Dear Homey Developer Support,
I am developing a custom driver for the RCS TBZ48 thermostat.
I will be leaving the country on the 9th of this month for more than six months. Since debugging requires physical access to the device within the same LAN, I urgently need to resolve the remaining platform-related issues before departure.
Repository:
I am encountering two platform-level inconsistencies.
Secure State Fluctuation
Observed:
-
Secure (S0) was successfully negotiated in earlier inclusion
-
Developer Tools now display secure state as: Secure: x
-
Facts:
-
COMMAND_CLASS_SECURITY detected
-
No identifier changes
-
Same hardware
-
Same controller
Questions:
Under what conditions does Homey mark secure state as “X”? -
-
Can Homey downgrade a node from secure to non-secure post-interview?
-
Does a failed nonce exchange result in non-secure classification?
-
Is there any caching behavior affecting secure display?
-
Battery Capability StatusFacts:
-
BATTERY_REPORT received
-
Battery percentage logged correctly
-
measure_battery capability implemented
Issue:
-
-
UI occasionally displays battery state as unknown
Questions:
-
Does Homey require explicit WAKE_UP handling for battery capability stability?
-
Is measure_battery dependent on polling configuration?
-
Does Homey mark battery as unknown if no report is received within a specific timeout?
-
Is registerCapabilityListener required for battery devices?
Condensed Log Extract:
Node Interview Complete
COMMAND_CLASS_BATTERY supported
COMMAND_CLASS_SECURITY supported
Battery Report received: 87%
Secure: false
Wake Up Notification received
I would appreciate clarification on:
-
Secure state evaluation logic
-
Battery capability state handling for sleeping devices
Given the time constraint before my departure, your prompt assistance would be greatly appreciated.