[Nut-upsuser] UPS hardware periodically resets

graeme vetterlein graeme.nut at vetterlein.com
Mon Nov 24 16:27:13 GMT 2025


I have a APC ups:


device.mfr: American Power Conversion
device.model: Back-UPS XS 700U


I use:

driver = usbhid-ups
port = auto
vendorid = 051d


And I set a few values:


offdelay = 300
ondelay = 70
lowbatt = 33


This all works fine.


However I noticed that often these values would get reset:


initial value: ups.delay.shutdown: 300

Later value:   ups.delay.shutdown: 20


Since this happened a lot, I added a cronjob to daily check it. It 
triggers about every 8-10 days.

I can reset these with:

# systemctl restart nut-driver at qnapups


and everything is fine.


But from logs:

Nov 16 17:38:49 zbox systemd[1]: Started nut-driver at qnapups.service - 
Network UPS Tools - device driver for NUT device 'qnapups'.
Nov 16 17:39:05 zbox usbhid-ups[1526738]: sock_connect: keeping default 
synchronous mode
Nov 22 19:15:20 zbox usbhid-ups[1526738]: ups_status_set: seems that UPS 
[qnapups] is in OL+DISCHRG state now. Is it calibrating or do you 
perhaps want to set 'onlinedischarge' option? Some UPS models (e.g. 
CyberPower UT series) emit OL+DISCHRG when offline.
Nov 23 21:05:22 zbox usbhid-ups[1526738]: nut_libusb_get_interrupt: No 
such device (it may have been disconnected)
Hint: Some lines were ellipsized, use -l to show in full.


at 9:00 on the 24th my cronjob noticed the values had reset. So


Just after Nov 22 19:15:20 (ie 9:00 on 23rd) ...the values were still OK

Just after Nov 23 21:05:22 (ie 9:00 on 24th) ... the values changed.


My personal theory is that the USB powered device in the APC box 
"crashed" and rebooted, resetting some values on the way.


Now I don't think that NUT can do a whole lot about this .... probably 
my simplest fix is to trigger the "systemctl restart" when I spot the 
change.


But it does suggest to me , it would be useful if I could trigger some 
actions ( e.g. systemctl restart  ) when NUT detects some error conditions:


     1: the OL+DISCHRG ... ?

     2: the (temporary) loss of the USB device


--


Graeme









-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/nut-upsuser/attachments/20251124/a74437f1/attachment.htm>


More information about the Nut-upsuser mailing list