Does anyone know if this has potential to cause issues with the Alexa Custom Skill? Not using the built-in one.
1 Like
Wow, the number of changes is huge! What a great project. Many thanks to all those who’ve contributed.
Nest Protect integration fails after this update.
1 Like
Seems almost every integration in the HACS fails. Nest Protect integration fails too.
ZHA crashed after update:
Logger: homeassistant.config_entries
Source: components/zha/core/cluster_handlers/__init__.py:133
First occurred: 23:41:32 (1 occurrences)
Last logged: 23:41:32
Error setting up entry SkyConnect Multi-PAN for zha
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/config_entries.py", line 387, in async_setup
result = await component.async_setup_entry(hass, self)
File "/usr/src/homeassistant/homeassistant/components/zha/__init__.py", line 122, in async_setup_entry
await zha_gateway.async_initialize()
File "/usr/src/homeassistant/homeassistant/components/zha/core/gateway.py", line 230, in async_initialize
self.async_load_devices()
File "/usr/src/homeassistant/homeassistant/components/zha/core/gateway.py", line 238, in async_load_devices
zha_device = self._async_get_or_create_device(zigpy_device, restored=True)
File "/usr/src/homeassistant/homeassistant/components/zha/core/gateway.py", line 570, in _async_get_or_create_device
zha_device = ZHADevice.new(self._hass, zigpy_device, self, restored)
File "/usr/src/homeassistant/homeassistant/components/zha/core/device.py", line 408, in new
zha_dev = cls(hass, zigpy_dev, gateway)
File "/usr/src/homeassistant/homeassistant/components/zha/core/device.py", line 153, in __init__
self._endpoints[ep_id] = Endpoint.new(endpoint, self)
File "/usr/src/homeassistant/homeassistant/components/zha/core/endpoint.py", line 109, in new
endpoint.add_all_cluster_handlers()
File "/usr/src/homeassistant/homeassistant/components/zha/core/endpoint.py", line 140, in add_all_cluster_handlers
cluster_handler = cluster_handler_class(cluster, self)
File "/usr/src/homeassistant/homeassistant/components/zha/core/cluster_handlers/general.py", line 358, in __init__
super().__init__(cluster, endpoint)
File "/usr/src/homeassistant/homeassistant/components/zha/core/cluster_handlers/__init__.py", line 133, in __init__
attr_def: ZCLAttributeDef = self.cluster.attributes_by_name[
KeyError: 'on_off'
Same here. Home Assistant has removed ALL exposed devices from both Alexa and Google. If you open the Alexa app then you can see the devices because Alexa does not delete devices that have been added. They stay as failing devices. So you can use the Alexa App to look up what you exposed. But it took me an hour to expose all the devices again.
For some reason the default list does not show all domains. Like input_booleans. You can add them with the blue expose entities button. They get listed in the window that opens. And once added you see them in the main window. A bit annoying that not all entities get listed.
Not a very elegant upgrade path. I really like the new UI but I cannot understand why the already exposed entities were not imported.
12 Likes
After update I’m getting this error (integration is not loading).
‘ERROR (MainThread) [homeassistant.setup] Setup failed for custom integration smartthinq_sensors: Requirements for smartthinq_sensors not found: [‘cchardet>=2.1.7’].’
Can anyone suggest something? Thanks
Edit.
Issue solved after re-downloading most recent integration version.
I am using m5 atom echo and voice assist vis cloud works fast. However, my speakers in the m5 dont work, on both devices i own. Does anyone else have this issue?
2 Likes
The Alexa media custom skill works fine it seems
Seems the component depends on a python lib that is not included. Raise a bug with the authors github
1 Like
Also Ecowit stopped working.
Setup failed for custom integration ecowitt: Unable to import component: cannot import name ‘async_get_registry’ from ‘homeassistant.helpers.entity_registry’ (/usr/src/homeassistant/homeassistant/helpers/entity_registry.py)
1 Like
Its still using the old registry methods that were deprecated in Feb 2021. The code needs to be updated to use async_get instead of async_get_registry. Please open an issue with the custom component.
Edit: looks like there is already an issue fail to load in beta 2023.5.b0 · Issue #149 · garbled1/homeassistant_ecowitt · GitHub
1 Like
I just dropped the Roborock HACS integration thinking I’d get the same thing back with map etc…but so far nothing…just start/stop and mop intensity… 
2 Likes
Yup. Did the upgrade and every phone and tablet in the house went mental as all the entities were dropped. Then mental again when I added them again. And now every person is mental as all their routines are broken since dropping the entities also stripped them out of the routines. So they all need to be set up again. Bit of a pain to say the least…it’s year of the voice alright…but all the voices here have an annoying high pitched whine…
12 Likes
same as when the reolink came out, not all can be added in the same release. Only one platform can be added at first.
Updated and my ELKM1 and Notion integrations failed. No amount of restart/reboots seems to fix them and nothing in breaking changes about those. Reverted to 2023.4.5 and all working again. Will update and grab diagnostics on my next try.
I had a strange issue on my Intel NUC running HAOS. I have my data on an attached NVMe, copied there using the HA function in the settings menu.
Right after the update and running the new version, I got this scary message. Not sure what to do:
REPAIR ISSUE
{reference} is a filesystem with the name ‘hassos-data’ and is not the active data disk. This can cause Home Assistant to choose the wrong data disk at system reboot.Use the fix option to rename the filesystem to prevent this. Alternatively you can move the data disk to the drive (overwriting its contents) or remove the drive from the system.
Confirming, all entities vanished from Alexa - and new UI to add them is pretty spooky and bugged.
Also, ONVIF integration stopped working with my Arenti camera, couldn’t get it back to work.
Rolled back for now, as badly I want to try new voice things, it doesn’t worth it.
4 Likes
Now the “old way of using History page” is officially removed:
Earlier there were TWO use-cases:
- Open the page → check History of all exposed entities (defined in “history.yaml”).
- Select some entity → check its History.
Now we have ONE use-case:
- Open the page → select one/or more entities → check its/their History.
There were proposals in Community like “give back a possibility to show pre-defined entities” - but this functionality was simply removed.
This change in the change log states:
Refactor handling of exposed entities for cloud Alexa and Google assistant:
- Existing cloud Alexa and Google assistant entity settings will be migrated from cloud store to the entity registry
So if any of you have to re-register anything with Alexa and Google Assistant, then there should be a bug.
Has any of you made an issue on Github already?
Something funky from beta that hasn’t been closed out yet: