So I have a rule that turns lights off with no motion for 8 mins and to cancel on truth.
It works fine, however I noticed it didn’t run the last time somebody left the room. It was almost an hour and the light never went off. It’s like the rule never ran??
I'm trying to do this as well, but perhaps slightly different. I have a big room with one light, so it takes three motion sensors to cover it.
I tried Motion Lighting Apps, but since I don't want motion to turn the light on, this doesn't work (it looks like it might, if you leave the "turn light on" unpopulated, but it never worked for me).
So now I'm on Rule Machine, Triggered Rule which seems nice.
I have trigger events: light turns on.
Conditions: all motion sensors ALL inactive.
a rule gets created here. Not sure why.
Actions for True: turn off that initial light (it's a switch/dimmer, but I call it switch in both places), after a delay of 2 minutes. (something I haven't been able to test properly is, if any of the three motion sensors goes back to active in these 2 minutes, I presume the rule is STILL triggered, but now the action will NOT happen... once the conditions are met again (ALL inactive) the two minute timer starts again? I hope?)
So this happens to work every other time I've tested it. The times it does not work, the STATUS of the light in the devices and dashboards turns OFF, but the light itself is still ON. The logs show this "happens" too. I haven't set up sniffers here yet, but I'm guessing this one attempt to send this OFF message fails every other time (I have noisy networks). It seems there has to be some solution for this? I think I have seen somewhere a "retry" check box somewhere but I don't see that here.
This light is a Generic Z-Wave dimmer. I guess I'll try it now on a Zigbee one and see if that works better. It seems the system should try and resend messages that aren't ack'd or better, do an instant read back and confirm/resend? OR am I just missing something? As I said, my networks are very noisy but it does seem like I'm seeing this sort of problem all over.
With a triggered rule, when it is triggered, the conditions are evaluated once and only once. so the motion change will not be seen. I think you'd be better off using a regular rule where evaluations are made with each change in your defined conditions.
Thanks @bjcowles... that addresses my parenthetical. I am rewriting the rule now. I agree, this feels better already "code" wise.
However, as I'm writing it, I'm sitting under the light that is on, but the rule as I described it above has apparently run so the HM thinks it is off. Clearly the rule won't work now... I'll finish rewriting and then toggle the lights and motion sensors and start testing again but I think my original question still stands...
As far as I can tell when I use drivers from Hubitat, I should not need to poll... (right?) but statuses get out of sync anyway. What is the right solution for that?
Well, that depends. What is the make and model of the dimmer. Z-wave plus devices should report their status immediately. Many older z-wave dimmers (non-plus) do not and need to be polled. Hubitat has a built-in app called Z-Wave Poller just for these older ones.
Hi, thanks again. It's a Jasco/GE. I know the situation you describe well. I thought this was a Plus, but looking at the product ID I guess it is not (0x3031 vs 0x3038). I've added it to the poller and I'll see if this improves things. So a device that pairs as a Generic Z-Wave Dimmer, and is a Z-Wave Plus device will be configured to report?
As I've been sitting here with the non-triggered rule running, the light has been turning off randomly. And now with the polling in there, and I'm watching the motion sensors in a dashboard, the light still went off, so something else is wrong here. It feels like re-triggering of one of the motions does not stop the two minute timer from triggering.
Thanks again... so "smart" means it is Z-Wave plus? Is this true for "GE Smart Fan Control"? I'm pretty sure this is not true... a number of mine are old and this is how they are all showing up.
It looks like this might be right for the dimmers and switches. Thanks!