[Nut-upsdev] tripplite smart2000rmxl2u product id 3014

[Charles Lepple]

On Fri, Oct 17, 2008 at 11:55 AM, Ken Teh <teh at anl.gov> wrote:
> I have it sort of working.  I was able to run the usbhid-ups driver with the productid as you suggested but I have to run everything as root.  Here's a quick summary of results:
>
> (1) The distro's 52.nut rules file generates error messages in the syslog.  So, I rewrote them.  But this had no effect.  I was able to see a mapped device /dev/hiddevXX with owner/permissions I specified but the  nut driver and programs still do not run without the '-u root' option.  Does the driver open a device file in /dev or does it walk the usb tree somewhere else?  In /sys?

/dev/hiddev* is not used by usbhid-ups (it is Linux-specific, and
until the 2.6 kernels, was not capable of distinguishing between
several of the collections of UPS variables).

Depending on your distro version, the udev rules probably need to
modify the node in /dev/bus/usb.

> (2) Running everything as root, I was able to generate notify events when I take the UPS on line power and off line power.
>
> (3) I also checked with a upsmon slave and everything works as expected.
>
> Today was the smoke test day.  I took the UPS off line power, partially drained the battery with a electric heater, and monitored the battery charge with upsc.  It kept dropping down to 1 and stayed there like forever.  upsmon never went through executing the low battery notify event or the shutdown.  I noticed that the UPS LEDs indicated adequate power (it was still green) so this leads me to the next questions:

What fraction of the UPS power capacity was the heater wattage?

> (1) It seems that the driver does not read the UPS correctly.  At least it does not read the battery charge correctly.

In order to see what is going on, we would probably need to look at
the debug logs during a discharge cycle. (Please compress them before
sending them to the list.)

> (2) What actually triggers the shutdown and low battery events in upsmon?  Is it the battery charge or something else?

It is the UPS's "low battery" signal, in combination with an "on
battery" condition. (This prevents problems when the battery is low
after the power returns, but the UPS has decided to turn on the load
anyway.)

NUT does not use the battery.charge value to trigger a shutdown.

> I'd appreciate a quick response and suggestions on moving forward.  If I cannot resolve this quickly, I will have to give up on nut and use something else.

Bear in mind that, for many of us, working on NUT is not endorsed by
our employer, so responding quickly is often not an option during
business hours.

Also, we are doing the best we can with no formal documentation on
Tripp Lite's implementation of the HID PDC protocol, and without
direct access to the particular hardware that you are using.

-- 
- Charles Lepple