[Nut-upsuser] Eaton/MGE 1500 Ellipse + NUT 2.4.1 on openSUSE 11.2

Arjen de Korte nut+users at de-korte.org
Wed Mar 3 19:30:19 UTC 2010

Citeren Roger Price <roger op rogerprice.org>:

> Pessimist: Shut down quickly as soon as power fails. When power  
> comes back, restart the server, but be ready to stop again as soon  
> as the power fails again. Hope to be able to stop as often as the  
> power fails.
> Since I live in a region with frequent short power outages, I choose  
> the pessimist strategy.

If you're living in a region with frequent short power outages, you'd  
be better served with a UPS that supports delaying restart of the load  
until the batteries have recharged to a certain minimum level. Your  
entry-level MGE/Eaton Ellipse series UPS doesn't have that, but the  
MGE/Eaton Evolution has. This will guarantee that the system will  
always have enough juice to complete a power-up and power-down  
sequence if the power returns and fails again at the worst possible  

Some comments on remarks on the page you created (thanks for that by the way)

   "Experience shows that detection of the USB interface to an Eaton/MGE UPS
    is a delicate matter, and best done during the initial power-on sequence."

The above is limited to the Eaton/MGE Ellipse series devices with  
combined USB/serial ports also known as USBS ports (RJ-45). It is not  
a problem with other MGE/Eaton devices.

   "Data stale : If the command upsc Eaton-66781 produces the result "Error:
    Data stale", try restarting the ups daemon with the command rcupsd  

This is typical for usbhid-ups driver and the MGE/Eaton Ellipse series  
with NUT versions before 2.4.2. It has to deal with a reply from the  
UPS that was misunderstood by the driver, occasionally resulting in a  
storm of reconnects. This effect is limited to the MGE/Eaton Ellipse  
series, other MGE/Eaton devices are not affected by this problem.

Best regards, Arjen
Please keep list traffic on the list

More information about the Nut-upsuser mailing list