stefancrain / folding-at-home Goto Github PK

View Code? Open in Web Editor NEW

38.0 2.0 8.0 75 KB

A Folding@Home Docker container with GPU support

License: MIT License

Dockerfile 69.58% Makefile 30.42%

docker foldingathome folding-at-home amd64 nvidia-docker opencl gpu-support cpu covid-19 alzheimers-disease

folding-at-home's Introduction

Hi there 👋

folding-at-home's People

Contributors

Stargazers

Watchers

Forkers

daniel-j-h neofob mudler felipeportella egerszegilaszlo 1e100 ngoc7sao9 broknfutr

folding-at-home's Issues

Passkey

hello, is it possible to set an passkey to the Startup options?

Failed reading core package header

I have recently been receiving this error in the logs:

15:42:13:WU00:FS00:Downloading core from http://cores.foldingathome.org/lin/64bit-avx2-256/a8-0.0.12/Core_a8.fah
15:42:13:WU00:FS00:Connecting to cores.foldingathome.org:80
15:42:13:ERROR:WU00:FS00:Exception: Failed reading core package header.
15:42:14:WU01:FS01:Downloading core from http://cores.foldingathome.org/lin/64bit/22-0.0.20/Core_22.fah
15:42:14:WU01:FS01:Connecting to cores.foldingathome.org:80
15:42:14:ERROR:WU01:FS01:Exception: Failed reading core package header.

F@H dashboard shows both CPU and GPU processing in the yellow "ready" stage.

I would appreciate any help. Thanks in advance.

`docker run` aborting with error "could not select device driver "" with capabilities: [[gpu]]"

("Host") OS: Ubuntu 18.04
Docker version: 19.03.6 (package 'docker.io' version '19.03.6-0ubuntu1~18.04.1' installed from the Ubuntu standard distro repositories)

Command: sudo docker run --name folding-at-home --gpus all -p 7396:7396 stefancrain/folding-at-home:latest --user=REDACTED --team=REDACTED --power=full --gpu-usage=100 --cpu-usage=100

Result:

docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]].
ERRO[0000] error waiting for container: context canceled

EDIT: "Host" GPU seems to be supported by F@H (ie, I can find it on GPUs.txt:

01:00.0 VGA compatible controller: NVIDIA Corporation GP104GLM [Quadro P5000 Mobile] (rev a1) (prog-if 00 [VGA controller])
        Subsystem: Hewlett-Packard Company GP104GLM [Quadro P5000 Mobile]

Also, NVIDIA drivers are installed and working (ie, glxinfo and glxgears work), also clinfo works and lists the NVIDIA adapter above.

is this normal or an issue?

08:33:09:WARNING:WU00:FS00:Failed to get assignment from '18.218.241.186:80': No WUs available for this configuration                                    
08:33:09:ERROR:WU00:FS00:Exception: Could not get an assignment  ```

CLI example missing --gpus option for nvidia?

Thanks for this!

It seems that Docker-native GPU support seems to require passing an option like --gpus all to docker run - my GPU is not used without it. I'm on Docker 19.03.6.

https://github.com/NVIDIA/nvidia-docker#usage

Include k8s manifest file

You want want to put this in the repo, I don't know if I will be running it for awhile because not having GPU really limits my WUs

---
kind: Deployment
apiVersion: apps/v1
metadata:
  namespace: default
  name: folding-at-home
  labels:
    app: folding-at-home
spec:
  replicas: 2
  revisionHistoryLimit: 3
  selector:
    matchLabels:
      app: folding-at-home
  template:
    metadata:
      labels:
        app: folding-at-home
    spec:
      containers:
        - name: folding-at-home
          image: stefancrain/folding-at-home:7.5.1
          imagePullPolicy: IfNotPresent
          command: ["/bin/tini", "--", "/usr/bin/FAHClient", "--web-allow=0/0:7396", "--allow=0/0:7396", "--smp=true"]
          args: ["--user=Devin_Buhl", "--team=241985", "--gpu=false", "--power=full"]
          resources:
            requests:
              cpu: 100m
              memory: 256Mi
            limits:
              cpu: 1000m
              memory: 2Gi
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
            - matchExpressions:
              - key: node-role.kubernetes.io/worker
                operator: In
                values: 
                  - "true"
---
apiVersion: v1
kind: Service
metadata:
  name: folding-at-home
spec:
  selector:
    app: folding-at-home
  type: NodePort
  ports:
    - port: 7396
      nodePort: 30160
      targetPort: 7396

Dependency Dashboard

This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.

Open

These updates have all been created already. Click a checkbox below to force a retry/rebase of any.

Update ubuntu Docker tag to v24

Detected dependencies

docker-compose

docker-compose.yml

dockerfile

build/Dockerfile

ubuntu 22.04

github-actions

.github/workflows/folding-at-home.yml

actions/checkout v4

crazy-max/ghaction-docker-buildx v3

.github/workflows/schedule.yml

Check this box to trigger a request for Renovate to run again on this repository

Docker compose missing support for NVIDIA runtime

Similar to #3, but not so easily resolved.

It looks like there is a long open issue in compose #6691 to support the NVIDIA gpu runtime.

ERROR:exception: clWaitForEvents

I need some help.
I'm having erros when folding with GPU.

19:32:53:WU01:FS01:Requesting new work unit for slot 01: READY gpu:0:GP108 [GeForce GT 1030] from 40.114.52.201

19:32:53:WU01:FS01:Connecting to 40.114.52.201:8080

19:32:53:ERROR:WU00:FS00:Exception: Server did not assign work unit

19:33:25:WU01:FS01:Downloading 29.70MiB

19:33:35:WU01:FS01:Download 2.95%

19:33:41:WU01:FS01:Download 22.73%

19:33:47:WU01:FS01:Download 39.14%

19:33:53:WU01:FS01:Download 54.29%

19:33:59:WU01:FS01:Download 82.49%

19:34:03:WU01:FS01:Download complete

19:34:03:WU01:FS01:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:11777 run:0 clone:3636 gen:15 core:0x22 unit:0x00000020287234c95e73c43565f3e52a

19:34:03:WU01:FS01:Starting

19:34:03:WU01:FS01:Running FahCore: /usr/bin/FAHCoreWrapper /app/cores/cores.foldingathome.org/v7/lin/64bit/Core_22.fah/FahCore_22 -dir 01 -suffix 01 -version 705 -lifeline 6 -checkpoint 15 -gpu-vendor nvidia -opencl-platform 0 -opencl-device 0 -cuda-device 0 -gpu 0

19:34:03:WU01:FS01:Started FahCore on PID 16

19:34:03:WU01:FS01:Core PID:20

19:34:03:WU01:FS01:FahCore 0x22 started

19:34:04:WU01:FS01:0x22:*********************** Log Started 2020-04-04T19:34:03Z ***********************

19:34:04:WU01:FS01:0x22:*************************** Core22 Folding@home Core ***************************

19:34:04:WU01:FS01:0x22:       Type: 0x22

19:34:04:WU01:FS01:0x22:       Core: Core22

19:34:04:WU01:FS01:0x22:    Website: https://foldingathome.org/

19:34:04:WU01:FS01:0x22:  Copyright: (c) 2009-2018 foldingathome.org

19:34:04:WU01:FS01:0x22:     Author: John Chodera <[email protected]> and Rafal Wiewiora

19:34:04:WU01:FS01:0x22:             <[email protected]>

19:34:04:WU01:FS01:0x22:       Args: -dir 01 -suffix 01 -version 705 -lifeline 16 -checkpoint 15

19:34:04:WU01:FS01:0x22:             -gpu-vendor nvidia -opencl-platform 0 -opencl-device 0 -cuda-device

19:34:04:WU01:FS01:0x22:             0 -gpu 0

19:34:04:WU01:FS01:0x22:     Config: <none>

19:34:04:WU01:FS01:0x22:************************************ Build *************************************

19:34:04:WU01:FS01:0x22:    Version: 0.0.2

19:34:04:WU01:FS01:0x22:       Date: Dec 6 2019

19:34:04:WU01:FS01:0x22:       Time: 21:20:17

19:34:04:WU01:FS01:0x22: Repository: Git

19:34:04:WU01:FS01:0x22:   Revision: f87d92b58abdf7e6bf2e173cfbc4dc3e837c7042

19:34:04:WU01:FS01:0x22:     Branch: core22

19:34:04:WU01:FS01:0x22:   Compiler: GNU 4.8.2 20140120 (Red Hat 4.8.2-15)

19:34:04:WU01:FS01:0x22:    Options: -std=gnu++98 -O3 -funroll-loops

19:34:04:WU01:FS01:0x22:   Platform: linux2 4.9.87-linuxkit-aufs

19:34:04:WU01:FS01:0x22:       Bits: 64

19:34:04:WU01:FS01:0x22:       Mode: Release

19:34:04:WU01:FS01:0x22:************************************ System ************************************

19:34:04:WU01:FS01:0x22:        CPU: Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz

19:34:04:WU01:FS01:0x22:     CPU ID: GenuineIntel Family 6 Model 23 Stepping 10

19:34:04:WU01:FS01:0x22:       CPUs: 4

19:34:04:WU01:FS01:0x22:     Memory: 7.79GiB

19:34:04:WU01:FS01:0x22:Free Memory: 5.66GiB

19:34:04:WU01:FS01:0x22:    Threads: POSIX_THREADS

19:34:04:WU01:FS01:0x22: OS Version: 4.15

19:34:04:WU01:FS01:0x22:Has Battery: false

19:34:04:WU01:FS01:0x22: On Battery: false

19:34:04:WU01:FS01:0x22: UTC Offset: 0

19:34:04:WU01:FS01:0x22:        PID: 20

19:34:04:WU01:FS01:0x22:        CWD: /app/work

19:34:04:WU01:FS01:0x22:         OS: Linux 4.15.0-91-generic x86_64

19:34:04:WU01:FS01:0x22:    OS Arch: AMD64

19:34:04:WU01:FS01:0x22:********************************************************************************

19:34:04:WU01:FS01:0x22:Project: 11777 (Run 0, Clone 3636, Gen 15)

19:34:04:WU01:FS01:0x22:Unit: 0x00000020287234c95e73c43565f3e52a

19:34:04:WU01:FS01:0x22:Reading tar file core.xml

19:34:04:WU01:FS01:0x22:Reading tar file integrator.xml

19:34:04:WU01:FS01:0x22:Reading tar file state.xml

19:34:04:WU01:FS01:0x22:Reading tar file system.xml

19:34:04:WU01:FS01:0x22:Digital signatures verified

19:34:04:WU01:FS01:0x22:Folding@home GPU Core22 Folding@home Core

19:34:04:WU01:FS01:0x22:Version 0.0.2

19:34:24:WU01:FS01:0x22:Completed 0 out of 2000000 steps (0%)

19:34:24:WU01:FS01:0x22:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900

19:34:26:WU01:FS01:0x22:ERROR:exception: clWaitForEvents

19:34:26:WU01:FS01:0x22:Saving result file ../logfile_01.txt

19:34:26:WU01:FS01:0x22:Saving result file checkpt.crc

19:34:26:WU01:FS01:0x22:Saving result file science.log

19:34:26:WU01:FS01:0x22:Folding@home Core Shutdown: BAD_WORK_UNIT

19:34:57:WARNING:WU01:FS01:FahCore returned: BAD_WORK_UNIT (114 = 0x72)

19:34:57:WU01:FS01:Sending unit results: id:01 state:SEND error:FAULTY project:11777 run:0 clone:3636 gen:15 core:0x22 unit:0x00000020287234c95e73c43565f3e52a

Thank you for your work.

Move to 7.6.9 appears to not have created a publicly available image on docker-hub

Thanks for making this build! Super nice to be able to containerize this on my homelab.

I see that you moved the version file over to 7.6.9, but there doesn't appear to be a new tag on dockerhub. The version there it still is at 7.5.1 and hasn't been updated recently. Perhaps something went wrong with the Github workflow?