Dear community,
I’m struggling with a strange double event issue that I’m hoping you might be able to point me in the right direction to resolve or mitigate:
**SUMMARY:✱
I click my apple Watch complication to open my garage door, it shows that it recognized the tap (by showing the “working spinner” for a second), but nothing happens. So I press the complication button a second time and the garage door opens. However, 13 min later, the garage door opens “by it self”, presumably because one of the taps finally got processed. I’m trying to figure out how to avoid this delayed double event in the future.
**SW VERSIONS:✱
iPhone 11
iPhone 17 pro
iOS Home Assistant Companion App 2026.2.1
Installation method: Home Assistant OS
- Core 2026.3.3
- Supervisor 2026.03.2
- Operating System 17.1
- Frontend 20260312.0
**CONFIGURATION:✱
- Script “Toggle Garage Door”
- Google Garage Door Shelly switch_0 [Garage Door Shelly]
- NOTE: I’m using a “Shelly Plus 1 UL” as a relay for the garage door
- iOS Home Assistant App → Settings → Apple Watch → Configuration → Toggle Garage Door (i.e., calls the above script)
**ACTION:✱
- On the apple Watch I click the Home Assistant Complication → Then I press on the “Toggle Garage Door” item. This shows the “working spinner” next to the item.
- Nothing happened for some reason, so I press it a second time and the garage door opens.
- I left the garage in my car, but importantly I closed the door using the Apple Home integration via CarPlay, so the execution for the close was a different path than above.
- The garage door reopens seemingly by itself approx 13 min later.
**ADDITIONAL DETAILS:✱
The “Trace timelines” from the Toggle Garage Door script shows slightly different sequences as soon below:
First trace at 1:11:19pm
Second trace of the delayed garage door opening “by itself”
**QUESTIONS:✱
- Where is the “script.open_garage_door” shown above? It’s not in my Scripts list, so I’m confused what that even is.
- Should I be doing something to flush a queue somewhere in case this happens in the future? Presumably there was an issue somewhere between my watch and my phone, or between the phone and the HA server that caused the delay. However, it’s very interesting that the first trace is different than the second trace, so I think it gives clue as to what’s going on. However, I don’t have a good understanding of how the HA system queues different requests, so I’m feeling a little stuck.
THANKS ALL!

