On Jan 10, 2015, at 11:07 AM, Steve Ballantyne <[email protected]> 
wrote:

> On Thu, Jan 8, 2015 at 11:37 PM, Charles Lepple <[email protected]> wrote:
>> Although this tends to happen more with x86 systems and/or USB3, you might 
>> want to see if the disconnects are less frequent with a shorter or 
>> higher-quality USB cable.
> 
> The more I look into this ... the more I think that it's exclusively
> Tripp-Lite weirdness!  There are some things that I explored as far as
> this being a Raspberry Pi problem, or a Raspbian problem, or even a
> Debian problem.  But any problems I could find out there have long
> since been resolved with confirmed bug fixes.

I forgot to mention, sometimes there are additional clues in the dmesg output 
that might point to cable issues-- but I am not holding my breath, since the 
RPi USB interface is somewhat different than UHCI or OHCI.

> I noticed there have been a few people over the years on the NUT list
> that have other models of Tripp Lite and have had this same issue of
> random disconnections.  So long as Tripp-Lite is not sharing their
> code with us, I think this will remain a mystery.  At some point they
> are going to *have* to either update their Linux binaries, or drop
> them altogether.  The specific distributions that they support with
> their proprietary software is loooong out of date.  Perhaps at that
> point they will open source the code?

It can't hurt to ask.

There have been a few references to other people with more than just one or two 
Tripp Lite UPSes deployed somewhere (e.g. Minutekey) so it might be worth 
pointing that out to a Tripp Lite sales rep as a potential distinguishing 
factor (especially now that Eaton has ended back their open source support).

> Also there seems to be some speculation (or perhaps you know for a
> fact) that the Tripp Lite actually includes a serial to USB converter
> inside of the device.  I have a stockpile of these awaiting
> deployment.  Maybe I will open one and look inside.

Just speculation on my part, given the similarities between the serial 
protocol, and the non-HID USB messages. The converter might actually be 
absorbed into a microcontroller (that could share code with the serial-only 
microcontrollers), but I haven't cracked the case on my UPS to check.

-- 
Charles Lepple
clepple@gmail




_______________________________________________
Nut-upsuser mailing list
[email protected]
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsuser

Reply via email to