Problem with my UPS on NUT (HomeAssistant)

networkupstools / nut

The Network UPS Tools repository. UPS management protocol Informational RFC 9271 published by IETF at https://www.rfc-editor.org/info/rfc9271 Please star NUT on GitHub, this helps with sponsorships!

https://networkupstools.org/

Other

1.79k stars 339 forks source link

Problem with my UPS on NUT (HomeAssistant) #2353

Open studioIV opened 4 months ago

studioIV commented 4 months ago

I am new to Homelab and just got my new ups Powerwalker VI 1100 CW but I can't find a driver that works with NUT(Running in HomeAssistant). To be honest I found this driver ( nutdrv_qx ). But most of the time the ups disconnect randomly and show up as anavailable. I use this computer as a proxmox server: M83 Desktop (ThinkCentre) SFF - Type 10AH.

jimklimov commented 4 months ago

Hello, so far I see not too many details to follow up on; however a few thoughts did come to mind:

Check our Wiki, e.g. https://github.com/networkupstools/nut/wiki/Troubleshooting-eventual-disconnections-(Data-stale)
Some PowerWalkers may be based on CPS products, so https://github.com/networkupstools/nut/wiki/CyberPower-Systems-(CPS)-know-how specifically can apply (maybe increasing polling frequency can help avoid that controller chip going to sleep)
I can only assume that you use a newest build of HA NUT plugin (third-party project as far as NUT is concerned) which may be based on some relatively recent NUT 2.8.x packaged release - so https://github.com/networkupstools/nut/wiki/Building-NUT-integration-for-Home-Assistant may apply if you endeavour to build your own, or https://github.com/networkupstools/nut/wiki/Changing-NUT-daemon-debug-verbosity to set up more detailed debug logging to catch those disconnections.

andyevans-eng commented 3 weeks ago

I have setup my Powerwalker UPS using the NUT integration, it kept losing connection with the stale data error after a few hours, I have configured the maxage and polling interval etc and it would run for about 20 hours before getting the same error. After reading through many forums I changed the polling interval to pollonly = "enabled" It worked perfectly for about 36 hours, but am now back to the stale data problem. I really need to get this working properly as I have safety automations that depend on the UPS data, which obviously don't work with stale data. HA running on PI5 8Gb Ram, everything up to date. Any suggestions please? Yes I can create an automation to restart when UPS becomes unavailable, but that defeats the object, plus I don't want my HA out of action for a minute or 2 almost every day. Thanks in advance, Andy

jimklimov commented 3 weeks ago

One more idea that comes to mind is adding cron scripts (generally, maybe some HA actions would be more idiomatic for you) to restart NUT and/or just try to soft-reset the USB port.

Check examples of such ideas at: