[Nut-upsuser] Fwd: Data stale error after short while

[hyouko at gmail.com]

mmh.. I looked quickly at the logs:
- from time to time some character is swapped with something other
(just like in the 'F' reply of your previous logs), but this shouldn't
be a big problem;
- then, suddenly, the queries timeout (just like when you relaunch the
driver after a stale) - this is the problem.

We need to make sure this isn't something vm-related (if so, having
almost zero experience with it, I'm not the right person to help):
- can you test it with the same os out of a vm? or
- can you setup a vm with an old version of ubuntu server (I'm
thinking of 10.04) and see if it behaves differently? (We had atleast
one somewhat similar issue in the past - with a different driver -,
claimed to be caused by usb/kernel interactions, GitHub #122)
- you previously said you were not able to restart blazer_usb but
successfully (albeit unintentionally) shut down the UPS after the
driver stopped working: is this reproducible? Can you log the launch
of blazer_usb with the 'k' switch (and still with -DDD) after the
driver stops working?

Also:
- have you tried playing with poll interval ('i' option)?
- do you see any difference if you launch, as root, the driver with
the '-u root' option?