In general it isn't happening because home automation events aren't happening at super fast speeds. How quickly can you open and close a door? Not 5 msecs.
You didn't really answer my question: Could the variable that triggers the rule (LED_Refresh) have been set as a consequence of the rule running, or as a consequence of the exact same events as triggered the rule (e.g., some other rule sets it)? It's mighty odd that happens 5 msecs later than prior instance. If this is truly a random event, that's awfully suspicious timing. You wouldn't be able to reproduce it. But if it is induced, then you can change your rules to fix this.
I have four apps for HVAC that flip it which are clean.
One other app has a lot of device triggers so I suspected it.
I had added a manual button to test with, but the button also triggered a light, so I was getting two very close together.
I fixed that and I can no longer cause the issue with my button,
BUT, I canât find anything now that would have caused issue without my button. But it was happening.
So I have added log messages in the five rules that change the variable and will wait for another occurrence.
Thanks, Bruce, for the explanation, but I still think that my âgoodnightâ or âleavingâ rules could generate enough changes rapidly to maybe be an issue.
Lots of events isn't a problem. Circularity of logic is a huge problem, and totally self inflicted. I don't know if that's what your problem was, but it sure looks like it.