From 0bce1e1bed2ca77c406550f07f2ec49a5d74838f Mon Sep 17 00:00:00 2001
From: "matthias.lotz" <matthias.lotz@hobbyhimmel.de>
Date: Thu, 19 Mar 2026 19:11:03 +0100
Subject: [PATCH] fix(device-availability): fix timeout monitor and
 bridge-restart race conditions

- status_monitor: add availability_managed set; _monitor_loop skips devices
  in this set so the LWT/availability topic is the sole online/offline source
- device_manager: register device with status_monitor.set_availability_managed()
  so the monitor actually skips them (previously the monitor had no knowledge
  of DeviceManager.availability_managed)
- mqtt_bridge: remove blanket 'reset all devices to offline' on bridge restart;
  this was causing a race condition where the cron reset state AFTER the bridge
  had already sent device_online events via retained MQTT messages;
  stale running session cleanup is kept (still needed)
---
 .../open_workshop_mqtt/models/mqtt_bridge.py     | 16 ++++++++--------
 iot_bridge/core/device_manager.py                |  2 ++
 iot_bridge/utils/status_monitor.py               | 10 ++++++++++
 3 files changed, 20 insertions(+), 8 deletions(-)

diff --git a/extra-addons/open_workshop/open_workshop_mqtt/models/mqtt_bridge.py b/extra-addons/open_workshop/open_workshop_mqtt/models/mqtt_bridge.py
index 29ac598..03964dd 100644
--- a/extra-addons/open_workshop/open_workshop_mqtt/models/mqtt_bridge.py
+++ b/extra-addons/open_workshop/open_workshop_mqtt/models/mqtt_bridge.py
@@ -191,17 +191,17 @@ class MqttBridge(models.Model):
                     ),
                 })
 
-                # Bridge came back online → push fresh config and reset device states.
-                # This ensures Odoo and Bridge are in sync after a restart.
+                # Bridge came back online → push fresh config and close stale sessions.
+                # Device states are NOT reset here - instead we rely on the bridge
+                # sending device_online/device_offline events after reconnect.
+                # (Resetting here would race with the device_online events that
+                # the bridge sends immediately on startup via retained MQTT messages.)
                 if was_offline:
                     _logger.info(
-                        f"Bridge {bridge.name} came back online – pushing config and resetting device states"
+                        f"Bridge {bridge.name} came back online – pushing config and closing stale sessions"
                     )
-                    # Reset stale device states to 'offline' so UI is consistent
-                    # until the bridge reports real events.
-                    devices = self.env['ows.mqtt.device'].sudo().search([('active', '=', True)])
-                    devices.write({'state': 'offline'})
-                    # Also close any stale 'running' sessions
+                    # Close any stale 'running' sessions - they couldn't have ended
+                    # cleanly while the bridge was offline
                     stale_sessions = self.env['ows.mqtt.session'].sudo().search([
                         ('status', '=', 'running')
                     ])
diff --git a/iot_bridge/core/device_manager.py b/iot_bridge/core/device_manager.py
index 8ead701..7593ee5 100644
--- a/iot_bridge/core/device_manager.py
+++ b/iot_bridge/core/device_manager.py
@@ -151,6 +151,8 @@ class DeviceManager:
                 self.device_map[avail_topic] = device_id
                 self.mqtt_client.subscribe(avail_topic)
                 self.availability_managed.add(device_id)
+                if self.status_monitor:
+                    self.status_monitor.set_availability_managed(device_id)
                 logger.info("availability_topic_subscribed", device_id=device_id, topic=avail_topic)
 
                 status_topic = pcfg.get("status_topic", "")
diff --git a/iot_bridge/utils/status_monitor.py b/iot_bridge/utils/status_monitor.py
index 0b109bb..0792774 100644
--- a/iot_bridge/utils/status_monitor.py
+++ b/iot_bridge/utils/status_monitor.py
@@ -64,6 +64,9 @@ class DeviceStatusMonitor:
         self.devices: dict[str, DeviceStatus] = {}
         self.lock = threading.Lock()
 
+        # Devices managed by LWT/availability topic - timeout monitor skips them
+        self.availability_managed: set[str] = set()
+
         # Background thread
         self.monitor_thread: threading.Thread | None = None
         self.stop_flag = threading.Event()
@@ -152,6 +155,10 @@ class DeviceStatusMonitor:
                     self._emit_device_online(device_id, now)
                     self._save_status()
 
+    def set_availability_managed(self, device_id: str):
+        """Register a device as LWT-managed so the timeout monitor skips it."""
+        self.availability_managed.add(device_id)
+
     def mark_online_silent(self, device_id: str):
         """Mark device as online and update last_seen WITHOUT emitting an event.
 
@@ -232,6 +239,9 @@ class DeviceStatusMonitor:
 
                 with self.lock:
                     for device_id, device in self.devices.items():
+                        # Skip devices whose online/offline is managed by LWT
+                        if device_id in self.availability_managed:
+                            continue
                         # Check if device timed out
                         if device.is_online:
                             elapsed = now - device.last_seen