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

[infowire]

Still getting the same arrows with doing default xorg. Don't seem to do anything as far as crashes go.

Code:

ERROR: Error assigning value 75 to attribute 'GPUTargetFanSpeed' (m1-desktop:0[fan:10]) as specified in assignment
       '[fan:10]/GPUTargetFanSpeed=75' (Unknown Error).


ERROR: Error assigning value 75 to attribute 'GPUTargetFanSpeed' (m1-desktop:0[fan:11]) as specified in assignment
       '[fan:11]/GPUTargetFanSpeed=75' (Unknown Error).

Is it ok if i leave the setting like this ? I tried my cards by the type of card, they seems to be alright 4 hours. Now im going to see if they play nice with each other.

Code:

 if [ $NUM_GPU_BLW_THRSHLD -gt 12 ]

EDIT:
All the cards temperature control hang again. Blah.

That error is from tempcontrol not watcdog,

Its weird, I had this error once when I added some cards to a rig, but restoring default xorg solved it.
May be try to re-install the image

Edit:
Are you using salfter switcher?

no i am not.


I got some other error after all night running good

Code:

Sun May 13 00:30:08 MST 2018 - Lost GPU so restarting system. Found GPU's:
Unable to determine the device handle for GPU 0000:08:00.0: GPU is lost.  Reboot the system to recover this GPU

but does not reboot since i don't have a reboot file or maybe it freezes.


EDIT: the os freezes and restarts the gui kinda slow... is that a sign of a bad card ?

I think the ssd went bad. Doesn't seem to write in hdd raw copy... even tho windows drive management says its healthy

[1715382095]

1715382095

[chem1990]

That lost temperature is almost evrytime too much OC on GPU. try lowering values. If persists than change riser...

[urnzwy]

Well after much frustration with my rigs crashing, I thought I had it figured out but the problem continued.

Originally I had setup SMOS but really didn't care for the simplicity of the interface or should I say lack there of and found nvOC.

As my rigs are remote in another state, it was even more difficult to diagnose.

For ease, I had my brother-in-law load up SMOS on both rigs and they have been running for over 4 days with under-volting (100W 1070 & 1070 Ti / 175W 1080 & 1080 Ti) and overclocking (-50 core 1070/1070 Ti, 200 core 1080 & 1080 Ti with 1100 Mem for the 1070/1070 Ti and 1000 for the 1080 & 1080 Ti). Mining ETH.

Whatever my issue was, it was an issue within nvOC. The question is why and what was causing the crashing issue.

Later this week, I plan on loading up a fresh copy of nvOC and see if the issue continues. Unfortunately I cleared the USB sticks we were using that had the nvOC OS on them. Had I though about it, it would have been good to have them to see if anyone could identify any issues. The only thing I really did was re-compile the miners.

It is difficult to pin-point when the issue really showed up as we were having to re-set the miners at least once a week and then eventually even would crash once daily. Both at the same time even on different algorithms. I tried multiple different mining programs, no over clocks, half overclock, full overclock. Nothing seemed to matter. 18-24 hours the crash was inevitable.

I wish I had more information on this, but wanted to get it into the forum in-case this ever happened to anyone else. I will update once I re-load up nvOC and see what happens.

[infowire]

I don't know if they are related errors:

Code:

Calymore log:
15:01:25:998	f1bb9700	NVML: cannot get current temperature, error 15
15:01:25:998	f1bb9700	NVML: cannot get fan speed, error 15

Watchdog:
Mon May 14 15:01:23 MST 2018 - Lost GPU so restarting system. Found GPU's:
Unable to determine the device handle for GPU 0000:0A:00.0: GPU is lost.  Reboot the system to recover this GPU

MINER_TEMP_CONTROL

ERROR: Error assigning value 50 to attribute 'GPUTargetFanSpeed'
       (m1-desktop:0[fan:9]) as specified in assignment
       '[fan:9]/GPUTargetFanSpeed=50' (Unknown Error).


ERROR: Error assigning value 50 to attribute 'GPUTargetFanSpeed'
       (m1-desktop:0[fan:10]) as specified in assignment
       '[fan:10]/GPUTargetFanSpeed=50' (Unknown Error).

I did the default xorg. Nothing.

My 9th and 10th card's do not have a fan speed control in the nvidia panel.
Also only x0 - x8 screens too.
 
https://preview.ibb.co/kKZaGJ/1.png

This is how it's supposed to be. All the cards 0-8 have a fan speed control bar.
https://preview.ibb.co/cNwBqd/2.png




I have been messing with it for 2 days, installed the OS again... nothing

[TAKCuCT]

I don't know if they are related errors:


I have been messing with it for 2 days, installed the OS again... nothing

Did you try to change drivers? Try from 384.

[infowire]

I don't know if they are related errors:


I have been messing with it for 2 days, installed the OS again... nothing

Did you try to change drivers? Try from 384.

No i haven't, current drivers are 387.34. So i should go back? All my rigs are 387.34, but this one have different cards in it.  
Just download the older drivers and run the install ? or do i have to do something else ? Anyone have upgraded 390.48? https://www.geforce.com/drivers/results/132530

------------------------------------- EDIT

So it was one card that caused the NVML Error "NVML: cannot get fan speed, error 15" i removed it and everything seems fine. I plugged in the culprit card on windows and its fine.... for now. Yes i tried switching risers before that.

 So to summarize my ordeal.

• Bad SSD - Slow running miner.
• A card that gave me an NVML error. EVGA 1060 Don't know if it was driver's fault or what.

[chem1990]

hello
I see no activity lately here, did something happen? another thread or project at the end?

thanks

[papampi]

hello
I see no activity lately here, did something happen? another thread or project at the end?

thanks

Just that nvOC working smooth and there are not much users with problems ...

[papampi]

nvOC dev team are working hard on next release with so many improvements and new features.
You can have a look at nvOC GitHub

Suggestions for improvement and new features are most welcome

[Tigel]

nvOC dev team are working hard on next release with so many improvements and new features.
You can have a look at nvOC GitHub

Suggestions for improvement and new features are most welcome

Fist of all, quite impressive, haven't checked for a while as things have been running quite well with 2.0. But:
Algorithm and card specific OC settings maybe? You know, for us who use rigs with mixed cards.

--
Tigel

[papampi]

nvOC dev team are working hard on next release with so many improvements and new features.
You can have a look at nvOC GitHub

Suggestions for improvement and new features are most welcome

Fist of all, quite impressive, haven't checked for a while as things have been running quite well with 2.0. But:
Algorithm and card specific OC settings maybe? You know, for us who use rigs with mixed cards.

--
Tigel

Global+GPU_OFFSET  and Algo+GPU_OFFSET are available for next version (Thanks to Lukepicci):

Code:

POWERLIMIT_MODE="GLOBAL"    # GLOBAL, GPU_SPECIFIC, ALGO_SPECIFIC (Pro Settings: GLOBAL_with_GPU_OFFSET, ALGO_SPECIFIC_with_GPU_OFFSET)
 	 
OVERCLOCK_MODE="GLOBAL"     # NO, GLOBAL, GPU_SPECIFIC, ALGO_SPECIFIC (Pro Settings: GLOBAL_with_GPU_OFFSET, ALGO_SPECIFIC_with_GPU_OFFSET)
 	 
# Modes:  
# NO disables overclock
# GLOBAL mode means no individual gpu- or algo-specific settings are used
# GPU_SPECIFIC let global setting be overridden by individual gpu-specific setting
# ALGO_SPECIFIC let global setting be dynamically overridden by algo-specific setting
# Pro Settings are not for the faint-hearted. Dont use if you dont know what they are. 

[Tigel]

Global+GPU_OFFSET  and Algo+GPU_OFFSET are available for next version (Thanks to Lukepicci):

Great, now I just need find time to test it!

--
Tigel

[papampi]

Global+GPU_OFFSET  and Algo+GPU_OFFSET are available for next version (Thanks to Lukepicci):

Great, now I just need find time to test it!

--
Tigel

19.21 is work in progress and not ready for beta testing yet ...

[sizzlephizzle]

But it's getting close! Great collaboration and collaborators.

Lots of improvements and a leaner crisper codebase.
Siz

[sizzlephizzle]

Have we tested funakoshi miner?

https://github.com/funakoshi2718/funakoshi-miner

[papampi]

Have we tested funakoshi miner?

https://github.com/funakoshi2718/funakoshi-miner

Good catch, let us know how it goes

[papampi]

Have we tested funakoshi miner?

https://github.com/funakoshi2718/funakoshi-miner

Good catch, let us know how it goes

On my test rig with 2x1060 + 1x1070 hash rate is a bit lower than dstm with same OC +PL

dstm:

Code:

2018-05-23 10:35:21 PM|   GPU0  64C  45% |  412.9 Sol/s   413.8 Avg   222.2 I/s | 4.18 S/W   98 W |  1.10  100  252
2018-05-23 10:35:21 PM|   GPU1  59C  45% |  295.1 Sol/s   290.7 Avg   155.8 I/s | 3.75 S/W   75 W |  0.79  93.3 277
2018-05-23 10:35:21 PM|   GPU2  66C  45% |  280.9 Sol/s   281.6 Avg   150.8 I/s | 3.62 S/W   79 W |  0.65  100  253
2018-05-23 10:35:21 PM|   ============== |  988.9 Sol/s   986.1 Avg   528.8 I/s | 3.85 S/W  254 W |  2.54  97.9 260

funakoshiMiner:

Code:

[2018-May-23 22:38:47] [GPU-0 (15 sec)] 212 I/s, 409 Sols/s
[2018-May-23 22:38:47] [GPU-1 (15 sec)] 145 I/s, 278 Sols/s
[2018-May-23 22:38:47] [GPU-2 (15 sec)] 142 I/s, 270 Sols/s
[2018-May-23 22:38:47] [TOTAL (15 sec)] 499 I/s, 957 Sols/s, 1 Accepted, 0 Rejected

[Stubo]

Have we tested funakoshi miner?

https://github.com/funakoshi2718/funakoshi-miner

Good catch, let us know how it goes

On my test rig with 2x1060 + 1x1070 hash rate is a bit lower than dstm with same OC +PL

dstm:

Code:

2018-05-23 10:35:21 PM|   GPU0  64C  45% |  412.9 Sol/s   413.8 Avg   222.2 I/s | 4.18 S/W   98 W |  1.10  100  252
2018-05-23 10:35:21 PM|   GPU1  59C  45% |  295.1 Sol/s   290.7 Avg   155.8 I/s | 3.75 S/W   75 W |  0.79  93.3 277
2018-05-23 10:35:21 PM|   GPU2  66C  45% |  280.9 Sol/s   281.6 Avg   150.8 I/s | 3.62 S/W   79 W |  0.65  100  253
2018-05-23 10:35:21 PM|   ============== |  988.9 Sol/s   986.1 Avg   528.8 I/s | 3.85 S/W  254 W |  2.54  97.9 260

funakoshiMiner:

Code:

[2018-May-23 22:38:47] [GPU-0 (15 sec)] 212 I/s, 409 Sols/s
[2018-May-23 22:38:47] [GPU-1 (15 sec)] 145 I/s, 278 Sols/s
[2018-May-23 22:38:47] [GPU-2 (15 sec)] 142 I/s, 270 Sols/s
[2018-May-23 22:38:47] [TOTAL (15 sec)] 499 I/s, 957 Sols/s, 1 Accepted, 0 Rejected

Thanks, Pap. How many times have the seen the claims of "much faster than others" with these miners only to find otherwise?

[urnzwy]

Well after much frustration with my rigs crashing, I thought I had it figured out but the problem continued.

Originally I had setup SMOS but really didn't care for the simplicity of the interface or should I say lack there of and found nvOC.

As my rigs are remote in another state, it was even more difficult to diagnose.

For ease, I had my brother-in-law load up SMOS on both rigs and they have been running for over 4 days with under-volting (100W 1070 & 1070 Ti / 175W 1080 & 1080 Ti) and overclocking (-50 core 1070/1070 Ti, 200 core 1080 & 1080 Ti with 1100 Mem for the 1070/1070 Ti and 1000 for the 1080 & 1080 Ti). Mining ETH.

Whatever my issue was, it was an issue within nvOC. The question is why and what was causing the crashing issue.

Later this week, I plan on loading up a fresh copy of nvOC and see if the issue continues. Unfortunately I cleared the USB sticks we were using that had the nvOC OS on them. Had I though about it, it would have been good to have them to see if anyone could identify any issues. The only thing I really did was re-compile the miners.

It is difficult to pin-point when the issue really showed up as we were having to re-set the miners at least once a week and then eventually even would crash once daily. Both at the same time even on different algorithms. I tried multiple different mining programs, no over clocks, half overclock, full overclock. Nothing seemed to matter. 18-24 hours the crash was inevitable.

I wish I had more information on this, but wanted to get it into the forum in-case this ever happened to anyone else. I will update once I re-load up nvOC and see what happens.

OK...

So update time. It turns out that the issue all along was a faulty graphics card. In moving to SMOS, I had left the mining screen up on my computer and just happen to walk into an error on GPU 12. The rig crashed, and it took the other rig out as well. We replaced the riser and it crashed again. SO we removed the card completely.

It's been over 4 days running constantly for BOTH rigs and not a single crash or issue..

With this. We are moving back to a fresh copy of nvOC tomorrow as I really do dislike SMOS. But I wanted to update all. So, check your GPU's if you have a crashing issue you can't figure out! We are going to try and replace all other parts and re-insert the card and hope it's not an actual issue with the card. But I have a feeling it is. So it may be going back to TRY and get a replacement. I guess we will see.

[Stubo]

Well after much frustration with my rigs crashing, I thought I had it figured out but the problem continued.

Originally I had setup SMOS but really didn't care for the simplicity of the interface or should I say lack there of and found nvOC.

As my rigs are remote in another state, it was even more difficult to diagnose.

For ease, I had my brother-in-law load up SMOS on both rigs and they have been running for over 4 days with under-volting (100W 1070 & 1070 Ti / 175W 1080 & 1080 Ti) and overclocking (-50 core 1070/1070 Ti, 200 core 1080 & 1080 Ti with 1100 Mem for the 1070/1070 Ti and 1000 for the 1080 & 1080 Ti). Mining ETH.

Whatever my issue was, it was an issue within nvOC. The question is why and what was causing the crashing issue.

Later this week, I plan on loading up a fresh copy of nvOC and see if the issue continues. Unfortunately I cleared the USB sticks we were using that had the nvOC OS on them. Had I though about it, it would have been good to have them to see if anyone could identify any issues. The only thing I really did was re-compile the miners.

It is difficult to pin-point when the issue really showed up as we were having to re-set the miners at least once a week and then eventually even would crash once daily. Both at the same time even on different algorithms. I tried multiple different mining programs, no over clocks, half overclock, full overclock. Nothing seemed to matter. 18-24 hours the crash was inevitable.

I wish I had more information on this, but wanted to get it into the forum in-case this ever happened to anyone else. I will update once I re-load up nvOC and see what happens.

OK...

So update time. It turns out that the issue all along was a faulty graphics card. In moving to SMOS, I had left the mining screen up on my computer and just happen to walk into an error on GPU 12. The rig crashed, and it took the other rig out as well. We replaced the riser and it crashed again. SO we removed the card completely.

It's been over 4 days running constantly for BOTH rigs and not a single crash or issue..

With this. We are moving back to a fresh copy of nvOC tomorrow as I really do dislike SMOS. But I wanted to update all. So, check your GPU's if you have a crashing issue you can't figure out! We are going to try and replace all other parts and re-insert the card and hope it's not an actual issue with the card. But I have a feeling it is. So it may be going back to TRY and get a replacement. I guess we will see.

Thanks for the feedback. I am glad you finally figured it out. Hopefully it will be smooth sailing for a long time with nvOC and healthy HW.