fightforlife / docker_hddfancontrol Goto Github PK

this is happening because mapping is done in container.
Solution could be to export the mapping.

Sorry for this question here but i am using hddfancontrol via docker and it works like a charm.
Since ive recently been using my homeserver also to transcode video, i ran into problems with high CPU temperatures.
Could someone give me a hint what to do with my docker-compose-file to also monitor CPU?
I dont care if this mix of CPU and HDD temperatures may also lead to permanent higher fan speeds.
Im just not very into this topic and am wondering if there is a way to just add the device CPU similar to my hdds:

version: "3"
services:
  hddfancontrol:
    image: fred92/hddfancontrol:master
    container_name: hddfancontrol
    restart: unless-stopped
    volumes:
      - /lib/modules:/lib/modules:ro
    privileged: true
    cap_add:
      - SYS_MODULE
    environment:
      - DEVICES=/dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdc2 #POSSIBLE TO ADD CPU HERE???
      - PWM_DEVICES=/sys/class/hwmon/hwmon2/pwm1 /sys/class/hwmon/hwmon2/pwm2
      - PWM_START=70 80
      - PWM_STOP=20 30
      - MIN_TEMP=30
      - MAX_TEMP=50
      - MIN_FAN=15
      - INTERVALL=60
      - SPINDOWN_TIME=900
      - LOG_PATH=/var/log/hddfancontrol.log

I’ve set this Docker image up on my TrueNAS Scale machine, and it works great briefly, but then displays the below message, cranks the fan to 100% and the container status shows “deploying”.

2023-02-16 05:09:46.205750+00:00ERROR [Main] ValueError: invalid literal for int() with base 10: ''
2023-02-16 05:09:46.205892+00:00INFO [Fan #1] Setting fan speed to 100%

Not really sure what could be causing that, guessing perhaps it misreads a drive temperature or something? I’m not familiar enough with Docker to really troubleshoot much, and TrueNAS makes it difficult to run things natively.

Hello,

I am trying to deploy this in TrueNAS Scale but it's been quite a journey and I still didn't get it to work so I'm now begging for your help.

On my system when I run sensors this is part of what I get

When I run the container I get the following error:

The variables I am using in the container are shown in the image. I changed from hwmon6 to hwmon5 because for some reason on the container it starts with hwmon0 and on the system itself it starts with hwmon1.

I'm not sure if I am doing this right or not. I checked the BIOS and the fan is exposed as PWM. It was Auto but I changed it to PWM.

Can you help me please?

Is it possible to not use the spindown feature? I tried leaving the env variable SPINDOWN_TIME empty but that produces an error from hddfancontrol because the docker entry script tries to pass a value. The same applies to the log file.

Furthermore it would be great if passing smartctl would be optional because if I understand the code here correctly this makes it impossible to use the new drivetemp kernel module based mode for fetching the temperature.

Hello,
I have been using this container for months and it works excellently.
However, as known, every reboot of linux "/dev/sd*" changes.
Having an m.2 sata disk, I need not to monitor it, as it reports high and wrong temperature.
Wanting to use persistent parameters, such as /dev/disk/by-id, the container goes "no such file or directory: '/dev/disk/by-id/***'" error.

I solved by mounting the /dev volume on the container '/dev:/dev:ro'
However, a cleaner solution valid for all use cases would be better (I use it on TrueNAS Scale)