Hi all, I have an issue that I’m not sure how it starts.
I got a Ups Cyberpower 425v 255w Cp425sLG that has the option to connect by USB to my home assistant system. I was very happy as the integrations was very easy, just plug in, reset my machine and it detect it automatically, I was running like this by about 1 month.
After this time, I have changed from 0.114 to 0.115 updates (And I have been installing all the 0.115 updates as they are released). The break of the system do not correlates to any update.
I start getting the following loop on the system console.
As you can see the USB port is detected as a cyberpower device, and then the USB conection is ended and it loops all the day.
I decide to remove the integration, disconect everything and try again, with no success. Even after removing the integration and rebooting, I can’t no longer set it up, as the hardware do not detect the UPS. but the console records remaing the same as the previews photo.
My system is the standar Hassos image for proxmox, mounted in a virtual machine.
When I open my proxmox enviroment and check on harder the USB port is passed out to HA and it says its a UPS.
The interesting part is that my log on HA do not show any error.
How can I debug this issue?
IF I need to add more information, please let me know.
I have a bunch of Cyberpower UPS’s and this is a known bug with the nut USB driver. It seems that the CyberPower controller likes to reset the USB connection on a specific interval. Some people have posted (on other forums) some work arounds, but they didnt work for me. The only thing I could do to get it to stay connected for a long period of time is to create a cron job to restart the nut-driver services every 20 minutes. Overkill? yes. But I have not had any data gaps (and therefore not gaps in shutdown protection) since I did this.
This is my crontab entry on the machine its connected to:
Thanks gregg, not sure this will work for me, as the speed of the restart is every few seconds.
Did you have any link to those sites? maybe any of those fixes hopefully will work for me.
The extrangest thing is that it work correctly for me for more than a month and sudenly it stop working.
Thanks again for your comments.
Thanks I will take a look on that and try to find more information.
I will connect this to a ubuntu PC I have to see if this problem is as showed on your link or if is something related to the interaction with my NUC PC or home assistant.
Hi dproffer, I try to check your commendation on my system and was not able to find the folder, I do not have it on the proxmox machine main system nor on the HA drivers.
Not sure if someone else have a different recommendation.
Good hunting! I am not familar with proxmox virtualization system. I run NUT on a rpi right in the raspbian linux. Somewhere NUT has to have a ups.conf file, I would doubt it will run without it. Hope you can find it as it does sound like the core problem you are facing is similar to the one I had.
My ST625 was failing due to constant USB disconnects, crashing the Pi zero it was connected to as well. Changing the poll to 15 seconds has made it very stable. 4 days now!
@dproffer – our units have a current wattage / load. In the PPB website, it’s listed in their javascript API as Output, titled “watts” – however, I couldn’t find anything in the latest MIB file that showed this.
Second, the “current load” MIB item which says “a percent of the load” is constantly at 27%, and doesn’t seem to be changing even when the display is. I was hoping to reverse engineer this percentage to show current watts load, but it’s static.
Sorry, I do not have any experience with MIB or direct API to any specific UPS vendor’s data. I just consume the data that NUT app extracts from my Cyberpower UPS’s:
