Is there anything I’m doing wrong or is assis: xyz on or xyz off really not working and I always have to (type/say) turn xyz on, turn xyz off? Or only a problem with German language? Example domain: light.
I’m really not used to use this longer version via Alexa and to be honest, it would be better to have it as short as possible as well in internal HA assist.
Were there any power level tuning for bluetooth in this release?
I had zigbee and BT adapter connected to the same hub, very close to each other and I know - not ideal. Up to this release both radios were working OK, but after the upgrade the zigbee can’t create the network if the BT dongle is also plugged.
Indeed a great integration, and the possibility to add static price components would be a good improvement! I’'m new here, so what is te right way to aks for such a improvement?
After database migration, the table “states” no longer have a date value available in columns last_changed or lastupdated, so my sql max/min serach no longer works: SELECT MAX(cast(state as decimal)) AS value FROM states WHERE entity_id='sensor.temperatura_giardino' AND state!='unknown' AND datetime(last_updated) >= datetime(date('now'));
How to re-enable it?
As strange it may seem, it worked. I guess that’s because date(now) without stating hour/minute was considering today at 00:00, so everything greater than that is today’s value…
Thanks, I’ve changed it to SELECT MAX(cast(state as decimal)) AS value FROM states WHERE entity_id='sensor.temperatura_giardino' AND state!='unknown' AND datetime(last_updated_ts,'unixepoch') >= datetime(date('now'));
Since updating to 2023.2.2 Home Assistant runs for a couple of hours and then stops responding.
I have to unplug the hardware to get it running again.
Any Ideas.
I did also update to the latest Zigbee2MQTT recently, but I think the instability started only after HA update.
Any ideas?
Yes, I have looked in the logs, but they only start with the HA restart.
Haven’t looked at the file system if there are older log files yet.
There is a MQTT error that is logged every 80 milliseconds which I have posted in the integrations thread already.
What a wonderful release. Updating the MariaDB tables took forever, and there was no telling when it was done.
Even better was that the recorder just broke for the entire period, even after the updating was done, and it took a restart of HA to bring it back and start logging things.
Great stuff.
If you have something writing to the log file every 80ms, I would imagine it could very quickly either swamp the bus of fill your storage (especially if you use an SD Card)
Are you able to edit your configuration to reduce the logging level to ‘critical’ to prevent the log spam until the issue is resolved?
If you’re using PostgreSQL in your SQL sensor use SELECT TO_TIMESTAMP(last_updated_ts) AS last_updated to extract human readable date from last_updated_ts column.
i don’t know if you encountered same issues as i, but few updates ago (well actually on beta) there i only had to say “turn on/off hallway” … Now i have to say " turn on/off hallway Light" , so it seems like they have added “domän/” as “required” in the request
Edit: which to some degree makes sense … i.e (turn on hallway. Light/ run hallway. Automation/Scene " or turn on hallway. vacuum/temperature/heating/xxx/yyy )