New core integration Alexa devices

Configuration
21

The Custom commands feature has not been implemented yet. When it is, it will allow you to emulate a voice command with a script action. The example you have referenced is just that… a way to get the same action out of an Alexa device as saying “Alexa, Play B.B.C. on Tunein”.

1 Like
22

Well, the functionality is currently limited. Many of us were hoping for more control, especially of devices connected to Alexa. Hoping to see this grow and add more features, eventually replacing the Alexa Media Player in HACS. Would be great to be able to control things like Echo Glow lamps for use in HA automations. Looking forward to seeing this mature.

2 Likes
23 
Logger: homeassistant.config_entries
Source: config_entries.py:749
First occurred: 8:18:57 AM (1 occurrence)
Last logged: 8:18:57 AM

Error setting up entry [email protected] - amazon.in for media_player
Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/config_entries.py", line 749, in __async_setup_with_context
    result = await component.async_setup_entry(hass, self)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/homeassistant/homeassistant/components/media_player/__init__.py", line 497, in async_setup_entry
    return await hass.data[DATA_COMPONENT].async_setup_entry(entry)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/src/homeassistant/homeassistant/helpers/entity_component.py", line 219, in async_setup_entry
    raise ValueError(
    ...<2 lines>...
    )
ValueError: Config entry [email protected] - amazon.in for alexa_media.media_player has already been setup!

The integration page shows

Needs attention

No devices or entities

Failed setup, will retry: ‘networkDetail’

Getting this error while trying to setting alexa integration.

Kindly guide.

24

I gave this a go today to see if it worked any better than the HACS AMP which is fundamentally broken at the moment with regards to the “last called” function, all “get history” requests are bounced with timeout/too many requests errors.

I am in the UK.

This integration reports that it failed to setup, will retry… I turned on debug and saw that it is actually logging in fine, and loads of device info is coming back (far too much sensitive data in it to post the whole thing), but right at the end of each burst of activity when it retries is this:

2025-07-04 10:23:41.377 DEBUG (MainThread) [aioamazondevices] GET request: https://alexa.amazon.co.uk/api/phoenix with payload None [json=False]
2025-07-04 10:23:41.441 DEBUG (MainThread) [aioamazondevices] Cookies from headers: {}
2025-07-04 10:23:41.441 DEBUG (MainThread) [aioamazondevices] Response 429 for url https://alexa.amazon.co.uk/api/phoenix with content type: application/json
2025-07-04 10:23:41.442 DEBUG (MainThread) [homeassistant.components.alexa_devices] Finished fetching [email protected] data in 1.569 seconds (success: False)
2025-07-04 10:23:41.970 ERROR (ImportExecutor_0) [homeassistant] Error doing job: Unclosed client session (None)
2025-07-04 10:23:41.977 ERROR (ImportExecutor_0) [homeassistant] Error doing job: Unclosed connector (None)

429 being a “too many requests” error. So it seems that this is also now broken by whatever Amazon did a week or so ago…

25

Yes,

My Dutch config isn’t working 2 right now.
Check: Alexa Devices Integration - #13 by WarLion

It’s a complex challenge.

26

Since my last password change from my Amazon Account I get the following message in this Integration: Einrichtungsfehler, wird erneut versucht: ‘NoneType’ object is not subscriptable

Remove the Integration and readd it doesn’t fix the issue. Any ideas?

27

Its a known problem. There is a fix and it has been merged, but not yet released.

28

While some people are willing to wait for core releases to fix broken integrations, sometimes integrations are essential in smart homes and people require immediate fixes. @wmaker is correct, the fix is completed and merged, which leaves you a few options.

  1. Wait for the next core release

  2. SSH into the HA container, update the manifest.json for alexa_devices to the latest release, and run pip install aioamazondevices==(release version). Latest release is currently 6.4.0.

  3. Install the latest release as a custom component.

I typically do #2, but I’m comfortable with CLI. Most people tend to do #3 for it’s simplicity. However, #3 requires that you delete your custom component before you update to the next core release. Whereas, any core update will override any changes made by doing #2.

1 Like
29

I had the same problem and looks like HA 10.2 which just came out fixes the problem.

While it seems to be working, I keep getting a bunch of these errors in the log (occurs once every 30 seconds or so):

Logger: aioamazondevices
Source: components/alexa_devices/coordinator.py:58
First occurred: 11:10:59 PM (66 occurrences)
Last logged: 11:43:30 PM

    Malformed endpoint data received: {'errors': [{'message': 'Too many requests', 'path': ['listEndpoints'], 'extensions': {'errorCode': 'UNOFFICIAL_CLIENT_TOO_MANY_REQUESTS', 'isRetryable': False, 'classification': 'DataFetchingException'}}], 'data': None, 'extensions': {'requestID': 'PZZX2ARKB6WST4GN3X50', 'duration': 15}}

I tried reloading the integration and also reconfigure (re-authenticating my amazon account + OTP) but it doesn’t seem to help.

What may be the problem?

Thanks

1 Like
30

The developer thinks its an issue with Amazon’s backend.

1 Like
31

Thanks! Also saw your comment in that github thread about temporarily disabling the logging until the issue is resolved. Very useful!