[Nut-upsuser] LiebertPSP

[Charles Lepple]

On Sep 24, 2011, at 5:48 AM, Tim Gould wrote:

> Apologies for resurrecting an old thread and likely poor formatting  
> - I wasn't subscribed to the list at the time and have copied it out  
> of the archives (http://lists.alioth.debian.org/pipermail/nut-upsuser/2010-July/006104.html 
> ).

No problem. Thanks for linking to the old threads.

> I've checked out trunk and there seems to be no progress there. I'm  
> happy to help debugging the driver, is there still some way we can  
> progress this?

One difference that isn't as obvious in the latest code is in the  
debug output. We changed the "Value:" format from fixed-width to  
automatic decimal/scientific notation, so we should be able to see if  
the "ACPresent" values are truly 0, or just 1e-7 (which would have  
looked like 0.000000).

The fun part will be finding a heuristic that determines whether the  
exponents are wrong. Unfortunately, the two Liebert models have the  
same USB product/vendor ID pair.

>> Tue Jul 13 19:54:03 UTC 2010
>> Citeren Michelle Wright <michelle.wright op gmail.com>:
>>
>> > $ sudo /lib/nut/usbhid-ups -DDD -a maxfun
>> > Network UPS Tools - Generic HID driver 0.34 (2.4.3)
>> > USB communication driver 0.31
Can you stop the original driver, and re-run it as indicated above?  
Debug level 3 (as shown) should be sufficient. You can add something  
like "| tee /tmp/liebert-`date -I`.txt" to log to both the screen and  
a log file.

After the driver gets into the polling loop, press Ctrl-C. Please gzip  
the log and attach it to your reply.

[...]
>
>> >   0.350081     Report[buf]: (2 bytes) => 07 0d
>> >   0.350095     Path: UPS.PowerSummary.PresentStatus.ACPresent,  
>> Type:
>> > Feature, ReportID: 0x07, Offset: 2, Size: 1, Value: 0.000000
>> >   0.350106     Report[buf]: (2 bytes) => 07 0d
>> >   0.350120     Path: UPS.PowerSummary.PresentStatus.ACPresent,  
>> Type:
>> > Input, ReportID: 0x07, Offset: 2, Size: 1, Value: 0.000000
>>
>> Although I didn't expect otherwise, the UPS is reporting that AC is
>> not present (that's why you see NUT reporting OB instead of OL). Try
>> if these value change to '1.000000' if you unplug the mains.
> I tried this - unfortunately no they do not. I get this with power  
> disconnected:
>    0.422018	Report[buf]: (2 bytes) => 07 0d
>    0.422034	Path: UPS.PowerSummary.PresentStatus.ACPresent, Type:  
> Feature, ReportID: 0x07, Offset: 2, Size: 1, Value: 0.000000
>    0.422046	Report[buf]: (2 bytes) => 07 0d
>    0.422061	Path: UPS.PowerSummary.PresentStatus.ACPresent, Type:  
> Input, ReportID: 0x07, Offset: 2, Size: 1, Value: 0.000000
>
> With the upsd and upsmon running, I get a console message that I'm  
> on battery which stays that way regardless of whether the AC is  
> connected or not.