v1.0.5
What's Changed
Four reliability fixes addressing root causes found after v1.0.4. Node entities now survive restarts correctly via a persistent cache, ENUM sensors no longer crash on DucoBox startup placeholder values, and the discovery scan reliably finds all nodes regardless of their ID or response time.
Changes
- Fixed:
ValueError: Sensor provides state value '-'— DucoBox returns-as a mode/state placeholder during startup; ENUM sensors now reject values not in their options list and also clear any stale-value persisted by RestoreSensor from a previous session - Fixed: Node entities showing
unavailablewithrestored: trueafter restart — node list is now persisted to HA storage after each successful poll and pre-loaded at startup, so all previously-known nodes are present on the first coordinator tick and entity instances are created immediately - Fixed: Slow-responding nodes consistently missed by discovery scan — per-node fetch timeout raised from 2s to 10s
- Fixed: Node IDs 11–49 never scanned — discovery range expanded from two partial ranges to the full
range(2, 101)
Upgrade Notes
- A full Home Assistant restart is required (not just a config entry reload)
- On first restart after upgrade, all nodes will be re-discovered and the cache populated; entities will be stable on all subsequent restarts