[Nut-upsuser] Tripp Lite SMART1000RM2U

[Dawning Sky]

On Dec 10, 2012 5:16 AM, "Charles Lepple" <clepple at gmail.com> wrote:
>
> [please keep the list CC'd. thanks]

Sorry.  I was new to the list and didn't notice the reply-to wasn't set to
the list.

>
> On Dec 10, 2012, at 3:31 AM, Dawning Sky wrote:
>
> > On Sun, Dec 9, 2012 at 5:30 PM, Charles Lepple <clepple at gmail.com>
wrote:
> >> On Dec 7, 2012, at 3:01 AM, Dawning Sky wrote:
> >>
> >>> 1. battery.charge doesn't report correct charge level when the UPS is
> >>> On Batter.  It just reports 0.  Even though it will report 100 when on
> >>> line power.
> >>
> >> I wonder if this is connected with the following:
> >>
> >>> ups.debug.S: 31 34 30 00 64 30 0d '140.d0.'
> >>
> >> As I remember it, '4' is the status character representing "self-test
status is unknown". You might need to do a deep discharge test first before
the state-of-charge constants are programmed into the UPS.
> >>
> >> If you can start a deep discharge test from the front panel, I'd try
that first. Otherwise, there is a "test.battery.start" command in NUT which
might be what you're looking for. However, bear in mind that this driver
was written without the benefit of protocol documentation.
> >>
> >
> > Charles,
> >
> > Thanks for getting back to me.  I thought I did a battery test from
> > the front panel some time ago.  Just to be sure, I did another test
> > from the panel and a test by test.battery.start.  It doesn't seem to
> > help with the battery.charge.  Part of upsc output is
> >
> > battery.charge: 0
> > battery.test.status: Battery OK
> > battery.voltage: 25.20
> > ...
> > ups.debug.S: 31 30 30 00 00 30 0d '100..0.'
>
> OK. Not really sure what to tell you - that 5th byte is what is converted
to state-of-charge, and your original post had 64 when battery.charge was
100%.
>
> If you have a Windows system handy (might work in VirtualBox or VMWare),
you can compare with the regular Tripp Lite software.

I don't have a Windows.  But Tripp Lite claims PowerAlert runs on Linux.
Maybe I'll give it a try when I get a chance.

>
> > In any case, I can probably live with this.  But I'm really troubled
> > with the fact that the NUT daemon didn't issue the shutdown command,
> > even though Do you have any thoughts on this?  Is it possible that the
> > user "nut" doesn't have the privilege to shutdown?
>
>
> I misread the part where you mentioned that you got the low battery
signal - I was thinking "on battery".
>
> upsmon is designed to be started as root, and when it forks, it leaves
behind a parent process that retains root privileges:
>
> http://www.networkupstools.org/docs/man/upsmon.html#_reloading_nuances
>
> If SHUTDOWNCMD is a script, check permissions on it. (Even root needs
execute permissions on scripts - it's only the read/write permissions that
it can override.) Also, the next section in the above URL mentions 'upsmon
-c fsd', which can be used to test the shutdown sequence without draining
the UPS.

The permission is 755 and should be correct.  The shutdown command was the
default "/sbin/shutdown -h +0".  I did a FSD test.  The computer was shut
down but not the UPS.

>
> Are you building from source, or installing a package?

It was installed from Fedora RPMs.

>
> --
> Charles Lepple
> clepple at gmail
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20121210/8d109771/attachment-0001.html>