auiayuiox

bummer, only found this thread after buying a BX1200MI for my NAS and immediately getting this error and googling it to end up here. I started with it connected to a USB 2.0 port - I've switched to a 3.2 gen 1 port and will see if that makes any difference. if the error can be safely ignored, then whatever, but i'm not sure if the increasing NUMXFERS are normal or part of this bug as some other posts suggest - it seems to be doing a self test every 5 minutes, which doesn't seem crazy to me? UPDATE: this might help some others - I am pretty sure my motherboard uses an Asmedia controller for the USB 2.0 ports, and the one 3.2 Gen 1 port it has. I had this issue manifest with the UPS on all these ports. I switched to the motherboard's USB 3.2 Gen 2 ports, which, from reviewing lsusb, go directly via the Intel B660 chipset. As soon as I switched to one of these ports, I noticed the same thing as other posters: the UPS went to 100% (from 99%), and NUMXFERS stopped iterating. The automatic self tests I was seeing every 5 minutes have stopped as well, so CUMONBATT has not gone up either. So take a look at what USB controller you're connecting it via. But most importantly it looks like, at least in my case, it's unnecessarily performing self tests while the issue manifests. UPDATE 2: I'm getting, on Unraid via apcupsd, the behaviour people above described when using NUT on separate infra to manage the UPS. So I am still getting the battery failure warnings using the USB port that goes directly via the Intel controller, but NUMXFERS and CUMONBATT have not gone up at all since switching to this port, and charge remains at 100%. Hope this helps someone else.

This started happening with my Ubuntu 22.04 VM after I did an apt upgrade (maybe new stuff in linux-firmware?). I pass through the iGPU for access to QuickSync. I can get the VM to boot if I force shut it down and start it up again cold.