Platform esphome does not generate unique IDs error

[khughes5]

Hi Guys,

Again, sorry if this is another newb question/problem but i have been trying to resolve issues with my installation, and can't for the life of me figure it out.

Control of my AC works through the website instantly

however control via HA was really hit and miss for me.. for example if i use the climate control panel item with the system in an off state, change mode to "Cool", it thinks about it for a couple of seconds and then resets back to off. During the last 2 weeks i have probably had it work for a total of 1 day.. but that was fleeting..

I started checking logs and found this repetitive error message:

Platform esphome does not generate unique IDs. ID D4:D4:DA:9D:0D:8C-select-homeac_mode is already used by select.esp_samsung_ac_homeac_mode - ignoring select.esp-samsung_ac_homeac_mode

My ESP at the time was called esp_samsung_ac and my climate name was "My Home" with the sensors named HomeAc_temp, HomeAC_target, HomeAC_mode etc.

In the past, i HAVE renamed the ESP device (usually by creating a new device in ESPhome, editing the config then re-uploading via OTA), and these error messages seemed to indicate it remembered the old configuration.. I even tried removing all the old records from .storage\core.entity_registry.. as well as trying to rename the device and sensor names again..

To make things more insulting.. i gave up today and created a whole new HomeAssistant installation from scratch, installed ESP-Home, then created a new device, copied my old yaml contents in (using the newly generated encryption key), created a manual firmware.. then uploaded it through the OTA on the web interface... I also then had to adopt the device into ESP-Home in discovered devices.. Somehow, this still results in the error message:

now my ESP is called "ESP-samsungac", my climate called "SamsungAC" and sensors as "AC_temp", "AC_target" etc.. My Climate controls still do not function reliably.. i really don't know what else to do.. short of throwing away the M5 stack i have it installed on and trying everything from scratch again.. but this seems stupid :(

here is my config yaml:

esphome:
  name: esp-samsungac
  friendly_name: ESP-SamsungAC

substitutions:
  name: esp-samsungac

esp32:
  board: m5stack-atom
  framework:
    type: arduino

web_server:
  port: 80

# Enable logging
logger:

# Enable Home Assistant API
api:
  encryption:
    key: "xxxxxxxxxxxxxxxxxxx"

ota:
  password: "73571d78e2b33fbbee49a53066f0436b"
  safe_mode: true
  reboot_timeout: 10min
  num_attempts: 5
  on_error:
    then:
      - logger.log:
          format: "OTA update errorr %d"
          args: ["x"]

wifi:
  ssid: !secret wifi_ssid
  password: !secret wifi_password

  # Enable fallback hotspot (captive portal) in case wifi connection fails
  ap:
    ssid: "Esp-Samsungac Fallback Hotspot"
    password: "xxxxxxxxxxxxxxxxxxx"

captive_portal:

uart:
  tx_pin: GPIO19
  rx_pin: GPIO22
  baud_rate: 9600
  parity: EVEN
  #debug:
  #  direction: BOTH
  #  dummy_receiver: true
  #  after:
  #      timeout: 200ms
  #  sequence:
  #    - lambda: UARTDebug::log_hex(direction, bytes, ' ');

external_components:
  - source: github://lanwin/esphome_samsung_ac@main
    components: [ samsung_ac ]

samsung_ac:
  devices:
    - address: "20.00.00" # for non NASA it sould be like "c4"
      # Each property below are optional - choose what you want to use.

      # You could use the ESPHome/Homeassist climate control
      climate:
        name: "SamsungAC"

      # And/or use seperate controls/sensors
      room_temperature:
        name: AC_temp
      target_temperature:
        name: AC_target
      power:
        name: AC_power
      mode:
        name: AC_mode
        
      # Only supported on NASA devices
      room_humidity:
        name: AC_humidity    

# DEVICE INFO
text_sensor:
  - platform: wifi_info
    ip_address:
      name: "${name} IP address"
      icon: mdi:ip-network

sensor:
  - platform: wifi_signal
    name: "${name} signal"
    update_interval: 360s
  
  - platform: uptime
    name: "${name} uptime"
    update_interval: 360s

button:
  # Restart the ESP
  - platform: restart
    name: "Restart ${name}"

[lanwin]

Thank you for your detailed report. Well I guess this could also explains this issue #6 (comment) .

I will look into this issue soon.

Can you please edit the issue and wrap the yaml part into a code Block? Look in the edit tool bar for <>.

[lanwin]

A did not noticed this is a discussion. This is clearly an issue. Could you convert it to one?

[khughes5]

Done! Sorry I wasn't confident enough that it wasn't something I had screwed up while messing with HA.. I am about 3 months in.. massive learning curve 🤣