Actually I think .4 was pulled due to ZHA issues. Wait for the next update.
Yes it was
In this case it didnât need a database restore to fix, just roll back the version to .3.
But I take your point.
Iâm curious too. Weâre expected to scour the release notes to identify anything which might impact us before updating. But so many of them tell us absolutely nothing about whatâs changed. Just writing one line like âBump to xyzâ is no help whatsoever.
Basically it should just be considered as âhouse keepingâ ⌠i.e updating version numbers etc., but ofcause that could be do to a prior bugfix, minor changes/cleaning in code, moving the code âto more suitable place in the hierarchyâ , or as you mention moving to future release(for various reasons)
Right, it just give one more to think/wonder about , maybe itâs better to just âoverseeâ these, or maybe instead start counting how many times they are âbumpingâ during each releases
The update from 2023.10.3 to 2023.10.4 âbrokeâ my Shelly WLAN devices. They donât show up in homeassistant anymore.
AND I have the issue that the restore to 2023.10.3 is not working. After starting the restore and the system (odroid XU-4) goes to reboot, HA did not came back. After over 30 minutes I plugged it off and on again. Than itâs booting 2023.10.4 again.
I hope to get a 2023.10.5 soonâŚ
In my mind âbumpâ refers to updating external code/libraries, whereby you should refer to the other developers change logs to figure out what changes or not.
Copying changelogs to your software even though itâs actually changes by the external library developers would feel like taking credit for the changes.
Thatâs how I see this.
How did you restore? Try
ha update --version 2023.10.3
I used the restore function of the HMI. Do you think it is different to CMD?
There is a issue to my Shelly problem: ESPHome UnhandledAPIConnectionError after 2023.10.4 on 32-bit systems using noise encryption ¡ Issue #102473 ¡ home-assistant/core ¡ GitHub
Odroid XU-4 is a 32-bit system, there is issue on that side.
[core-ssh ~]$ ha core update --version 2023.10.3 --backup
Processing... Done.
Command completed successfully.
Thank you for that hint. Executing the restore via command line worked for me.
I did the upgrade a few hours ago from 2023.9.2 to 2023.10.4.
Everything went smoothly!
@d77 I have 13 Shellys form different generations and types and using the native integration (CoIOT on the old ones and outbound websockets for the Gen 2) and everything continue to works as it was.
Thank you all for this awesome platform!
Ok, why do you run esphome on shelly? Shelly has a great integration.
2023.10.5 available, I will wait a little before pushing the button
Having problems updating from 2023.10.4 to 2023.10.5. Continues to say âInstallingâ after about 16 minutes. The log shows the following, in order:
Logger: homeassistant.components.hassio.handler
Source: components/hassio/handler.py:584
Integration: Home Assistant Supervisor (documentation, issues)
First occurred: 3:58:21 PM (4 occurrences)
Last logged: 3:58:21 PM
Timeout on /core/info request
Timeout on /os/info request
Timeout on /supervisor/info request
Timeout on /store request
Logger: homeassistant.components.hassio
Source: components/hassio/__init__.py:556
Integration: Home Assistant Supervisor (documentation, issues)
First occurred: 3:58:21 PM (1 occurrences)
Last logged: 3:58:21 PM
Can't read Supervisor data:
Iâm running HA on a Raspberry Pi 4 in Docker.
For an update, it finally did finish the install (I think) after about 20 minutes. Whew, but not sure if it was successful. Only error I see in the log now is:
Logger: homeassistant.components.hassio
Source: components/hassio/websocket_api.py:138
Integration: Home Assistant Supervisor (documentation, issues)
First occurred: 4:20:16 PM (1 occurrences)
Last logged: 4:20:16 PM
Failed to to call /ingress/session - 'NoneType' object is not iterable
Canât get Local Assist Pipeline from the beginning. Got errors from Whisper addon
ERROR:asyncio:Task exception was never retrieved
future: <Task finished name='Task-29830' coro=<AsyncEventHandler.run() done, defined at /usr/local/lib/python3.9/dist-packages/wyoming/server.py:28> exception=ConnectionResetError('Connection lost')>
Traceback (most recent call last):
File "/usr/local/lib/python3.9/dist-packages/wyoming/server.py", line 35, in run
if not (await self.handle_event(event)):
File "/usr/local/lib/python3.9/dist-packages/wyoming_faster_whisper/handler.py", line 45, in handle_event
await self.write_event(self.wyoming_info_event)
File "/usr/local/lib/python3.9/dist-packages/wyoming/server.py", line 26, in write_event
await async_write_event(event, self.writer)
File "/usr/local/lib/python3.9/dist-packages/wyoming/event.py", line 114, in async_write_event
await writer.drain()
File "/usr/lib/python3.9/asyncio/streams.py", line 387, in drain
await self._protocol._drain_helper()
File "/usr/lib/python3.9/asyncio/streams.py", line 190, in _drain_helper
raise ConnectionResetError('Connection lost')
ConnectionResetError: Connection lost
ERROR:asyncio:Task exception was never retrieved
future: <Task finished name='Task-29831' coro=<AsyncEventHandler.run() done, defined at /usr/local/lib/python3.9/dist-packages/wyoming/server.py:28> exception=ConnectionResetError('Connection lost')>
Traceback (most recent call last):
File "/usr/local/lib/python3.9/dist-packages/wyoming/server.py", line 35, in run
if not (await self.handle_event(event)):
File "/usr/local/lib/python3.9/dist-packages/wyoming_faster_whisper/handler.py", line 45, in handle_event
await self.write_event(self.wyoming_info_event)
File "/usr/local/lib/python3.9/dist-packages/wyoming/server.py", line 26, in write_event
await async_write_event(event, self.writer)
File "/usr/local/lib/python3.9/dist-packages/wyoming/event.py", line 114, in async_write_event
await writer.drain()
File "/usr/lib/python3.9/asyncio/streams.py", line 387, in drain
await self._protocol._drain_helper()
File "/usr/lib/python3.9/asyncio/streams.py", line 190, in _drain_helper
raise ConnectionResetError('Connection lost')
ConnectionResetError: Connection lost
Whisper addon config is :
model: medium-int8
language: fr
beam_size: 1 (I tryed to change value but no changes)
Piper addon config is (mostly default, except language):
voice: fr_FR-siwis-medium
speaker: 0
length_scale: 1
noise_scale: 0.667
noise_w: 0.333
max_piper_procs: 2
debug_logging: true
update_voices: true
Tryed Franch and English but Whiser is listening for almost 75 second before respond âI didnât understandâ
HA is installed via the official generic_x86-64 image on an Intel NUC
Using the latest HA version 2023.10.5
Looks like the MyQ integration is broken again with the latest releases. Looks like Chamberlain messed with their API again. Would you recommend to scrap the MyQ integration and replace it with something else?
It would be not optimal, as I spent money with adding a second MyQ garage door not too long ago.
I installed the Homebridge Docker container with the MyQ Homebridge plugin. At that point the Homebridge Docker and my door were discovered by the HASS Homekit Device integration.
Still cloudy but the Homebridge MyQ plugin doesnât seem to have the same problems as the Home Assistant add-on.
A few extra hoops but this is just a stop-gap until my ratgdo shield shows up.
I finally got fed up trying to keep myq running and switched to the Yolink device (attached it to my Chamberlain opener). No problems since then and yolink supports the HA integration themselves.
In most cases the âbumpâ is directly an action to fix actual issues. The entries in the release notes are the headlines from the PRs and not the issues they solve.
I wish the many devs would think about the release note and give their PR a more meaningful title related to the issue instead of these very little useful bump notes. You have to spend half an hour following referenes to the issues. It does not take extra time for a dev to write a better title on a PR.
Please consider bumping less and write what you fix in the title instead.
Sometimes a PR is just keeping in sync with upstream and then the bump is meaningful