Improve resiliency of the code

Dockerfile

@@ -1,7 +1,7 @@
 FROM python:3.12-alpine
 
 # Install socat
-RUN apk add --no-cache socat
+RUN apk add --no-cache socat nano
 
 # Set the working directory in the container
 WORKDIR /usr/src/app

src/entrypoint.sh

@@ -3,31 +3,48 @@
 # Function to handle SIGTERM
 cleanup() {
     echo "INFO:SeplosBMS:Container stopped, cleaning up and exiting..."
-    # Send SIGTERM to the Python script
+    pkill socat
     kill -s SIGTERM $!
     wait $!
 }
 
 # Trap SIGTERM and call cleanup function
 trap cleanup SIGTERM
 
-# start socat for master if RS485_MASTER_REMOTE_IP and RS485_MASTER_REMOTE_PORT are set
+start_socat() {
+    local ip=$1
+    local port=$2
+    local device=$3
+    echo "Starting socat for device $device at ${ip}:${port}"
+    socat pty,link=$device,raw tcp:$ip:$port,retry,interval=.2,forever &
+    socat_pid=$!
+    sleep 2  # Give socat time to establish the connection
+    if ! kill -0 $socat_pid 2>/dev/null; then
+        echo "ERROR: Failed to start socat for device $device"
+        return 1
+    fi
+    echo "socat started successfully for device $device"
+}
+
+# Start socat for master and slaves
 if [ -n "$RS485_MASTER_REMOTE_IP" ] && [ -n "$RS485_MASTER_REMOTE_PORT" ]; then
-    echo "starting socat for master rs485 vcom ${RS485_MASTER_REMOTE_IP}:${RS485_MASTER_REMOTE_PORT}"
-    socat pty,link=/tmp/vcom0,raw tcp:$RS485_MASTER_REMOTE_IP:$RS485_MASTER_REMOTE_PORT,retry,interval=.2,forever &
+    start_socat "$RS485_MASTER_REMOTE_IP" "$RS485_MASTER_REMOTE_PORT" "/tmp/vcom0" || {
+        echo "Failed to start socat for master"
+        exit 1
+    }
 fi
 
-# start socat for slaves if RS485_SLAVES_REMOTE_IP and RS485_SLAVES_REMOTE_PORT are set´
 if [ -n "$RS485_SLAVES_REMOTE_IP" ] && [ -n "$RS485_SLAVES_REMOTE_PORT" ]; then
-    echo "starting socat for slaves rs485 vcom ${RS485_SLAVES_REMOTE_IP}:${RS485_SLAVES_REMOTE_PORT}"
-    socat pty,link=/tmp/vcom1,raw tcp:$RS485_SLAVES_REMOTE_IP:$RS485_SLAVES_REMOTE_PORT,retry,interval=.2,forever &
+    start_socat "$RS485_SLAVES_REMOTE_IP" "$RS485_SLAVES_REMOTE_PORT" "/tmp/vcom1" || {
+        echo "Failed to start socat for slaves"
+        exit 1
+    }
 fi
 
-# start the script
-echo "start the script"
+echo "Starting the script"
 
-# run the script in background
+# Run the Python script in the background
 python fetch_bms_data.py > /proc/1/fd/1 2> /proc/1/fd/2 &
 
-# wait for the script to exit
-wait $!
+# Wait for any process to exit
+wait $!

src/fetch_bms_data.py

@@ -87,7 +87,7 @@ def get_config_value(var_name, return_type=str) -> int | float | bool | str | No
 
     # Logging setup and config
 
-    logging.basicConfig()
+    logging.basicConfig(format='%(asctime)s %(levelname)s: %(message)s', level=logging.INFO)
     logger = logging.getLogger("SeplosBMS")
 
     if get_config_value("LOGGING_LEVEL").upper() == "ERROR":
@@ -108,12 +108,19 @@ def get_config_value(var_name, return_type=str) -> int | float | bool | str | No
     MQTT_TOPIC = get_config_value("MQTT_TOPIC")
     MQTT_UPDATE_INTERVAL = get_config_value("MQTT_UPDATE_INTERVAL", return_type=int)
 
+    def on_mqtt_connect(client, userdata, flags, rc):
+        if rc == 0:
+            logger.info("Connected to MQTT (%s:%s, user: %s)", MQTT_HOST, MQTT_PORT, MQTT_USERNAME)
+        else:
+            logger.error("Failed to connect to MQTT Broker (%s:%s, user: %s): %s ", MQTT_HOST, MQTT_PORT, MQTT_USERNAME, rc)
+
+    def on_mqtt_message(client, userdata, msg):
+        logger.info(f"MQTT message received: {msg.topic} {msg.payload}")
+
     mqtt_client = mqtt.Client()
     mqtt_client.username_pw_set(MQTT_USERNAME, MQTT_PASSWORD)
-    mqtt_client.on_connect = logger.info(
-        "mqtt connected (%s:%s, user: %s)",
-        MQTT_HOST, MQTT_PORT, MQTT_USERNAME
-    )
+    mqtt_client.on_connect = on_mqtt_connect
+    mqtt_client.on_message = on_mqtt_message
 
     # Home Assistant auto-discovery config
 
@@ -1121,30 +1128,34 @@ def read_serial_data(self):
     # fetch battery-pack Telemetry and Telesignalization data
     i = 0
     while True:
-        current_battery_pack = battery_packs[i]["pack_instance"]
-        current_address = battery_packs[i]["address"]
-
-        # fetch battery_pack_data
-        current_battery_pack_data = current_battery_pack.read_serial_data()
-
-        # if battery_pack_data has changed, update mqtt stats payload
-        if current_battery_pack_data:
-            logger.info("Pack%s:Sending updated stats to mqtt.", current_address)
-            mqtt_client.publish(f"{MQTT_TOPIC}/pack-{current_address}/sensors", json.dumps({
-                **current_battery_pack_data,
-                "last_update": datetime.now().strftime('%Y-%m-%d %H:%M:%S')
-            }, indent=4))
-        else:
-            logger.info("Pack-%s:Data not changed, skipping mqtt update.", current_address)
-
-        logger.info("Sending online status to mqtt")
-        mqtt_client.publish(f"{MQTT_TOPIC}/availability", "online", retain=False)
-
-        # query all packs again in continuous loop or with pre-defined wait interval after each circular run
-        i += 1
-        if i >= len(battery_packs):
-            time.sleep(MQTT_UPDATE_INTERVAL)
-            i = 0
+        try:
+            current_battery_pack = battery_packs[i]["pack_instance"]
+            current_address = battery_packs[i]["address"]
+
+            # fetch battery_pack_data
+            current_battery_pack_data = current_battery_pack.read_serial_data()
+
+            # if battery_pack_data has changed, update mqtt stats payload
+            if current_battery_pack_data:
+                logger.info("Pack%s:Sending updated stats to mqtt.", current_address)
+                mqtt_client.publish(f"{MQTT_TOPIC}/pack-{current_address}/sensors", json.dumps({
+                    **current_battery_pack_data,
+                    "last_update": datetime.now().strftime('%Y-%m-%d %H:%M:%S')
+                }, indent=4))
+            else:
+                logger.info("Pack-%s:Data not changed, skipping mqtt update.", current_address)
+
+            logger.info("Sending online status to mqtt")
+            mqtt_client.publish(f"{MQTT_TOPIC}/availability", "online", retain=False)
+
+            # query all packs again in continuous loop or with pre-defined wait interval after each circular run
+            i += 1
+            if i >= len(battery_packs):
+                time.sleep(MQTT_UPDATE_INTERVAL)
+                i = 0
+        except Exception as e:
+            logger.error(f"Error in main loop: {e}")
+            time.sleep(10)
 
 # catch exceptions related to the initial connection to the serial port
 except serial.SerialException as e: