Journey event  ·  6 Sep 2026

Non-climate Zigbee datapoints (CO₂, lux, contact, occupancy, power) never become fleet entities — catalog-only roles are dead ends

VERIFIED LIVE 2026-09-06 (PR #197 hotpatched to dsc-brain.local, QA device via mosquitto_pub — no real CO₂ hardware bound yet): bound 0xqa00000000c02 → co2_tent, published {co2:812

  state copied from the tracker, never improved

  event date 2026-09-06

Event date 2026-09-06
Version found
Version fixed
Expected kit firmware
Release file
SHA-256
Served count
Source https://app.notion.com/3d32b4cda37081a3a926e1ec2af6040a

VERIFIED LIVE 2026-09-06 (PR #197 hotpatched to dsc-brain.local, QA device via mosquitto_pub — no real CO₂ hardware bound yet): bound 0xqa00000000c02 → co2_tent, published {co2:812,voc:30,linkquality:100} on zigbee2mqtt/qa_co2 → zigbee_by_role.co2_tent {co2:812, voc:30, kind:gas}; /fleet?include_hass=true shows sensor.dsc_zigbee_co2_tent_co2=812 ppm, _voc=30 ppb, _linkquality=100 lqi; /settings/automations/targets lists all three under entities; Climate ‘Zigbee by role’ renders the new ‘Other sensors’ row (co2_tent · 4×8 · qa_co2 · CO₂ 812 ppm · VOC 30 ppb — screenshot taken); a v2 banner rule on sensor.dsc_zigbee_co2_tent_co2 > 800 fired on the next evaluation tick and cleared at co2=400; bindings restored afterwards. — BUILT: zigbee_by_role carries every bound role with a kind; fleet_state exports each datapoint generically (number → sensor with catalog unit, bool → binary_sensor, string → sensor); unbound devices never exported; automation_rules age/targets cover every zigbee entity; 8 formerly dead-end roles now consume:true. 8 tests in test_zigbee_datapoints.py. — ORIGINAL: to_hass_states only exposed temperature/humidity; 8 of 21 roles did nothing after binding.

All events  ·  DSC  ·  CannaLib  ·  Site  ·  Firmware releases

Empty rows are honest

A blank cell means the tracker has no value for it. Nothing on this page is filled in to look complete.