April 19, 20251 yr Author 1 hour ago, anibis2 said: I replaced my tried-and-true fan control scripts with this docker earlier in the week and it's working great. Will it ever be possible to control NVIDIA GPU fans without having to use privileged? Just curious if it's just a flat no it can't happen, or something that's on the roadmap. I know this is still new, and it's already the best Unraid fan controller by a large margin. I didn't even know this platform existed. Thanks for bringing it to Unraid! I honestly don't know if it will be, it's out of my knowledge level. When I was looking into it I think there's essentially a mismatch between how CoolerControl prefers to handle GPUs versus how the Unraid driver works, and CoolerControl also just mainly operates with the assumption it has full access to the system. If someone figures it out I'd be happy to update the documentation. I spent a lot of time trying to adjust the container itself or even adjust the stuff I was mounting to the container and had no luck.
April 21, 20251 yr I just want to say thank you for all the work you have done on this. I just swapped motherboards because my gigabyte was defective and throwing memory errors. The new one's (Asrock TRX40 Creator) sensors weren't supported by hwmon, so I bought a Corsair iCue Commander Core XT, which liquidctl recognized immediately through CoolerControl, and set up the drive temps in hwmon. Last step is to allocate the correct mounts so that I don't have to run the container in privileged mode. All this to say I won't need to worry about my expensive drives cooking themselves to death, OR need a jet engine running in my office anymore.
April 22, 20251 yr This Nvidia power saver script seems to stop CoolerControl when --runtime=nvidia --gpus=all is added to Extra Parameters. https://github.com/SpaceinvaderOne/nvidia_powersave
April 22, 20251 yr Author 1 hour ago, MrSwe said: This Nvidia power saver script seems to stop CoolerControl when --runtime=nvidia --gpus=all is added to Extra Parameters. https://github.com/SpaceinvaderOne/nvidia_powersave What do you mean by stop? The container shuts down?
April 23, 20251 yr 7 hours ago, guniv said: What do you mean by stop? The container shuts down? Yes. And i get this in the logs. /tmp/appimage_extracted_d63f8c624a560e45bd3d28e317211f27/AppRun: line 4: 25 Killed exec -a $ARGV0 $APPDIR/coolercontrold "$@" Edited April 23, 20251 yr by MrSwe
April 23, 20251 yr Author 9 hours ago, MrSwe said: Yes. And i get this in the logs. /tmp/appimage_extracted_d63f8c624a560e45bd3d28e317211f27/AppRun: line 4: 25 Killed exec -a $ARGV0 $APPDIR/coolercontrold "$@" Can you share what the log shows for your Nvidia powersave script? I also use this script and have not encountered this issue, and this is what I get when I run the script: Listing NVIDIA Graphics Cards in your server... GPU 0: NVIDIA GeForce GTX 1070 (UUID: GPU-87c9b28c-e03f-b8fe-ea22-2ef8621b71c0) GPU 1: NVIDIA GeForce GTX 980 (UUID: GPU-b238369c-709f-7a8f-8efb-550113bb6937) Initial Power Usage and State -------------------------------------------- GPU 0 -- NVIDIA GeForce GTX 1070 Power Usage -- 9.24 W Power State -- P8 -------------------------------------------- GPU 1 -- NVIDIA GeForce GTX 980 Power Usage -- 15.32 W Power State -- P8 -------------------------------------------- Checking for process termination eligibility... At least one GPU is being used by a Docker container. Will not terminate GPU-related processes.
April 23, 20251 yr 2 hours ago, guniv said: Can you share what the log shows for your Nvidia powersave script? I also use this script and have not encountered this issue, and this is what I get when I run the script: Listing NVIDIA Graphics Cards in your server... GPU 0: NVIDIA GeForce GTX 1070 (UUID: GPU-87c9b28c-e03f-b8fe-ea22-2ef8621b71c0) GPU 1: NVIDIA GeForce GTX 980 (UUID: GPU-b238369c-709f-7a8f-8efb-550113bb6937) Initial Power Usage and State -------------------------------------------- GPU 0 -- NVIDIA GeForce GTX 1070 Power Usage -- 9.24 W Power State -- P8 -------------------------------------------- GPU 1 -- NVIDIA GeForce GTX 980 Power Usage -- 15.32 W Power State -- P8 -------------------------------------------- Checking for process termination eligibility... At least one GPU is being used by a Docker container. Will not terminate GPU-related processes. Listing NVIDIA Graphics Cards in your server... GPU 0: NVIDIA GeForce GTX 1060 6GB (UUID: GPU-4d618c93-4167-e443-0634-c0ef94fc0ba2) Initial Power Usage and State -------------------------------------------- GPU 0 -- NVIDIA GeForce GTX 1060 6GB Power Usage -- 10.69 W Power State -- P8 -------------------------------------------- Checking for process termination eligibility... No Docker containers are using the GPUs. Terminating non-Docker processes using NVIDIA devices... USER PID ACCESS COMMAND /dev/nvidia-uvm: root 2535408 F.... coolercontrold /dev/nvidia0: root 2535408 F.... coolercontrold /dev/nvidiactl: root 2535408 F.... coolercontrold GPU 0 is idle and not used by any Docker container. Optimising for power efficiency. GPU 0 optimisation attempt 2 confirmed. GPU 0 optimisation attempt 3 confirmed. Waiting for GPUs to adjust to new power states... -------------------------------------------- Final Power Usage and State -------------------------------------------- GPU 0 -- NVIDIA GeForce GTX 1060 6GB Power Usage -- 10.69 W Power State -- P8 -------------------------------------------- Note: Power states range from P0 (maximum performance) to P8 (lowest power state).
April 23, 20251 yr Author 2 hours ago, MrSwe said: Listing NVIDIA Graphics Cards in your server... GPU 0: NVIDIA GeForce GTX 1060 6GB (UUID: GPU-4d618c93-4167-e443-0634-c0ef94fc0ba2) Initial Power Usage and State -------------------------------------------- GPU 0 -- NVIDIA GeForce GTX 1060 6GB Power Usage -- 10.69 W Power State -- P8 -------------------------------------------- Checking for process termination eligibility... No Docker containers are using the GPUs. Terminating non-Docker processes using NVIDIA devices... USER PID ACCESS COMMAND /dev/nvidia-uvm: root 2535408 F.... coolercontrold /dev/nvidia0: root 2535408 F.... coolercontrold /dev/nvidiactl: root 2535408 F.... coolercontrold GPU 0 is idle and not used by any Docker container. Optimising for power efficiency. GPU 0 optimisation attempt 2 confirmed. GPU 0 optimisation attempt 3 confirmed. Waiting for GPUs to adjust to new power states... -------------------------------------------- Final Power Usage and State -------------------------------------------- GPU 0 -- NVIDIA GeForce GTX 1060 6GB Power Usage -- 10.69 W Power State -- P8 -------------------------------------------- Note: Power states range from P0 (maximum performance) to P8 (lowest power state). Thanks! I am writing new documentation now on this. In short the way the script works won't recognize coolercontrol as "used by a docker container" so I've edited the script to carve out a coolercontrol exception: https://github.com/guniv/CoolerControl-Docker/blob/main/tutorial/nvidia_power.sh I didn't catch this because I always have something on my server's GPU so thanks for reporting. Edited April 23, 20251 yr by guniv
April 23, 20251 yr 13 minutes ago, guniv said: Thanks! I am writing new documentation now on this. In short the way the script works won't recognize coolercontrol as "used by a docker container" so I've edited the script to carve out a coolercontrol exception: I didn't catch this because I always have something on my server's GPU so thanks for reporting. Nice, that worked. Thanks for the help.
April 24, 20251 yr Hey, I starter to have issues with CoolerControl very recently (past 2-3 days?). Basically one of the drives is being kept up and is doing this weird noise which I think is parking/unparking the head every few seconds. Turning off CoolerControl sorts it out, it stops immidiatelly. I did create a custom sensor combining all HDDs into one, but that was working for past ~month flawlessly, so probably not it? Log looks ok as well, the only error/warning are these ERROR coolercontrold::sleep_listener] Could not connect to DBUS, sleeping listener will not work! WARN coolercontrold::repositories::hwmon::hwmon_repo] Slow HWMon Device detected for: HUH72..... This device may be slow to update and respond. Any idea? Edited April 24, 20251 yr by rampa
April 24, 20251 yr Author 16 minutes ago, rampa said: Hey, I starter to have issues with CoolerControl very recently (past 2-3 days?). Basically one of the drives is being kept up and is doing this weird noise which I think is parking/unparking the head every few seconds. Turning off CoolerControl sorts it out, it stops immidiatelly. I did create a custom sensor combining all HDDs into one, but that was working for past ~month flawlessly, so probably not it? Log looks ok as well, the only error/warning are these ERROR coolercontrold::sleep_listener] Could not connect to DBUS, sleeping listener will not work! WARN coolercontrold::repositories::hwmon::hwmon_repo] Slow HWMon Device detected for: HUH72..... This device may be slow to update and respond. Any idea? Unfortunately this sounds like an issue with CoolerControl specifically and not the Docker container so I'll have to point you to them to ask for support if you still think it is CoolerControl causing the problem. However, I can say that I personally have not noticed CoolerControl causing my drives to spin back up after being slept and I haven't heard other reports of this so far (although there are temp reporting issues with spun down drives).
April 27, 20251 yr Thanks, I changed the polling rate from 1s to 5s, that looks like did the trick. I'm not really sure why has it started happening, but it looks like it's mostly gone 👍
May 13, 20251 yr Hello, Did you delete the plugin from CA and "re-add" it? It suddenly popped up as new apps and don't say installed on my screen eventho I have it. Do I need to delete the old one and reinstall the "new"?
May 13, 20251 yr Author 3 hours ago, Stabster said: Hello, Did you delete the plugin from CA and "re-add" it? It suddenly popped up as new apps and don't say installed on my screen eventho I have it. Do I need to delete the old one and reinstall the "new"? I edited the CA template earlier today to use an image from Docker Hub instead of from GitHub which I guess on CA's end makes it think it's a brand new app. You can update your repository on your container and it should mark it as installed again on CA, but I will continue to update the GitHub image and it shouldn't matter to existing users!
May 14, 20251 yr I can just install the new version, it's not an issue it just depends if you prefer to keep the docker hub release or the github one
May 21, 20251 yr This is such a cool container!!! Thank you.I just started using it for monitoring but hitting a bit of a snag. Basically I can monitor just fine but can't apply new profiles to my fans. The error in the console is "coolercontrold::api] Internal Error: Error on it8613:fan2 for Manual Control - Unable to set pwm_enable for "/sys/class/hwmon/hwmon7/pwm2_enable" to 1. Most likely because of a limitation set by the driver or a BIOS setting; Error: Read-only file system (os error 30)". Now, I tried resetting it to read-write in the container config but it doesn't seem to make any difference.I am out of time to troubleshoot it this morning as I have to get to work... but just to confirm my suspicion; given this error will I have to run this container privileged in order to control my fan profiles? I suspect so; I tried echoing "1" to /sys/class/hwmon/hwmon7/pwm2_enable and it worked fine, but the container clearly can't for whatever reason.
May 21, 20251 yr Author 2 minutes ago, Sinister Crayon said:This is such a cool container!!! Thank you.I just started using it for monitoring but hitting a bit of a snag. Basically I can monitor just fine but can't apply new profiles to my fans. The error in the console is "coolercontrold::api] Internal Error: Error on it8613:fan2 for Manual Control - Unable to set pwm_enable for "/sys/class/hwmon/hwmon7/pwm2_enable" to 1. Most likely because of a limitation set by the driver or a BIOS setting; Error: Read-only file system (os error 30)". Now, I tried resetting it to read-write in the container config but it doesn't seem to make any difference.I am out of time to troubleshoot it this morning as I have to get to work... but just to confirm my suspicion; given this error will I have to run this container privileged in order to control my fan profiles? I suspect so; I tried echoing "1" to /sys/class/hwmon/hwmon7/pwm2_enable and it worked fine, but the container clearly can't for whatever reason.Yeah, if you are passing through the it8613 directory (as gone over in the setup guide) and it's still throwing this error you may need to run the container privileged.
May 21, 20251 yr 8 hours ago, guniv said:Yeah, if you are passing through the it8613 directory (as gone over in the setup guide) and it's still throwing this error you may need to run the container privileged.Thanks so much. Yeah, I actually ended up deleting and re-deploying it, but also rebooted my NAS out of curiosity (and it wasn't sleeping disks properly so that was another issue entirely)... looks like it's all working perfectly now with no privileged mode required. Awesome :)
May 28, 20251 yr I was finally able to install it with update to unraid 7.1.2!Thanks a lot for the software! finally I can manage all my fans based on HDD temperature!
June 19, 20251 yr I'm facing an issue where after every Unraid host reboot, the HDD custom sensor resets to being empty. It does not reset if I just restart the container. The VRM custom sensor does not have this problem, only the HDD ones. Something to do with the drivetemp module perhaps? Anyone has any clues?I'm also running in privileged mode because without it, CoolerControl can't change the fan speed, only read it.
June 19, 20251 yr Author 47 minutes ago, Sak said:I'm facing an issue where after every Unraid host reboot, the HDD custom sensor resets to being empty. It does not reset if I just restart the container. The VRM custom sensor does not have this problem, only the HDD ones. Something to do with the drivetemp module perhaps? Anyone has any clues?I'm also running in privileged mode because without it, CoolerControl can't change the fan speed, only read it.My best guess here is that perhaps the CoolerControl container is starting before the script that enables drivetemp has a chance to start.Maybe one way around this would be to uncheck "Autostart" for the container in the Docker tab of Unraid and add this line to the drivetemp script:docker start CoolerControlSo that will start the container after drivetemp has been enabled. Edited June 19, 20251 yr by guniv
June 20, 20251 yr 22 hours ago, guniv said:My best guess here is that perhaps the CoolerControl container is starting before the script that enables drivetemp has a chance to start.Maybe one way around this would be to uncheck "Autostart" for the container in the Docker tab of Unraid and add this line to the drivetemp script:docker start CoolerControlSo that will start the container after drivetemp has been enabled.That did it. Thanks!I added a 5 seconds sleep just to be sure.sudo modprobe drivetemp sleep 5 docker start CoolerControl
June 30, 20251 yr It's the best fan control software I've used on Unraid. Autofan couldn't detect my SSD temperature to adjust fan speed, but this one can. When I finally got the fan speed to adjust based on SSD temperature, I couldn't believe I hadn't discovered this gem sooner! Many thanks to you and the Docker author for your work! However, why doesn't it seem to detect my HDD temperature? Is this a bug or something?
June 30, 20251 yr Author Just now, iti said:It's the best fan control software I've used on Unraid. Autofan couldn't detect my SSD temperature to adjust fan speed, but this one can. When I finally got the fan speed to adjust based on SSD temperature, I couldn't believe I hadn't discovered this gem sooner! Many thanks to you and the Docker author for your work! However, why doesn't it seem to detect my HDD temperature? Is this a bug or something?Please read this guide: https://github.com/guniv/CoolerControl-Docker/blob/main/setup.mdYou need to run some commands to enable HDD temperature.
June 30, 20251 yr 9 minutes ago, guniv said:Please read this guide: https://github.com/guniv/CoolerControl-Docker/blob/main/setup.mdYou need to run some commands to enable HDD temperature.Oh, it works now. I was so happy that I forgot it was mentioned in the documentation, haha. Thank you once again, sincerely.
.thumb.jpg.23e9326f2055355990482d64cc4f27a6.jpg)
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.