Nut upsdev - Nov 2005 - Re: TrippLite Omni1000LCD [Was: Belkin UPS software]

Dean K. Gibson wrote:
> 
> My Belkin UPS 1000 failed about a month ago;  all of a sudden it 
> reported a battery failure.  Since the Belkin logs showed this as in 
> instant transition from 99% to 22%, I assume a circuit board failure, 
> not a battery failure.  Similarly, one Belkin UPS 500 circuit board 
> failed after I used my 100W amateur radio transmitter (not connected to 
> the UPS), and another Belkin UPS 500 I gave my father has shown 
> problems, 
Yes, my experience with Belkin has been that their hardware and
software is crap. It barely works and easily breaks. 
> so I decided to replace them with multiple TrippLite 
> OMNI1000LCD units.  Unfortunately, these use a USB connector, and 
> TrippLite's Linux software doesn't support recent versions of Linux
> using USB (later than RedHat 8), so I'm presently running them without
a
> software connection.
> 
> I see from a Google search that you had a conversation with someone on 
> the NUT-upsdev list about the OMNI1000LCD.  Do you know what the state 
> of any development is for supporting the OMNI1000LCD?
Yes, I was in the process of writing support for the OMNI1000LCD, but
there were some strange problems. It would actually be very helpful if
you could try to reproduce the steps that Joel Peshkin did (discussed
on the nut-upsdev mailing list in early November). I would be very
interested to know if you will run into the same problems or not. 

In principle, the newhidups driver (newest development branch) should
work for this device, but for some reason it wasn't able to read any
variables from the UPS, and we were unable to determine the reason. 

If you could post your findings to the list that would help.

-- Peter

P.S. please cc any reply to the list. 
> 
> Sincerely, Dean Gibson
> 
> On 2003-11-24 08:59, Peter Selinger wrote:
> > Dean K. Gibson wrote:
> >   
> >> Amazingly, the following DNS bug is also fixed:
> >> ----
> >> upsd: the upsd program does a DNS lookup from its own hostname to
an IP
> >> address, EVERY SECOND. Since I log DNS queries, this means 604,800
extra
> >> lines in my weekly DNS logs (ugh). Adding the required
hostname-to-IP
> >> mapping to my /etc/hosts file solved this problem. (Thanks to Dean
Gibson
> >> for this information).
> >> ----
> >>     
> >
> > Thanks for letting me know.
> >  
> >   
> >> Whatever else one might say about the Bulldog Plus software, I
must say
> >> that fixing and releasing my two reported bugs within two weeks is
not only
> >> commendable, but outright amazing in this day and age.
> >>     
> >
> > Yes, that is pretty nice. I didn't get such a favorable response
to my
> > own bug report. 
> >  
> >   
> >> -- Dean
> >>
> >> ps:  Sorry about all the web page updates (grin)!  If it's all
right with
> >> you, I'm perfectly happy to forward to you whatever info I
discover about
> >> the Belkin UPS and software, because I think it helps the Internet
to have
> >> that info all at one web site, to the extent practical.  However,
if you
> >> ever give up and/or go to an APC UPS, and/or decide to abandon
your Belkin
> >> web pages, I'd be willing to preserve them on my site.
> >>     
> >
> > Okay, no problem. I am happy to maintain the web pages for now. 
> >
> >   
> >> Oh, and yes, when using my ups-belkin startup script, the
S99Bulldog links
> >> should be removed.
> >>     
> >
> > OK. 
> >
> > -- Peter

Per your suggestion, I have joined the list.

I'm still running Fedora Core 1;  am I going to have to upgrade to a 
later Fedora Core (eg, 4) in order to test the current NET development 
software with the OMNI1000LCD?  Attempts to load a current (2.0.2) NUT 
RPM release on my system seems to want a bunch of new stuff, including 
libusb and a newer version of RPM and POPT.  Why RPM is involved I 
cannot fathom.

Sincerely, Dean


On 2005-11-23 08:42, Peter Selinger wrote:
> Dean K. Gibson wrote:
>   
>> My Belkin UPS 1000 failed about a month ago;  all of a sudden it
reported a battery failure.  Since the Belkin logs showed this as in instant
transition from 99% to 22%, I assume a circuit board failure, not a battery
failure.  Similarly, one Belkin UPS 500 circuit board failed after I used my
100W amateur radio transmitter (not connected to the UPS), and another Belkin
UPS 500 I gave my father has shown problems,
>>     
> Yes, my experience with Belkin has been that their hardware and software is
crap. It barely works and easily breaks.
>   
>> so I decided to replace them with multiple TrippLite OMNI1000LCD units.
Unfortunately, these use a USB connector, and TrippLite's Linux software
doesn't support recent versions of Linux using USB (later than RedHat 8), so
I'm presently running them without a software connection.
>>
>> I see from a Google search that you had a conversation with someone on
the NUT-upsdev list about the OMNI1000LCD.  Do you know what the state of any
development is for supporting the OMNI1000LCD?
>>     
> Yes, I was in the process of writing support for the OMNI1000LCD, but there
were some strange problems. It would actually be very helpful if you could try
to reproduce the steps that Joel Peshkin did (discussed on the nut-upsdev
mailing list in early November). I would be very interested to know if you will
run into the same problems or not.
>
> In principle, the newhidups driver (newest development branch) should work
for this device, but for some reason it wasn't able to read any variables
from the UPS, and we were unable to determine the reason.
>
> If you could post your findings to the list that would help.
>
> -- Peter
>
> P.S. please cc any reply to the list. 
>