You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
HA Discovery can fail silently per-network (e.g. 401 Network not found
on startup, the v1.8.1 fix recovers via retry but ultimately gives up
after 8 attempts). Users had no surface in HA itself for whether
discovery was healthy; the only signal was a log warning. Anyone hitting
the startup race had to trawl the add-on logs to know what was wrong.
Add a per-network "Discovery (Network N)" diagnostic sensor with three
states: discovering, ok, paused. Published via HA MQTT Discovery under
the existing cgateweb Bridge device so it groups with other diagnostics
(Bridge Ready, MQTT Connected, etc.). Config payload published once;
state publishes are de-duplicated by previous value. State transitions:
- queueTreeRequest() -> discovering
- handleTreeEnd() success -> ok
- _handleTreeRequestFailure -> paused (only when retry budget exhausts)
Bumps to 1.8.4.
Copy file name to clipboardExpand all lines: homeassistant-addon/CHANGELOG.md
+5Lines changed: 5 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -5,6 +5,11 @@ All notable changes to the C-Gate Web Bridge Home Assistant add-on will be docum
5
5
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
6
6
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
7
7
8
+
## [1.8.4] - 2026-05-04
9
+
10
+
### Added
11
+
-**Per-network discovery health sensor**: HA Discovery now publishes a "Discovery (Network N)" diagnostic sensor for each configured network, with three states — `discovering` (request in flight or retry pending), `ok` (last TreeXML succeeded), or `paused` (retry budget exhausted from the v1.8.1 startup-race protection). The sensor lives under the existing cgateweb Bridge device in HA, so users can see at a glance whether auto-discovery is healthy without trawling the add-on logs. State publishes are de-duplicated; the HA Discovery config payload is published once per network.
this.logger.warn(`Received a new TreeXML start before previous tree completed; dropping incomplete tree for network ${this.activeTreeSession.network}`);
@@ -349,16 +423,18 @@ class HaDiscovery {
349
423
});
350
424
}else{
351
425
this.logger.info(`Parsed TreeXML for network ${networkForTree} (took ${duration}ms)`);
0 commit comments