I’m seeing a pattern when I edit a single scene and save it, all scenes and several entities become unavailable. Logbook shows it as one big blast in my logbook. This ends up triggering other things and makes a big mess. I think part of that is due to so misconfigured automations that will trigger on any state of a scene and I need to fix that. However, that does not explain the root issue with this “became unavailable” for about 100 things. Among the entities impacted, it spans multiple technologies (zwave, companion app, scenes, hue, etc.).
Yeah, I understand how scenes are applied when editing them in the UI, but that is unrelated here. I was editing a scene that was already active. Unrelated scenes and entities were thrown into an unavaialable state when I saved the scene I was editing.
It’s reloading the Scene integration; all scenes are affected.
Thou this actually sounds like a Bad Implementation ! , Maybe this should be change into i.e,
“Save Scene” “restart core” “apply changes”, like some other integrations !
After looking at the logs again. I see the scenes are getting set to unavailable and then immediately being set to the date/time they were last called. Does that happen with ever scene change?
I fond a Node-red node that was set to trigger on one of these scenes and ticked the boxes to ignore the trigger if previous state was unavailable.
It happens when the integration is reloaded.
For future reference, in an automation using a State Trigger, you can use the not_from
option.