The reason I am posting here is I am wondering if this is a ZHA bug and how can I start to debug it.
At random but usually after many days even over a week of reliable usage, one or more TRV-ZBs will stop processing updates. I get : ember status 102 timeout.
BUT the TRV-ZBs are still connected OK. If I press a button (auto mode toggle for example) HA knows about it immediately and updates the card view. If I turn up the temperature using the dial on the valve to 23 degrees the valve opens , HA knows about it immediately and even triggers my heating to come on with an automation. If I look in the Logbook, a valve in this state is still checking in perodically.
If I try to send them a command like set local temperature offset, or set target temperature, the command times out in the logs.
EITHER this is a firmware issue on the valve… putting it into some asymmetric state… sends commands but doesn’t receive them.
OR there is something not right in ZHA
Rebooting the valve (batteries out… in … calibrate…) resolves it …but this is getting tedious.
I am wondering where to start, can I collect detailed zigbee logs for one device and see if it casts light on what’s going on.
Is there something else I could do to prove - is it possible to send a command from another device on the same network or something ?
I have never tried to update the firmware on the co-ordinator - should I ? Its a sonoff zigbee stick
EZSP
by Silicon Labs
Firmware: 6.10.3.0 build 297