[Nut-upsuser] Tripplite UPS

[Arjen de Korte]

Citeren Lelsie Rhorer <lrhorer at satx.rr.com>:

> 	2. NUT 2.2.2-6.2  I'm not sure how to look up any extensions, unless
> you need me to look up the versions of libc6, libupsclient1,libusb, etc.
> The only other package I am running other than NUT (and of course its
> dependencies) is knutclient 0.9.4-1

This is important information (see below).

> 	3. Permanently.  That's why I had to write the script.

Sadly, this is a known problem with the tripplite_usb driver in  
nut-2.2.1 and nut-2.2.2. Instead of reconnecting, it will segfault  
when the connection to the UPS is lost. Depending on local  
circumstances, this may never happen or as often as once every few  
minutes. Upgrading to the latest version from the trunk, or to  
nut-2.2.0 should fix this problem.

> 	4. What Windows shows is correct.  It's really not hard to know a
> fully charged 24VDC battery pack should not read as 16V, and as I said, a
> voltmeter confirms the reported values are incorrect.  Irrespective of what
> Windows does, the Linux driver should report correct values.

If it was that simple as you're suggesting here, it would already be  
done. We're not reporting 'incorrect' values because we are too lazy  
to write decent drivers, but simply because we don't have the  
information available to calculate the correct values. Either learn to  
live with this, or switch to a vendor that actually does support NUT  
(by providing protocol information and/or hardware for driver  
development).

If the UPS is telling us that the battery voltage is 14.5 V, that's  
all we know about it. Apparently to get to the actual battery voltage,  
some calculation is involved. But since Tripplite doesn't tell us,  
there is no way to know if we need to add 12V to the reported value or  
multiply it by 1.8 (or any combination of multiplying with a  
coefficient and adding an offset value). Worst of all, the reported  
voltage may be correct (battery broken). How should we know?

The same goes for the reported values for input- and output voltage.

Best regards, Arjen
-- 
Please keep list traffic on the list