[Nut-upsuser] Ablerex MARS

[lonely wolf]

On 04/23/2006 07:43 PM, Carlos Rodrigues wrote:

>Yes, that's probably it. The value displayed is rounded to one decimal point.
>
>I'll too need to take a look at those logs (taken when the UPS is
>fully charged), posting them to the list is a good idea.
>
>On 4/23/06, Peter Selinger <selinger at mathstat.dal.ca> wrote:
>  
>
>>Could the reason be that the UPS reports voltage to more than one
>>decimal, and the value used by the driver to calculate the load is
>>actually more accurate than the value displayed?
>>
>>I tried answering my own question by looking at lonely wolf's logs,
>>but could not retrieve it. -- Peter
>>
>>Carlos Rodrigues wrote:
>>    
>>
>>>Now, that's strange. It should read 100%. In fact, just to be sure I
>>>faked the voltage here and it reads 100%...
>>>
>>>Can you give me more details about your setup? Version, architecture
>>>and stuff...
>>>      
>>>
nut was retrieved with "svn co svn://svn.debian.org/nut/trunk". Last 
tests have been performed using revision 410.
I have uploaded on my web site the last versions of the rpms I have 
packaged and used (source and binaries).
All tests have been performed on a fully updated Centos 4.3 running on 
AMD Athlon XP 2600+. The UPS is connected via a [green] DB-9 serial 
cable received with the UPS (actually the only cable included in the box 
; no power cables, not even for connecting the UPS to 220V ).
Testing method:
a)
- start upsilon (which leaves a daemon named rupsd running)
- use upsilon to retrieve the status (a snapshot of the screen as 
reported by upsilon is available at http://wdl.lug.ro/ablerex)
- stop upsilon
b)
- run megatec -DDD -a ablerex  1> megatec.debug 2>&1 
- start upsd (with no arguments)
- use upsc ablerex at localhost to retrieve status
c)
compare the results



Please let me know if you need more details or any other tests and I 
will do my best to supply them if possible.

    Manuel