2012/6/10 Tim Gould <tgo...@reverb.com.au> > I upgraded to trunk (3652) - it worked, but again with the seemingly > spurious RB during DISCHRG. >
more than a upsd trace, we need usbhid-ups (-DDDDD) to be able to see what's going on... Ie, is the device actually reporting RB (UPS.PowerSummary.PresentStatus.NeedReplacement == 1...) or not. > -DDD output attached. RB turns up just before battery.charge.low is > reached (around 1551 sec), at the same time LB turns up too. > > Once a slave client (on the same UPS) shuts down battery.charge goes back > up above battery.charge.low, so LB goes away, but RB stays. > > I didn't have upsmon running on this host so it ran until the battery was > depleted rather than shutting down. > the above use case should not be a problem, since an initiated shutdown (master + slave) will go to the completion. a variation, where the above result is desired, is load shedding, where you want to shutdown non critical loads, to get more runtime for the critical ones... > Hope this helps, thanks, Tim. > it indeed helps, thanks Tim. cheers, Arnaud -- Linux / Unix / Opensource Engineering Expert - Eaton - http://opensource.eaton.com Network UPS Tools (NUT) Project Leader - http://www.networkupstools.org Debian Developer - http://www.debian.org Free Software Developer - http://arnaud.quette.free.fr
