That log is before I upgraded to 2.2.3
No problem. Just installed v2.2.3.118 without issue. I'll keep an eye on it tomorrow and see what happens.
I did notice something strange. In preparation for removing dev:2 "Kitchen", I removed it from 2 rules, and the Alexa interface. However, when I open the device page for "Kitchen", it still says it is in use by one rule. But it's not in that rule's action - I explictly removed it earlier.
(Bedtime is a virtual switch. Activating it through Alexa turns off most of the lights in the house.)
My other oddity today. The hub logged a system event for sunset. But my automation trigger for sunset-20 minutes did not execute to turn on the exterior lights. Weird...
There's code new to 2.2.3 that, ironically, was supposed to fix sunrise event not firing. The community reported a bug affecting time calculation for rules like the one in your example, although we believe it's limited to the day of hub reboot. We'll have a fix shortly.
3 Likes
Hub is still going strong, no issues for two days solid on v2.2.3. Other than my wife asking me why the kitchen wasn't turning on/off with the other lights in the house, lol.
I just added the offending dev:2 back to Alexa and my automation rules. So that'll be the final test...
1 Like
I'm calling this one "Solved". It's been nearly 2 days since adding the offending device back into my rules, and everything is still working. Thanks for the help!
3 Likes
Ok, this may not be related to my previous issue, but I'll add it here for continuity. I've had the hub lock up again the last 2 nights, but it seems to be a token issue with the amazon/alexa API. Both times, I realized it because Alexa voice commands didn't work, and rule automations did not fire. Then I found I was unable to connect to the hub web page, and had to reboot it using the diagnostic interface. You can see the reset was completed each time when the Lutron telnet session is re-estalished.
Something similar also happened before this, with a few extra debug logs...
And, because I know how frustrating bug reports are when no background info is provided 
Are you using Alexa app or echo skill?
Amazon echo skill native to hubitat.
Go into the Alexa app on your phone, disable the Hubitat skill, re-enable the skill, login, and select your hub. It will re-issue a valid token.
Done. Hopefully this takes care of it!
Unfortunately I'm still having a nightly issue with the hub. Every night, around 8-9pm, I get something like this in the log, and the hub becomes non-responsive. I have to go into the diagnostic portal and reboot. Then it works until the next night.
Here are the last 3 nights error log entries...
2020-08-24 09:04:41.396 pm
Error making Call to refresh Token: Remote host closed connection during handshake
2020-08-23 08:21:48.463 pm
Error making Call to refresh Token: Remote host closed connection during handshake
2020-08-22 07:37:46.131 pm
Error making Call to refresh Token: api.amazon.com:443 failed to respond
Since this thread was already marked solved, should I start a new one for this issue?
