I now have this running in a docker too. Except for the pushing to HA part. I set the config.ini to use mqtt-homeassistant as reporting method. It pushes to mqtt and publishes in homeassistant/sensor/flower3/state, but it doesn’t get picked up by the mqtt integration, do you see this too?
What exactly do you mean by this?
I added:
- /var/run/dbus:/var/run/dbus
in the docker-compose file with the HA container.
I now don’t have the Error 2 anymore as mentioned above, but now I have the following:
Unable to fetch switchbot services data
I’e found that I need to update my RPI_GPIO HACs integration after a core update. Then reboot. it works fine again.
Yes your system should tell you that, which makes the whole thing really easy.
Not related to this HA 2022.7, but it is a common UI issue.
Within a few last months we observed changes in Menu, moving a “Restart” button, etc.
I propose a feature which someone may find very useful - scrollbars in Dev tool → Template.
Please have a look.
I have a E1810 that is not working. I can’t get it to pair now and seems to can’t finish a reconfigure.
Please open a new github issues and provide py-spy How to install Py spy on a HA OS instance, please instruct - #44 by Mariusthvdb, MySQL slow query logs, and callgrind.out files from Profiler - Home Assistant
I have broken templates and can’t map an ‘unknown’ attribute to 0 in 2022.7 like I could in 2022.6.
The original template appears appears to be mapping unknown states to 0 with the map('int') filter. I am trying to get a count of batters that are low, unknown or unavailable.
{{ expand('group.aqara_batteries', 'group.shelly_batteries') | map(attribute='state') | map('int') | select('le', 0) | list | length }}
I have tried map(attribute='state', default=0) but that fails because the state exists but is unknown.
I have tried piping it to int(default=0) but that changes it from an array that exists to a scalar.
I have tried map('int(default=0)') but that is an invalid filter.
I have tried replace('unknown', '0') but that treats the input array as a string.
Don’t know if I found a bug or was exploiting an undocumented feature. Either case, I haven’t found a solution.
Wes.
{{ expand('group.aqara_batteries', 'group.shelly_batteries') | map(attribute='state') | map('int', 0) | select('le', 0) | list | length }}
Supply the int filter with an explicit default value (see how in the example I posted). Without it, when it encounters the unknown value it will no longer automatically supply a default but generate an error.
1 Like
Thanks. Didn’t think of that syntax.
Wes.
My Stunning performance is severely frustrated lately by daily
Anyone else seeing this more than before? No clue what it could be yet…
What’s more , it’s only happening on my system using Nabu Casa Cloud… my duckdns systems never see this.
On mobile_app that is.
A whole day trying to fix my ha after updating to 2022.7.3 !
Very frustrated! The fan of my argon one case not working anymore. Lot’s of errors…
Also had a look at all entities… It’s a complete mess! When you select ‘all entities’, a lot ‘old’ entities appear. I found out when I wanted to rename a few, and got errors… Deleted them in de files found in the .storage folder, but they keep coming back.
Performance? Probably because there’s a performance increase because of python 3.10.
I still have a slow startup time…and. Errors everywhere… Even the new 'music assistant ’ is giving errors.
So frustrated…
Yup, I agree, ive been spending two evenings in a row to get everything working again. Primarily BLE integrations are a mess now. I knew Miflora was going to be a issue, but Switchbot wasn’t on the list. I simply can’t get it to work, it just doensn’t. Got a Pi4 running with docker, added dbus and everything but just doesn’t. Finally managed to get things working by using switchbot-mqtt.
What auth method are you using? If you auth using trusted_networks, the mobile app will do this everytime you go on a public network. You must do the auth using username/password when setting up the app or it will do this.
yeah, aware of those restrictions. thats why I always use username/password. But, most surprisingly, this happens all the time when on my home network…
127.0.0.1 indicating it is from an integration inside HA maybe? why not simply use the phones IP address
Every time I’ve had those errors it has been fixed by logging out, then back in again on my wall tablet which runs the HA app.
yeah, well that can’t be can it, because that specific device gets banned, so we need to delete the ip-bans.yaml, and restart before being able to re-log in.nThat’s why it’s such an annoying bug.
Before, I used the trick to toggle my VPN and thus change ip-address, but somehow that doesnt work anymore. Seems the App is clever enough to realize it’s still the same device 
\\homeassistant\config.storage\core.entity_registry was corrupted twice as a result of an upgrade from 2022.6.7 to 2022.7.x (I think I tried .2 and .3, but can’t remember for sure). Replacing a version from a backup from 2022.6.7 fixed one issue, but I could not figure out from the logs what else might have gone wrong as this only fixed the issue temporarily. I’ve been holding off updates since then reason.
I’m planning to upgrade again to 2022.7.3 today or tonight my partner is not at home, so it’s a good moment for a maintenance window . I will post here how the update went.