[Support] PTRFRLL - Docker images

[Masamoto]

On 3/10/2021 at 5:05 PM, PTRFRLL said:

 

Unfortunately, that's a limitation of how the Nvidia drivers are integrated in Unraid. Normally, you could use the nvidia-settings CLI but that errors out as Unraid runs headless.

You might try tweaking the memory-tweak flag for T-rex but I'm not sure how much that will help

Sadly Memory Tweak seems to only work on graphics cards with GDDR5 or GDDR5X memory only.

I can't make the edits stick somehow.

I am tempted to run a VM and try overclocking/underclocking there. 

[darkangelus]

Thanks, i wasted two hours last night trying to get a vm passthrough to work for mining and i wish i'd thought of checking docker.

Running fine on my 1650 super 4gb, hashrate of 40 at 80w.

[vomi21]

last night, i installed trex miner and i cant get ist to start. I keep getting some kind of cuda error:

 

--exit-on-cuda-error Forces miner to immediately exit on CUDA error.

 

sorry for not knowing what to do. Its my first try to mine something, because my unraid is running 24/7 anyway.

 

Thanks for help!

[brimnac]

On 2/17/2021 at 5:46 PM, Shao said:

Anyone know how to add: --mt Memory tweak mode?

The config in GUI is way different than the command line options.

 

Also looking for this. @PTRFRLL - is there something simple I'm missing? I keep getting barked at if I put --mt 4 or --mt=4 in the Extra Parameters, and can't figure out where else to make the change.

EDIT: Solution below.

Edited March 19, 2021 by brimnac

[gamer1pc]

Anyone know how I can add this to the config/command line c=LTC,mc=FIRO. I need it in order for the pool to convert my FIRO to LTC. Thanks in advanced for the help 

[brimnac]

@gamer1pc: try this, it got the Memory Tweak I was having issues with to work.

 

I modified the config.json file using Midnight Commander. It originally ended like this:

	"worker" : "unraid"
}

 

"worker" is no longer the last config option, so add a comma to the end and the new switches after.

    "worker" : "unraid",
    "mt" : 4
}

 

I don't know for certain, but I'd try something like this (put your worker information in the <name> variable, if applicable):

	"worker" : "<name>",
	"c" : "LTC",
	"mc" : "FIRO"
}

 

I'm getting some rejected shares, so I'm going to figure that out next. --mt works for me now, though, per my LOG file:

WARN: GPU #0(000100): Zotac GeForce GTX 1080, intensity set to 22, mtweak 3

 

EDIT: I did need to set it as PRIVILEGED in the docker setup page when using the "mt" flag. I was getting errors that it wasn't in Admin mode; that solved the issue.

Edited March 19, 2021 by brimnac

[gamer1pc]

@brimnac Thanks for the reply I'll give it a try

[Riverfrome]

Thanks for the work... It's was up and running together with the other nicehash rig in a matter of minutes.

 

That being said... This unraid server is my media server and I just want to run my Nvidia p2000 at like 50% in the background.... But it's running at 65/75w with about 80°c which I don't want...

 

Did anyone find how to limit the card?

 

I tried nvidia-smi -pl 50 (since my card is 75w)

 

But gets returned "50w is not a valid power limit which should be between 75 and 75w"

 

Any ideas ( sorry guys... I know this might be an easy fix)

 

Thanks

Edited March 18, 2021 by Riverfrome

[PTRFRLL]

9 hours ago, Riverfrome said:

Thanks for the work... It's was up and running together with the other nicehash rig in a matter of minutes.

 

That being said... This unraid server is my media server and I just want to run my Nvidia p2000 at like 50% in the background.... But it's running at 65/75w with about 80°c which I don't want...

 

Did anyone find how to limit the card?

 

I tried nvidia-smi -pl 50 (since my card is 75w)

 

But gets returned "50w is not a valid power limit which should be between 75 and 75w"

 

Any ideas ( sorry guys... I know this might be an easy fix)

 

Thanks

 

I believe you can view the min/max power limits of your card with the following command:

 

nvidia-smi -q -d POWER

 

Otherwise, you might play around with the memory tweak (--mt) setting for T-rex

[jaybee]

On 3/18/2021 at 2:37 PM, PTRFRLL said:

 

I believe you can view the min/max power limits of your card with the following command:

 

nvidia-smi -q -d POWER

 

Otherwise, you might play around with the memory tweak (--mt) setting for T-rex

 

Thanks for the support with this.

 

I've had a pay around with my 3070 and got it passed through to a VM where it does indeed hash at 61mh/s with the memory overclocked. In this docker it sits at 51/52mh/s. I have been successful in adjusting the power limit using nvidia-smi as you proposed above. This worked perfectly and halved my power usage. I was also able to set the core GPU clock also using nvidia-smi with no issues. I did have to set persistance mode on.

 

For the memory however, I can only see a command mentioned as like this:

 

nvidia-smi -i 0 -ac 2000,1000

 

But when I do that I get the following error:

 

Setting applications clocks is not supported for GPU 00000000:06:00.0.
Treating as warning and moving on.
All done.

 

So it seems adjusting memory clocks is not possible? I googled and also saw that if you have "N/A" in the output for "nvidia-smi -q" in a lot of places, this means that your GPU does not support that feature. I can see N/A against the application clocks part as per below output. Maybe this is why the above error occurs. 

 

 

 Clocks
        Graphics                          : 1005 MHz
        SM                                : 1005 MHz
        Memory                            : 6800 MHz
        Video                             : 900 MHz
    Applications Clocks
        Graphics                          : N/A
        Memory                            : N/A
    Default Applications Clocks
        Graphics                          : N/A
        Memory                            : N/A
    Max Clocks
        Graphics                          : 2100 MHz
        SM                                : 2100 MHz
        Memory                            : 7001 MHz
        Video                             : 1950 MHz
    Max Customer Boost Clocks

 

 

I can see mention of using "nvidia-settings" or xconfig or powermizer to adjust things like memory, but I think this is only possible on more feature rich Linux distros like ubuntu possibly. Have you any other ideas how we can overclock memory somehow? This would be a killer feature to have it set and running inside a docket as you say, since then it is not bound to a VM and is accessible as a shared resource for both mining at its full potential, as well as for plex transcoding duties. 

[_cr8tor_]

On 3/13/2021 at 2:53 AM, darkangelus said:

Thanks, i wasted two hours last night trying to get a vm passthrough to work for mining and i wish i'd thought of checking docker.

Running fine on my 1650 super 4gb, hashrate of 40 at 80w.

How?

I have a 1650 also and can not get it working.
Log below.

Note: I only want it working on the 1650. I also cant figure out how to exclude the 660 via unraid either but it seems to bypass it anyway. Or is that part of the problem?

 

 

Starting T-rex miner...
============================================================
Server: stratum2+tcp://lyra2z.usa-west.nicehash.com:3365
Algorithm: lyra2z
Wallet: 3MsjZ9Wr3HbQqXmwisnZcn4VYSfWTkAEYQ
Worker: Mining1
============================================================
20210323 00:51:58 T-Rex NVIDIA GPU miner v0.19.12 - [CUDA v11.10 | Linux]
20210323 00:51:58 r.34753df0aef5
20210323 00:51:58
20210323 00:51:58 NVIDIA Driver v460.67
20210323 00:51:58 CUDA devices available: 2
20210323 00:51:58
20210323 00:51:58 WARN: DevFee 1% (lyra2z)
20210323 00:51:58
20210323 00:51:58 URL : stratum2+tcp://lyra2z.usa-west.nicehash.com:3365
20210323 00:51:58 USER: 3MsjZ9Wr3HbQqXmwisnZcn4VYSfWTkAEYQ
20210323 00:51:58 PASS: x
20210323 00:51:58 WRK : Mining1
20210323 00:51:58
20210323 00:51:58 Starting on: lyra2z.usa-west.nicehash.com:3365
20210323 00:51:58 Authorizing...
20210323 00:51:58 Authorized successfully.
20210323 00:51:58 WARN: New difficulty: 16 (0.06250)
20210323 00:51:58 lyra2z block 596956, diff 2192.868
20210323 00:51:58 ERROR: Can't start miner, GeForce GTX 660 (CC 3) is not supported
20210323 00:51:58 WARN: GPU #0(000200): MSI GeForce GTX 1650, intensity set to 17
20210323 00:51:59 T-Rex finished.
init.sh: line 16: 6 Killed ./t-rex -c /config/config.json -a $ALGO -o $SERVER -u $WALLET -p x -w $WORKER

Edited March 23, 2021 by _cr8tor_

[PTRFRLL]

17 hours ago, _cr8tor_ said:

How?

I have a 1650 also and can not get it working.
Log below.

Note: I only want it working on the 1650. I also cant figure out how to exclude the 660 via unraid either but it seems to bypass it anyway. Or is that part of the problem?

 

 

Starting T-rex miner...
============================================================
Server: stratum2+tcp://lyra2z.usa-west.nicehash.com:3365
Algorithm: lyra2z
Wallet: 3MsjZ9Wr3HbQqXmwisnZcn4VYSfWTkAEYQ
Worker: Mining1
============================================================
20210323 00:51:58 T-Rex NVIDIA GPU miner v0.19.12 - [CUDA v11.10 | Linux]
20210323 00:51:58 r.34753df0aef5
20210323 00:51:58
20210323 00:51:58 NVIDIA Driver v460.67
20210323 00:51:58 CUDA devices available: 2
20210323 00:51:58
20210323 00:51:58 WARN: DevFee 1% (lyra2z)
20210323 00:51:58
20210323 00:51:58 URL : stratum2+tcp://lyra2z.usa-west.nicehash.com:3365
20210323 00:51:58 USER: 3MsjZ9Wr3HbQqXmwisnZcn4VYSfWTkAEYQ
20210323 00:51:58 PASS: x
20210323 00:51:58 WRK : Mining1
20210323 00:51:58
20210323 00:51:58 Starting on: lyra2z.usa-west.nicehash.com:3365
20210323 00:51:58 Authorizing...
20210323 00:51:58 Authorized successfully.
20210323 00:51:58 WARN: New difficulty: 16 (0.06250)
20210323 00:51:58 lyra2z block 596956, diff 2192.868
20210323 00:51:58 ERROR: Can't start miner, GeForce GTX 660 (CC 3) is not supported
20210323 00:51:58 WARN: GPU #0(000200): MSI GeForce GTX 1650, intensity set to 17
20210323 00:51:59 T-Rex finished.
init.sh: line 16: 6 Killed ./t-rex -c /config/config.json -a $ALGO -o $SERVER -u $WALLET -p x -w $WORKER

 

You need to exclude the 660. There's two ways to accomplish that.

1. In the Trex config.json, there's a devices property that you can use to specify which devices to use:

 

{
    ...
    "devices": "0,1",
    ...
}

 

You have two devices (0 and 1), the 660 is #1 so remove the 1 from the devices list:

{
    ...
    "devices": "0",
    ...
}

 

OR

 

2. You can add the NVIDIA_VISIBLE_DEVICES variable on the docker and only enter the UUID of the 1650 you want to use

Edited March 23, 2021 by PTRFRLL

[darkangelus]

On 3/23/2021 at 12:52 AM, _cr8tor_ said:

How?

I have a 1650 also and can not get it working.
Log below.

Note: I only want it working on the 1650. I also cant figure out how to exclude the 660 via unraid either but it seems to bypass it anyway. Or is that part of the problem?

 

 

Starting T-rex miner...
============================================================
Server: stratum2+tcp://lyra2z.usa-west.nicehash.com:3365
Algorithm: lyra2z
Wallet: 3MsjZ9Wr3HbQqXmwisnZcn4VYSfWTkAEYQ
Worker: Mining1
============================================================
20210323 00:51:58 T-Rex NVIDIA GPU miner v0.19.12 - [CUDA v11.10 | Linux]
20210323 00:51:58 r.34753df0aef5
20210323 00:51:58
20210323 00:51:58 NVIDIA Driver v460.67
20210323 00:51:58 CUDA devices available: 2
20210323 00:51:58
20210323 00:51:58 WARN: DevFee 1% (lyra2z)
20210323 00:51:58
20210323 00:51:58 URL : stratum2+tcp://lyra2z.usa-west.nicehash.com:3365
20210323 00:51:58 USER: 3MsjZ9Wr3HbQqXmwisnZcn4VYSfWTkAEYQ
20210323 00:51:58 PASS: x
20210323 00:51:58 WRK : Mining1
20210323 00:51:58
20210323 00:51:58 Starting on: lyra2z.usa-west.nicehash.com:3365
20210323 00:51:58 Authorizing...
20210323 00:51:58 Authorized successfully.
20210323 00:51:58 WARN: New difficulty: 16 (0.06250)
20210323 00:51:58 lyra2z block 596956, diff 2192.868
20210323 00:51:58 ERROR: Can't start miner, GeForce GTX 660 (CC 3) is not supported
20210323 00:51:58 WARN: GPU #0(000200): MSI GeForce GTX 1650, intensity set to 17
20210323 00:51:59 T-Rex finished.
init.sh: line 16: 6 Killed ./t-rex -c /config/config.json -a $ALGO -o $SERVER -u $WALLET -p x -w $WORKER

Hi, to answer your question to me. I realized afterwards that a 4gb card is now unusable for ETH since feburary or so and while it's mining it's not getting any shares at all.

I've switched to ravencoin which is hashing between 11-19mh/s at the moment. I was trying to look up how to overclock/underclock to get a better rate but now i've read the rest of the post i'll gie it a try.

[remati]

The Docker Hub URL https://registry.hub.docker.com/r/ghcr.io/meeb/tubesync seems to 404 in your template.

[PTRFRLL]

36 minutes ago, remati said:

The Docker Hub URL https://registry.hub.docker.com/r/ghcr.io/meeb/tubesync seems to 404 in your template.

 

Thanks for the heads up. I think that's an issue with the CA templates, they assume all Dockers are hosted on docker hub but in this case it's hosted on Github's Container registry

[luisalrp]

Hello,

 

I'm trying to run tubesync behind swag, but I'm getting the following error:

Quote

Bad Request

Contradictory scheme headers

The config file is pretty simple, but there is something I'm missing:

server {
    listen 443 ssl;
    listen [::]:443 ssl;
    server_name ts.*;
    include /config/nginx/ssl.conf;
    client_max_body_size 0;
    # enable for Authelia
    include /config/nginx/authelia-server.conf;
    location / {
        # enable for Authelia
        include /config/nginx/authelia-location.conf;
        include /config/nginx/proxy.conf;
        resolver 127.0.0.11 valid=30s;
        set $upstream_app tubesync;
        set $upstream_port 4848;
        set $upstream_proto http;
        proxy_pass $upstream_proto://$upstream_app:$upstream_port;
    }
}

Any ideas? Thanks!

[horphi]

Hello,

 

does somebody of you have the same problems with ptrfrll/nv-docker-trex?

 

 

Starting T-rex miner...
============================================================
Server: stratum+tcp://eu1.ethermine.org:4444
Algorithm: ethash
Wallet: 0x583fFc59BB730eAcAD88f3cAbF6E5a1958f4D74C
Worker: worker
============================================================
20210408 13:23:54 ERROR: Can't start miner, fan control is not supported on Linux
20210408 13:23:55 T-Rex finished.

 

 

I don't know where to setup the fan control.

MSI 2080ti

 

Best regards,

 

Phil

Edited April 8, 2021 by horphi

[PTRFRLL]

1 hour ago, horphi said:

Hello,

 

does somebody of you have the same problems with ptrfrll/nv-docker-trex?

 

 

Starting T-rex miner...
============================================================
Server: stratum+tcp://eu1.ethermine.org:4444
Algorithm: ethash
Wallet: 0x583fFc59BB730eAcAD88f3cAbF6E5a1958f4D74C
Worker: worker
============================================================
20210408 13:23:54 ERROR: Can't start miner, fan control is not supported on Linux
20210408 13:23:55 T-Rex finished.

 

 

I don't know where to setup the fan control.

MSI 2080ti

 

Best regards,

 

Phil

 

I have not seen that before. You might try raising an issue on the T-Rex Github:

https://github.com/trexminer/T-Rex

[doma_2345]

I have just set this up on a p4000 can anyone explain to me why the current and reported hashrates are so different, is this because of dev and pool fees?

 

[PTRFRLL]

1 hour ago, doma_2345 said:

I have just set this up on a p4000 can anyone explain to me why the current and reported hashrates are so different, is this because of dev and pool fees?

 

 

I believe current hashrate is based of a 24hr period. More here

[Alias_ZA]

Hi, 

 

Can someone help me setup two instances of this in docker?

 

I already have them running but instance 2 is using the config file of instance 1 even though the appdata paths are set correctly. I think something might be hard coded somewhere?

 

Below is the Unraid side - I want it to use the config file in /mnt/user/appdata/trex-miner-hayden/ but instead it seems to be using /mnt/user/appdata/trex-miner/ which is the first instance. 

 

Any ideas?

 

[Skylinar]

On 3/17/2021 at 8:56 PM, brimnac said:

@gamer1pc: try this, it got the Memory Tweak I was having issues with to work.

 

I modified the config.json file using Midnight Commander. It originally ended like this:

	"worker" : "unraid"
}

 

"worker" is no longer the last config option, so add a comma to the end and the new switches after.

    "worker" : "unraid",
    "mt" : 4
}

 

I don't know for certain, but I'd try something like this (put your worker information in the <name> variable, if applicable):

	"worker" : "<name>",
	"c" : "LTC",
	"mc" : "FIRO"
}

 

I'm getting some rejected shares, so I'm going to figure that out next. --mt works for me now, though, per my LOG file:

WARN: GPU #0(000100): Zotac GeForce GTX 1080, intensity set to 22, mtweak 3

 

EDIT: I did need to set it as PRIVILEGED in the docker setup page when using the "mt" flag. I was getting errors that it wasn't in Admin mode; that solved the issue.

 

Thanks for the insights. I'm messing around with this and a P2000 and can't even get it working. 

 

I've added

"mt" : 4

to the config after the "worker" and started the trex docker in PRIVILEGED mode, is there more I've missed? 

 

Quote

20210424 18:42:29 WARN: GPU #0(000c00): NVIDIA Quadro P2000, intensity 22

 

[nbbsn]

I just updated to the latest image of the Trex miner from April 26 and am no longer able to start the container due to a CUDA error.

 

ERROR: Can't start miner, can't initialize CUDA engine, cuda exception in [main, 609], CUDA_ERROR_NO_DEVICE. Is NVIDIA driver installed?

 

Nvidia driver (v455.45.01) has been installed for months ever since unraid 6.9.0 came out and is still working perfectly in jellyfin, emby and previous versions of Trex. 
I rolled back to the previous image from 4 weeks ago and it's working flawlessly again, so something in that latest build screwed up my GTX 1070.

Let me know if I can do anything to help troubleshoot

 

Edited April 29, 2021 by nbbsn

[PTRFRLL]

1 hour ago, nbbsn said:

I just updated to the latest image of the Trex miner from April 26 and am no longer able to start the container due to a CUDA error.

 

ERROR: Can't start miner, can't initialize CUDA engine, cuda exception in [main, 609], CUDA_ERROR_NO_DEVICE. Is NVIDIA driver installed?

 

Nvidia driver (v455.45.01) has been installed for months ever since unraid 6.9.0 came out and is still working perfectly in jellyfin, emby and previous versions of Trex. 
I rolled back to the previous image from 4 weeks ago and it's working flawlessly again, so something in that latest build screwed up my GTX 1070.

Let me know if I can do anything to help troubleshoot

 

 

Shoot. I updated the CUDA base image (to CUDA 11.2) in the latest update. The "easy" answer is to update your nvidia drivers. That said, I know that's not an acceptable answer for everyone. I guess I'll downgrade the base image back to 11.1 which should be compatible for all/most nvidia drivers.

Edited April 29, 2021 by PTRFRLL

[nbbsn]

18 minutes ago, PTRFRLL said:

 

Shoot. I updated the CUDA base image (to CUDA 11.2) in the latest update. The "easy" answer is to update your nvidia drivers. That said, I know that's not an acceptable answer for everyone. I guess I'll downgrade the base image back to 11.1 which should be compatible for all/most nvidia drivers.

Is there a performance benefit to using CUDA 11.2 for the cards/drivers where it's available? If so you could possibly offer a separate 11.1 build for compatibility, although that would introduce some additional overhead for maintaining the image.

 

/edit: thanks for the quick response, I appreciate it.
Just tested the new build you pushed and it works perfectly again.

Edited April 29, 2021 by nbbsn