2023.2: How can I Assist?

I was not aware of such issue, sorry (I normally skip version .0 to account for these type of issues).
I think the ‘bug’ is that the DB is busy with the migration so it can’t insert (or it can’t be read) to retrieve non-streamed values.

The challenge is, I understand, I can’t attempt any rollback/backup restore until the DB migration finishes in order to avoid DB corruption.

  • Will there be a log message notifying the migration is completed? Or otherwise how can it be known by a user?

Thanks for the quick reply!

I can’t find when the map card got timestamps. I can’t find it in the changes on this or the previous release but I’m quite sure this is something new.
Good work whoever did that! I have wanted this for a long time.

1 Like

Nope, the bug is related to unit conversion entities and it’s entirely a frontend issue.

If I recall correctly, it shows up in the logs at some point.

after the latest updates, I can no longer get home assistant to work on android 7.1. I had 2 wall tablets, and it doesn’t work anymore. is it no longer possible?

What exactly means „does not work“? Did you update to 2023.2.1 already?
It fixes at least an issue for older android browsers and fully kiosk browser…

According to a response in my own thread a problem with the Synology NAS integration has already been identified? I’ve hit it today as soon as I update to the latest Core.

When I revert to the previous version all problems go away.

I’m choosing to simply bypass this release, but in case it may help someone’s investigation, here’s the link to my thread that includes the log output:
https://community.home-assistant.io/t/synology-nas-integration-fails-after-update-to-core-2023-2-1/529653

It seems the migration finished.
The graphs are loading the data from these past hours so I guess is not the same bug, but indeed related to the migration.
Replying to my past self, I saw no logs of it being finished, but the notification that poped up in the notification tray, disappeared. So that would be it. (i.e. don’t dismis the notification and let HA do it so you’ll know when its done!).

1 Like

Petro, with the latest update that came out on 02/03/2023, it is now working.

I don’t know if these short intervals (i.e. show me the last ten minutes of history) are a new option for the 2023.2.x release and nobody knows how to configure it in the history card or should I better ask this question in the configuration/frontend forum because it works already in older versions and I have overseen something?

I’m also in the same situation. I’m usign MariaDB on Synology NAS DS216j and the last avaiable version is 10.3.29. Maybe it will be only an assumible performance issue because Recorder documentation still say it is compatible with MariaDB >= 10.3. But I’m still unsure to update to HA 2023.2.

Hi,
In my case i had to remove the surveillance station in the syno nas to have the dsm integration configured.
Something has changed probably .

I took the plunge and updated but I only have a v small db. Upgrade was smooth. No issues so far

Hi,

Has anyone experienced issues with icon colours with the stock Light Card with 2023.2.x?

Specifically, I have a Hue group and Tuya lights not showing the icon colours they previously did with respect to the lights on/off/brightness.

This is an example from 2023.1.7

image

And this is the same from 2023.2.x

image

both are using the default HA theme.

Kitchen - Dining is a Hue group and in 2023.1.x works as expected, but in the second screenshot it is black (in dark mode it is grey)
Lounge & Porch - Wall Lights is a tuya light
And interestingly Lounge & Porch - Chair is a hue single bulb which behaves as expected!

Using an Entities card or a Button card the entities’ icons behave as expected i.e as per the 2023.1.x screenshot.
Thanks

BTW the sink, oven, fireplace, island are all Magic Home LED strips which work as expected.

1 Like

Yes I see that too. Black is the new yellow :slightly_smiling_face: Shrug.

Just a short question. How long does the database migration of 2023.2 take?
I recently upgraded and did a reboot while migrating and i think my database was corrupted.
Now i will wait some more but i dont know if i get a succesful message or something like that.

Any way to identify the finished state? This is the only thing i see in logs: Database is about to upgrade from schema version: 30 to: 33

Also RAM usage is pretty high with 90-95% on my VM.

Same here. I’m stuck with MariaDB version provided by Synology. I’m afraid that issues would happen and so , I’ve removed the recorder DB from MariaDb and I’m back to the HA disk :frowning: .
If this is a wait until Synology updates the packege, Im afraid it will take time …

Seems like something worth revising in the release notes, eh?

And yes when I first read it I wondered how it would effect last_changed and last_updated, and if there would be less database entries if many small changes were less than the rounded precision.

If you use a custom theme (or card) then it needs updating.

If you use the default theme and a core card and this is happening then it is a frontend bug. Please report it.

Isn’t it in the esphome doco what to change to

The repair popup links to the documents just as you did.
Can’t see anything wrong with this.

For me it exchanged the encryption key by itself.
Perhaps that is because I had to upgrade my node also?