especially if they are AM models.  I haven't heard of anyone needing to do this with 1070s or 1060s; what type of GPUs are you using?

I've got:
2x asus dual 1070
3x EVGA superclock 1070
1x Nvidia 1070 FE in the x16 slot

Asrock z270 Killer SLI/AC mobo (no external power to mobo for pcie (no option for it))
dual 750w PSUs - 1x evga and 1x thermaltake
64gb USB 3 stick

No 1080s in this rig atm, though I had a 1080 ti in there yesterday in the x16 slot so no riser but the problems I've had arose with all 1070s

Before, I had it setup so that:
2 risers were powered on one molex power line
2 riser on 1 SATA lines
3x 1070 each on its own power line (450w); and
mobo/cpu all on the EVGA 750w psu; [all of that should be well within the acceptable range of draw from the PSU]

Then the other 3 1070 GPUs and 1 riser on molex were powered by the 750w thermaltake [again, well within the total power load taking overhead into account]

New setup splits the risers between the PSUs 2 go to separate lines on the EVGA, the other 3 on separate lines on the thermaltake, all on SATA power cables except 1 (though that shouldn't make a difference), and the risers are powered by the same PSU as the GPU main power is coming from whereas before they weren't.
the mobo/cpu are still on the primary EVGA PSU, and the thermaltake is connected by a Y splitter ATX cable.

The result is that the more even power load resolved the problem, though I can't be absolutely certain it was solved by splitting the risers to different lines exclusively.

[fullzero]

How many NVIDIA GPUs does nvOC support on 1 mainboard?

In theory 14x; I may need to update some system files the new Biostar and ASRock monster mobos.

see:

https://bitcointalk.org/index.php?topic=1854250.msg20018178#msg20018178


I have received a shipping notification for the Biostar; but I still need one of the ASRock H110 Pro BTC+ to test / ensure support:

https://www.newegg.com/Product/Product.aspx?Item=N82E16813157781

If a member has one arriving early this week (before Wednesday preferably):

I am offering double the newegg price for one + $20 for USPS Priority Mail Flatrate Box shipping (with insurance) to me:

$285 in BTC

Unless your Trust exceeds mine; I will pay upon receipt: I am also willing to use most trusted forum members as escrow if desired.

[fullzero]

Hi!

Are you already discussed about "ERROR: Looks like GPU0 (or 1 or any) are stopped. Restart attempt" ?
I have this error on one of my 1050Ti cards in about 10 minutes after start Zcash mining.
No overclocking. Just as it is. And after card error all system hangs up deadly if I'm trying to close terminal window. If don't - mining continues but without one card.
And this rig works under win10 for days and weeks w/o problems!

So.. how to fix it?

What are is your powerlimit / mobo / CPU / all GPUs / what COIN are you mining / what mining client are you using?

All default. Didn't touch anything. Powerlimit 100, all GPUs, Zcash on EWBF 0.3.4b from github.

not sure what you mean by this:

EWBF 0.3.4b from github.

EWBF releases his client via google drive and MEGA links.

Did you for some reason change the EWBF clients on nvOC?


A 1050ti has a max powerlimit of 75, setting a powerlimit of 100; sets no powerlimit as the cmd is invalid.

What mobo / CPU / all 1050tis?


If 3.4 is giving you problems you can try 3.3 by changing:

Code:

EWBF_VERSION="3_4"

to

Code:

EWBF_VERSION="3_3"

How many GPUs are you using?

Are you using risers, if so what kind are they, and how are you powering them?

If you are powering 1050tis using risers without using 6 pin pcie direct to the risers; they will never be stable.

[fullzero]

I think I've figured out why my profitability switcher was having problems with some coins.  I noticed it when I tried switching my rig from Genoil back to Claymore (I'm having trouble getting Genoil to run for more than 24 hours without crashing)...the script would run, and I'd end up with two or more instances of the Claymore miner running.

pgrep and pkill, when called with the -f option (to look at the whole command line), need to have "+" escaped for some reason...most likely the pattern is processed as a regular expression and not a simple string.  Any miner command string that included "stratum+tcp" (that's basically all of them except the equihash and Genoil daggerhashimoto miners) thus wasn't searchable or killable.

I've committed a fix that should take care of that.  It also prepends the search string with "^" to match on the miner process itself, not the screen process that spawned it.  (Killing the miner will automatically kill the parent screen process.)

Please test and ensure with confidence; that it is rock solid: then I will integrate these changes. 

It would be helpful if you did so with the integrated oneBash + switch version I made (swap the relevant part of switch with your updated code); so I can be sure it works as intended.

[fullzero]

Thanks for links.    

You bet..

If you get a second, i posted a while back about an issue I'm having in 0017 where one card runs at 66% capacity while the others are at 100%.  I'm just wondering if that's something you've seen before, and if there might be a straight forward solution to it.  If not, I'll probably just wait until I swap boards in a few days to mess with it since otherwise it's running solid as a rock with all 6 gpus.

Is the hashrate lower with the GPU running at 66%?

Hey FZ - just wanted to let you know I resolved the issue with my underpowered GPU... turns out having 2 risers on one SATA power line is frowned upon... once I put all the GPUs on their own lines the system is back up at 100%

With 1080 or 1080ti that is best; especially if they are AM models.  I haven't heard of anyone needing to do this with 1070s or 1060s; what type of GPUs are you using?

I've got:
2x asus dual 1070
3x EVGA superclock 1070
1x Nvidia 1070 FE in the x16 slot

Asrock z270 Killer SLI/AC mobo (no external power to mobo for pcie (no option for it))
dual 750w PSUs - 1x evga and 1x thermaltake
64gb USB 3 stick

No 1080s in this rig atm, though I had a 1080 ti in there yesterday in the x16 slot so no riser but the problems I've had arose with all 1070s

Before, I had it setup so that:
2 risers were powered on one molex power line
2 riser on 1 SATA lines
3x 1070 each on its own power line (450w); and
mobo/cpu all on the EVGA 750w psu; [all of that should be well within the acceptable range of draw from the PSU]

Then the other 3 1070 GPUs and 1 riser on molex were powered by the 750w thermaltake [again, well within the total power load taking overhead into account]

New setup splits the risers between the PSUs 2 go to separate lines on the EVGA, the other 3 on separate lines on the thermaltake, all on SATA power cables except 1 (though that shouldn't make a difference), and the risers are powered by the same PSU as the GPU main power is coming from whereas before they weren't.
the mobo/cpu are still on the primary EVGA PSU, and the thermaltake is connected by a Y splitter ATX cable.

The result is that the more even power load resolved the problem, though I can't be absolutely certain it was solved by splitting the risers to different lines exclusively.

Which driver version are you using?

Also do you know which GPU model was the one that was insufficiently powered?

[lbrasi]

I think I've figured out why my profitability switcher was having problems with some coins.  I noticed it when I tried switching my rig from Genoil back to Claymore (I'm having trouble getting Genoil to run for more than 24 hours without crashing)...the script would run, and I'd end up with two or more instances of the Claymore miner running.

pgrep and pkill, when called with the -f option (to look at the whole command line), need to have "+" escaped for some reason...most likely the pattern is processed as a regular expression and not a simple string.  Any miner command string that included "stratum+tcp" (that's basically all of them except the equihash and Genoil daggerhashimoto miners) thus wasn't searchable or killable.

I've committed a fix that should take care of that.  It also prepends the search string with "^" to match on the miner process itself, not the screen process that spawned it.  (Killing the miner will automatically kill the parent screen process.)

Please test and ensure with confidence; that it is rock solid: then I will integrate these changes. 

It would be helpful if you did so with the integrated oneBash + switch version I made (swap the relevant part of switch with your updated code); so I can be sure it works as intended.

I will be glad to test, just not sure what needs to be changed

[pixelizedchaos]

Verge is looking really good, is there any way we can add it to the list? https://bitcointalk.org/index.php?topic=1365894.0

[TenaciousJ]

Thanks for links.    

You bet..

If you get a second, i posted a while back about an issue I'm having in 0017 where one card runs at 66% capacity while the others are at 100%.  I'm just wondering if that's something you've seen before, and if there might be a straight forward solution to it.  If not, I'll probably just wait until I swap boards in a few days to mess with it since otherwise it's running solid as a rock with all 6 gpus.

Is the hashrate lower with the GPU running at 66%?

Hey FZ - just wanted to let you know I resolved the issue with my underpowered GPU... turns out having 2 risers on one SATA power line is frowned upon... once I put all the GPUs on their own lines the system is back up at 100%

With 1080 or 1080ti that is best; especially if they are AM models.  I haven't heard of anyone needing to do this with 1070s or 1060s; what type of GPUs are you using?

I've got:
2x asus dual 1070
3x EVGA superclock 1070
1x Nvidia 1070 FE in the x16 slot

Asrock z270 Killer SLI/AC mobo (no external power to mobo for pcie (no option for it))
dual 750w PSUs - 1x evga and 1x thermaltake
64gb USB 3 stick

No 1080s in this rig atm, though I had a 1080 ti in there yesterday in the x16 slot so no riser but the problems I've had arose with all 1070s

Before, I had it setup so that:
2 risers were powered on one molex power line
2 riser on 1 SATA lines
3x 1070 each on its own power line (450w); and
mobo/cpu all on the EVGA 750w psu; [all of that should be well within the acceptable range of draw from the PSU]

Then the other 3 1070 GPUs and 1 riser on molex were powered by the 750w thermaltake [again, well within the total power load taking overhead into account]

New setup splits the risers between the PSUs 2 go to separate lines on the EVGA, the other 3 on separate lines on the thermaltake, all on SATA power cables except 1 (though that shouldn't make a difference), and the risers are powered by the same PSU as the GPU main power is coming from whereas before they weren't.
the mobo/cpu are still on the primary EVGA PSU, and the thermaltake is connected by a Y splitter ATX cable.

The result is that the more even power load resolved the problem, though I can't be absolutely certain it was solved by splitting the risers to different lines exclusively.

Which driver version are you using?

Also do you know which GPU model was the one that was insufficiently powered?

I think driver 381.22 is installed right now.  The cards that were underpowered were EVGA superclock 1070s, I had two of them on one SATA power cable and neither one worked properly that way.  Once their risers were on their own lines they work great.  Now I'm not seeing any dips in power - all cards are 100% and pulling about 3.25 sol / watt.

[osnwt]

How tricky would it be to use the nvidia-settings commands from within oneBash to run the following command recursively for each card that's detected when it runs, rather than setting a specific power limit or clock offset?

Enable PowerMizer (Prefer Maximum Performance)

nvidia-settings -a '[gpu:0]/GPUPowerMizerMode=1'


essentially this, but with more efficient code because I never was good at foreach loops and such...

if [ $POWERLIMIT == "NO" ]
then
sudo nvidia-settings -a '[gpu:0]/GPUPowerMizerMode=1'
sudo nvidia-settings -a '[gpu:1]/GPUPowerMizerMode=1'
sudo nvidia-settings -a '[gpu:2]/GPUPowerMizerMode=1'
sudo nvidia-settings -a '[gpu:3]/GPUPowerMizerMode=1'
sudo nvidia-settings -a '[gpu:4]/GPUPowerMizerMode=1'
sudo nvidia-settings -a '[gpu:5]/GPUPowerMizerMode=1'
fi

I was going to rewrite exactly these parts of script since I also don't like such code. But before doing so I asked (in another rxOC thread) to put oneBash files to GitHub repository, so such changes will not be like a monkey work to copy/paste a lot. Until that I feel no motivation since I have only couple of rigs with 1050 running nvOC. It might change when I receive 12 PCIe MOBO (waiting for Biostar board from Aliexpress). But I am pretty sure that git is the best way to handle a lot of change requests and integrate them into this great distro or run just own branch.

PS. I was talking about putting oneBash only to a repository, not about full distribution.

[flminer]

Just tried nvOC v0017 on an ASROCK X370 Gaming K4 motherboard and it does not recognize the hardware.

I don't have a x370 mobo and haven't added support for the chipset. 

You can do this by:

Click Ubuntu button on top left and type:

u

Click on software updater

Install updates


If you can't get ubuntu to boot:

If the bios posts; you can access the grub loader menu by pressing

esc

continuously while booting (note holding it down doesn't usually work), then select boot in recovery mode. 

in recovery mode:

Enable networking

then install updates from the cmd prompt:

Code:

sudo apt-get update && sudo apt-get dist-upgrade --yes

and reboot

this should ensure your build has all known system files for your system.

Will give that a shot tonight. Appreciate it! 

[tempgoga]

Hey guys, any clue what might account for the difference in PCIE utilization? asking because GPU6 is my least stable card that requires the lowest clocks of the bunch and evidently it also has the lowest PCIE utilization.
https://i.imgur.com/pwxDFAQ.png

[Maxximus007]

Have been testing the new Claymore 9.7 on a 6 x 1070 rig. It is absolutely faster than before, but have had hangups on a regular basis. Lowering MC helps, but the avg speed will be a bit lower than Genoil (running latest from github). You have to calculate it a bit, since the dual mining (DCRI 20) will give some additional value, but you'll lose 2% because of dev fee.  
The main advantage of Claymore is that it will restart itself, for Genoil you need to use a restart script.
To me it seems Genoil is the better choice for now.

[tempgoga]

Have been testing the new Claymore 9.7 on a 6 x 1070 rig. It is absolutely faster than before, but have had hangups on a regular basis. Lowering MC helps, but the avg speed will be a bit lower than Genoil (running latest from github). You have to calculate it a bit, since the dual mining (DCRI 20) will give some additional value, but you'll lose 2% because of dev fee.  
The main advantage of Claymore is that it will restart itself, for Genoil you need to use a restart script.
To me it seems Genoil is the better choice for now.

7 x 1070 rig here, i've found that if my clocks are even slightly too high it doesn't matter which miner i'm using the rig just hangs forever unless i restart it remotely via SSH, i'm using dcri 40 and i think its the best setting right now, its about 31.5Mh/s for ETH and 410 Mh/s for SIA, on claymore 9.5 it was more like 29.8/390..

Whats your power limit and clocks if you dont mind me asking? i'm on 125pl/0/1400.

[bobers]

please help. How can i add dcri argument to my onebash settings? what and where exactly i should whrite?

[tempgoga]

please help. How can i add dcri argument to my onebash settings? what and where exactly i should whrite?

The dcri value is only for dual mining, if you're only mining Ethereum theres no point in using dcri, theres no point in even using claymore miner, genoil miner is just as fast but with no fee for strictly ETH mining.

if you are dual mining just search for "dcri" in oneBash, the default setting is 70, i recommend changing it to 40.

[Maxximus007]

please help. How can i add dcri argument to my onebash settings? what and where exactly i should whrite?

It's not set by default in oneBash So you can set it in $ETH_EXTENTION_ARGUMENTS. Use 40 or even 20 (depending your cards etc).

Code:

ETH_EXTENTION_ARGUMENTS=""    # add any additional claymore arguments desired here

So:

Code:

ETH_EXTENTION_ARGUMENTS="-dcri 40"    # add any additional claymore arguments desired here

[Maxximus007]

Have been testing the new Claymore 9.7 on a 6 x 1070 rig. It is absolutely faster than before, but have had hangups on a regular basis. Lowering MC helps, but the avg speed will be a bit lower than Genoil (running latest from github). You have to calculate it a bit, since the dual mining (DCRI 20) will give some additional value, but you'll lose 2% because of dev fee.  
The main advantage of Claymore is that it will restart itself, for Genoil you need to use a restart script.
To me it seems Genoil is the better choice for now.

7 x 1070 rig here, i've found that if my clocks are even slightly too high it doesn't matter which miner i'm using the rig just hangs forever unless i restart it remotely via SSH, i'm using dcri 40 and i think its the best setting right now, its about 31.5Mh/s for ETH and 410 Mh/s for SIA, on claymore 9.5 it was , more like 29.8/390..

Whats your power limit and clocks if you dont mind me asking? i'm on 125pl/0/1400.

These are pretty good numbers, Here on 115W, dcri 20, Claymore 9.7 30Mh/s and 300Mh/s SIA. Not very stable, have to lower my OC more. Stable with Genoil 0.11 (MC clock 200 down from Claymore) and gives me 31.5 Mh/s

[tempgoga]

Have been testing the new Claymore 9.7 on a 6 x 1070 rig. It is absolutely faster than before, but have had hangups on a regular basis. Lowering MC helps, but the avg speed will be a bit lower than Genoil (running latest from github). You have to calculate it a bit, since the dual mining (DCRI 20) will give some additional value, but you'll lose 2% because of dev fee.  
The main advantage of Claymore is that it will restart itself, for Genoil you need to use a restart script.
To me it seems Genoil is the better choice for now.

7 x 1070 rig here, i've found that if my clocks are even slightly too high it doesn't matter which miner i'm using the rig just hangs forever unless i restart it remotely via SSH, i'm using dcri 40 and i think its the best setting right now, its about 31.5Mh/s for ETH and 410 Mh/s for SIA, on claymore 9.5 it was , more like 29.8/390..

Whats your power limit and clocks if you dont mind me asking? i'm on 125pl/0/1400.

These are pretty good numbers, Here on 115W, dcri 20, Claymore 9.7 30Mh/s and 300Mh/s SIA. Not very stable, have to lower my OC more. Stable with Genoil 0.11 (MC clock 200 down from Claymore) and gives me 31.5 Mh/s

i see, i cant dual mine on anything less then 120 pl if i want the memory clocks to be 1000+ and still be stable.