FlamongOle Posted May 11, 2021 Author Share Posted May 11, 2021 On 5/3/2021 at 4:25 PM, Rolucious said: I updated the driver in the nvidia pluging to 465.27, but that caused nvidia-smi to not properly report the GPU's for some reason, resulting that the cards weren't able to overclock. In the end the only nvidia driver I found working was 460.73. It's currently working correctly Ye, had the same experience here. Not so much I can do about it. Maybe there's some conflict with the 465 drivers between docker and unraid even if it's built from the same source, or just a bug. Won't use time to investigate this as it's probably at the near end of the ethereum mining era. Stable 460 drivers work, so I'll write a note about it at the front page. Thanks for the info trhough. Quote Link to comment
FlamongOle Posted May 11, 2021 Author Share Posted May 11, 2021 12 hours ago, corgan said: in the latest release v1.3.13 of nsf they added premetheus /metric support of the api. Is it possible to use this on this container? If it requires additional start up flag, then probably not. If it just requires open ports/port mapping, you can set that up manually yourself in the template setup and check if you can connect to it. I won't be using more time with this docker container unless it stops working. Quote Link to comment
FlamongOle Posted May 11, 2021 Author Share Posted May 11, 2021 Front page updated with Nvidia driver version and the clock table with additional information below it. Quote Link to comment
corgan Posted May 12, 2021 Share Posted May 12, 2021 23 hours ago, olehj said: If it requires additional start up flag, then probably not. If it just requires open ports/port mapping, you can set that up manually yourself in the template setup and check if you can connect to it. I won't be using more time with this docker container unless it stops working. in the docs of the original nsfminer i found this Quote Introduction nsfminer implements an API (Application Programming Interface) interface which allows to monitor/control some of the run-time values endorsed by this miner. The API interface is available under the following circumstances: If you're using a binary release downloaded from the releases section of this repository If you build the application from source ensuring you add the compilation switch -D APICORE=ON Activation and Security Whenever the above depicted conditions are met you can take advantage of the API support by adding the --api-bind argument to the command line used to launch nsfminer. The format of this argument is --api-bind address:port where nnnn is any valid TCP port number (1-65535) and is required, and the address dictates what ip the api will listen on, and is optional, and defaults to "all ipv4 addresses". Examples: ./nsfminer [...] --api-bind 3333 so i think it must be started with --api-bind 3333 May this be possible with this container? Quote Link to comment
RazorX Posted June 30, 2021 Share Posted June 30, 2021 is a NVIDIA Quadro P1000 4GB enough ram for this or no? Quote Link to comment
FlamongOle Posted June 30, 2021 Author Share Posted June 30, 2021 9 minutes ago, RazorX said: is a NVIDIA Quadro P1000 4GB enough ram for this or no? No Quote Link to comment
letrain Posted July 6, 2021 Share Posted July 6, 2021 just installed.. every time the container starts it says driver mismatch and has to download drivers. i have the nvidia-plugin drivers installed that are listed in the op. the one in the docker is higher and says it has to download the lower version i have installed.. -Trying to get Nvidia driver version--- ---Successfully got driver version: 460.84--- ---Checking Xwrapper.config--- ---Configuring Xwrapper.config--- ---Driver version missmatch, currently installed: v465.19.01, driver on Host: v460.84--- ---Downloading and installing Nvidia Driver v460.84--- op says not to use 465... Quote Link to comment
FlamongOle Posted July 6, 2021 Author Share Posted July 6, 2021 9 hours ago, letrain said: just installed.. every time the container starts it says driver mismatch and has to download drivers. i have the nvidia-plugin drivers installed that are listed in the op. the one in the docker is higher and says it has to download the lower version i have installed.. -Trying to get Nvidia driver version--- ---Successfully got driver version: 460.84--- ---Checking Xwrapper.config--- ---Configuring Xwrapper.config--- ---Driver version missmatch, currently installed: v465.19.01, driver on Host: v460.84--- ---Downloading and installing Nvidia Driver v460.84--- op says not to use 465... Don't use that version on the host (if you want to OC), haven't tried 470. The container handles the rest and are setup with default nvidia driver which will likely be replaced regardless - hence the missmatch. Nothing to worry about what's happening inside the container. Quote Link to comment
SaranG Posted July 14, 2021 Share Posted July 14, 2021 (edited) Hello, I am getting some notice in logs every few minutes. 15:50:13 miner GPU 0 gave incorrect result. Lower overclocking values if it happens frequently. Normally it will show as below: 16:01:29 miner Job: 32957e43 blk: 12826862 eth-us-east.flexpool.io:5555 16:01:30 miner 0:21 A0:F25 60.46 Mh - cu0 60.46 55C 65% 129.86W 16:01:32 miner Job: ffc6a8b3 blk: 12826862 eth-us-east.flexpool.io:5555 16:01:35 miner 0:21 A0:F25 60.45 Mh - cu0 60.45 55C 65% 129.90W 16:01:35 miner Job: 4e13f4c0 blk: 12826862 eth-us-east.flexpool.io:5555 I am having RTX 3070 Gigabyte gaming. Here are my settings: Power: 130 GPU Clock: -503 Mem Clock: 2300 Fan Speed: 65% Nvidia Driver: v460.73.01 Please let me know how to fix this if anyone knows. Thanks Edited July 14, 2021 by SaranG Quote Link to comment
FlamongOle Posted July 14, 2021 Author Share Posted July 14, 2021 8 minutes ago, SaranG said: Hello, I am getting some notice in logs every few minutes. 15:50:13 miner GPU 0 gave incorrect result. Lower overclocking values if it happens frequently. Normally it will show as below: 16:01:29 miner Job: 32957e43 blk: 12826862 eth-us-east.flexpool.io:5555 16:01:30 miner 0:21 A0:F25 60.46 Mh - cu0 60.46 55C 65% 129.86W 16:01:32 miner Job: ffc6a8b3 blk: 12826862 eth-us-east.flexpool.io:5555 16:01:35 miner 0:21 A0:F25 60.45 Mh - cu0 60.45 55C 65% 129.90W 16:01:35 miner Job: 4e13f4c0 blk: 12826862 eth-us-east.flexpool.io:5555 I am having RTX 3070 Gigabyte gaming. Here are my settings: Power: 130 GPU Clock: -503 Mem Clock: 2300 Fan Speed: 65% Nvidia Driver: v460.73.01 Please let me know how to fix this if anyone knows. Thanks It sort of says it.. either lower mem clock, or increase power and/or fan speeds and see if that helps.. GPU's are individual, try different values.. Quote Link to comment
nimaim Posted July 23, 2021 Share Posted July 23, 2021 v460.73.01 is no longer available in the nvidia-driver plugin. Any other versions on the Unraid host confirmed working with this? Quote Link to comment
FlamongOle Posted July 23, 2021 Author Share Posted July 23, 2021 1 minute ago, nimaim said: v460.73.01 is no longer available in the nvidia-driver plugin. Any other versions on the Unraid host confirmed working with this? I haven't tried and likely won't. I'll leave the docker and the page up, but I will end the "support" now due to decreasing popularity in mining and the dockerhub won't build for free automatic anymore. Quote Link to comment
nimaim Posted July 23, 2021 Share Posted July 23, 2021 6 minutes ago, olehj said: I haven't tried and likely won't. I'll leave the docker and the page up, but I will end the "support" now due to decreasing popularity in mining and the dockerhub won't build for free automatic anymore. Understandable, thank you for your work on this! It was the only thing that worked for me to OC + set fan speed in a docker container. I'll migrate over to passing the cards through in a VM. 1 Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.