[Nut-upsuser] MGE UPS Comet EX RT 7 not setting LB

[Arnaud Quette]

2006/4/6, Philippe Marzouk <phil at ozigo.org>:
> On Thu, Apr 06, 2006 at 08:32:11AM +0200, Arnaud Quette wrote:
> > 2006/4/5, Philippe Marzouk <phil at ozigo.org>:
> > > ...
> > > This is now OK however there is a cosmetic issue:
> > >
> > > input.frequency: 500
> > >
> > > instead of 50.
> > >
> > > I checked on my Ellips units at home running Debian and Nut 2.0.3, the
> > > frequency is the expected 50 so this is specific to this unit it seems.
> >
> > known and logged bug ("mge-shut: manage exponent"):
> > https://alioth.debian.org/tracker/index.php?func=detail&aid=302073&group_id=30602&atid=411544
> >
> > I now prefer (and started yesterday evening) the long run approach,
> > that is to say creating a libshut.ch and begun to prepare the pseudo
> > merge into newhidups... After all, that was my first intention when I
> > forked mge-shut to create newhidups...
> >
>
> Wow, I still have to test the newest versions of newhidups on my
> Evolution as I am still running on serial :)

newhidups has evolved a lot, and is no more experimental for a long time now...

> I also noticed that I need to wait 1 minute when stopping mge-shut
> before the UPS will accept talking again. I had to add 'sleep 60' in my
> halt script else the upsdrvctl shutdown would not work.
>
> Is this a known behaviour ?

the known fact is that mge-shut (and newhidups) are very verbose, and
have  timeout that are protocol dependant. The worst case when
stopping should be 15 to 20 seconds, which is 3 times 5 sec for a
get_data that is failing.

Though I've improved the stop time reactivity, there are still rooms
for other things alike, so stay tuned.

Arnaud
--
Linux / Unix Expert - MGE UPS SYSTEMS - R&D Dpt
Network UPS Tools (NUT) Project Leader - http://www.networkupstools.org/
Debian Developer - http://people.debian.org/~aquette/
OpenSource Developer - http://arnaud.quette.free.fr/