My system runs on a Raspberry Pi with a Sheepwalk RPi3 adapter and RPi3a
splitters giving me 8 busses, with about 20 temperature sensors and 6
I/O modules, all powered. I was having a few problems loosing one or
more busses, after some investigation I split the power so I have a
power supply for the Pi and a separate supply for the 1-wire.
I had been running without any problems for the last couple of weeks,
until today, when again it locked up, lost all channels. That has
happened 3 times today.
When it happens running owdir returns the standard /uncached /settings
/system, etc but no devices. Reboot makes no difference, power cycle
and it all works again.
I swapped the RPi3 adapter for a spare after the second failure, so I
think that is eliminated. I have checked syslog, nothing significant
there. 1-wire is running at 4.95v which I would think is OK.
My own logging shows this sort of thing, this is for DHW_Top, it is
repeated for the other devices
<class 'pyownet.protocol.OwnetError'> [Errno 2] legacy - No such
<class 'pyownet.protocol.OwnetError'> [Errno 1] Startup - command
line parameters invalid: u'DHW_Top/power'
I don't understand how it can run for about 2 weeks without a fault and
then have 3 in a few hours.
Any ideas what the problem could be? How to fault find further?
One further thought, I have one temperature sensor which would work
sometimes but for periods of sometimes several days I could not read
it. Is it possible that that single sensor could bring down hte whole
network? I have just disconnected it.