I don’t immediately see the reason this would be the case. I think this is handled around line 475 in the code, and I don’t see any reason this wouldn’t fire twice. It would need a closer look and some debugging output to test the code path down to this section of code, but it is also possible that the device itself doesn’t fire this event if there isn’t a state change between “home” and “away”, in which case there would be nothing we could do. I don’t think I’ll get the chance to debug this very soon, but perhaps you can. You can try adding debug statements in the device handler and watching the “Live Logging” section of the web interface while you try to reproduce the issue. If the device handler never gets entered at all, then it won’t be possible to fix. However, if it does get entered, then something in the device handler is ignoring the second attempt, and we could fix that.