Last_Known State

[ElementZoom]

Hi, I notice that when I do a power cycle or rebooted the MSR-2, the ltr390_light always returns to unavailable and won't be available until I shine a light to it, or open the room light. Can you please add last_known state and apply it to the ltr390_light so it will still read the previous state after the power cycle / HA reboot?

Thanks

[bharvey88]

Hi, I notice that when I do a power cycle or rebooted the MSR-2, the ltr390_light always returns to unavailable and won't be available until I shine a light to it, or open the room light. Can you please add last_known state and apply it to the ltr390_light so it will still read the previous state after the power cycle / HA reboot?

Thanks

Can you confirm if you have the "reduce db reporting" toggle on? I think this might be related to the logic we're using with that enabled.

[ElementZoom]

Hi @bharvey88 yup I l have it turned on. Do I need to turn it off?

[bharvey88]

Hi @bharvey88 yup I l have it turned on. Do I need to turn it off?

Can you confirm that the issue goes away with it disabled? That way we can focus on improving that yaml after a reboot - i think it is "working as expected" because it's waiting until a change of 20 or higher lux is seen before it sends the update but we should be able to edit the yaml to make this behavior change a bit for edge cases like a reboot.

[ElementZoom]

Hi @bharvey88 yup I l have it turned on. Do I need to turn it off?

Can you confirm that the issue goes away with it disabled? That way we can focus on improving that yaml after a reboot - i think it is "working as expected" because it's waiting until a change of 20 or higher lux is seen before it sends the update but we should be able to edit the yaml to make this behavior change a bit for edge cases like a reboot.

Yeah I'll check when I get back home later. I've turned that off for now

[ElementZoom]

I think it works. Haven't received any failed automations for now.

[bharvey88]

Thanks - I am closing this ticket but I will talk to Trevor about improving the reduce_db_reporting yaml to account for reboots - likely this filter can be edited to work better for edge cases such as this.