Trigger when device no longer provided by integration | Home Assistant | Page 1

lucid turret Oct 21, 2025, 3:56 PM

#

Occasionally the Venstar Explorer Mini stops responding to HTTP. I can ping it, but web API calls fail. I tried to create a watchdog automation, triggered when the thermostat entity goes to Unavailable for 5 minutes. But when the situation happened, the automation did not run. It should have triggered earlier today, but didn't. I didn't notice since there were no notifications. Then I took the HA update to 2025.10.3. I just happened to poke around in the devices and noticed the thermostat was unavailable. Looked at the log and saw the failure to connect which is usually logged when the Venstar goes off in the weeds like this. The automation probably should have triggered again when HA was up for 5 minutes - although the state technically didn't change, it was unavailable from the time HA started running, so maybe not. Power cycled it and it came back. But I'm annoyed that the watchdog automation didn't work. Any recommendations?

trigger: state
entity_id:
  - sensor.thermostat_thermostat_temperature
to: unavailable
for:
  hours: 0
  minutes: 5
  seconds: 0

merry marsh Oct 21, 2025, 3:58 PM

#

in the future, please format YAML properly

red frigateBOT Oct 21, 2025, 3:58 PM

#

To format your text as code, enter three backticks on the first line, press Enter for a new line, paste your code, press Enter again for another new line, and lastly three more backticks.
```yaml
example: here
```
Don't forget you can edit your post rather than repeatedly posting the same thing.

merry marsh Oct 21, 2025, 3:59 PM

#

As you pointed out, it depends on whether the state actually changed to unavailable.

#

it probably started as unavailable and there was no state change

elfin harbor Oct 21, 2025, 4:02 PM

#

lucid turret Occasionally the Venstar Explorer Mini stops responding to HTTP. I can ping it, ...

you could use Uptime Kuma (as addon or running separately somewhere then integrated), one of its options is to monitor http being up and responding. which would then give you an entity via the integration.

lucid turret Oct 21, 2025, 4:02 PM

#

It must have been available at 8am when my "daytime" temp targets were applied, or it would have been a lot colder this morning. So unless it happened to go unavailable at the same time the upgrade restart happened, it seems like a transition was missed. I suppose that's possible ...

merry marsh Oct 21, 2025, 4:04 PM

#

you can see all the state transitions and timing in the logbook, and easier when you pull it up from the entity dialog

lucid turret Oct 21, 2025, 4:06 PM

#

All history for the device prior to the upgrade has vanished.

#

Oh I forgot to mention a step. After power cycling the thermometer it still wasn't in HA until I reloaded the integration.

#Trigger when device no longer provided by integration