Not sure what's going on--but I have a rule that triggers when my Ring (user app) device changes mode. This had been working well, but in the last day or two doesn't seem to be running.
This event shows it seems to have triggered, but nothing actually ran (this is the only event in the log for this RM App):
It does look like it triggered, but the “Log: ‘HSM…” message didn’t get logged? If you haven’t turned on all logging for the app, can you do that to see where it fails?
How about mode(off) did not in fact represent a state change at that time?
Look at App Status page, Application State for the value "prevState". Show that here. Then push the device through a mode change, and look at it again. Is this a Custom Attribute trigger? If so, in "prevState" you should look for the device id of the device, like this:
Notice the 1689=87.73 That's where it's showing the previous state for device 1689. It would look at that value and the new value (that you can log with %value%), to decide if its changed or not.
So sometimes it is picked up? If so, time to move on. You have something corrupt, and you can spend hours and hours trying to figure it out, and even then probably won't find it.
Please create a totally simple rule with Rule 5.0 that has the same trigger and the same Log action, and nothing else. Push the device through its changes, and see what that rule does.
So. That is rather odd--and somewhat points to what you said (the scary "corrupt" word).
As soon as I added my RM5 rule--the RM4 rule seems to have gotten its mind squared away...
Thanks for the tips--from the user side, it's hard to know if something isn't working due to a bug vs. something getting corrupted. In this case, it had been working perfectly up until one of the 2.2.8.xxx updates and it started getting dementia.
Guess I'll see how it does now--and look at re-creating the rule perhaps if it starts acting ill.