Well I tell you- I don't know WHAT is going on. I am close to a year in with Hubitat and a super happy customer but over the last couple months SOMETHING has gone on... I have basically lost almost all reliability. I've tried to compensate in numerous ways but it is getting to be unmanageable.
Basically what it amounts to are rules improperly evaluating and things coming on and going off at the wrong time. It is a little like there are things either running int he background on a different clock or phantom devices (like motion sensors) activating on their own.
For example... it is currently 2:30am but the following condition says TRUE and all my "evening" lights are on:
Is there a way I can clear everything out and get a fresh code load or anything like that? Things are going bad and the WAF is deteriorating which does not bode well.
There is an issue for conditions/triggers that relate to sunset/sunrise.
From what I understand, your rule will kick off a job at 12:01am, in order for Hubitat to figure out when is the next sunset. (Subscription to the Hub sunset event).
Therefore, if you modify a rule, say 30 minutes before sunset, it won't run that day because it has to figure out when sunset is (I guess it doesn't store it).
Look at the next scheduled time for that rule to run... you may notice a time of 12:01am.
That's a sure sign of this issue.
Here is my complete rule @billmeek
Thanks for taking a look. It is supposed to be Sunset to 930 but I already have a double check at 10pm in case it missed the 930... in this case everything seemed to work fine and the lights were off as they were supposed to be after 930pm... but then around 2am they all came on... which is weird b/c there isnt even a trigger point there. sigh... I think the whole system is jacked up right now... for some reason
From what I can see at this point, there is not a job scheduled for 12:01am assuming I am looking at the right place. I also don't see a sunset event in the scheduled area. Do you think I should?
I don't think you will see scheduled event for sunset unless you use an offset. Then it will put the offset time in as a scheduled job.
As for your issue have you checked to make sure your location and time zone are correct?
You should see something in Event Subscriptions if you go into the settings cog and and scroll down.
Have you tried the built-in Simple Lighting app? This is a simple automation; on at a specific time and off at a specific time and is well suited for that app. Itās not a fix for a problem that shouldnāt exist, but it may get you on a faster path to stability.
There should not be a scheduled job at 12:01 AM. That happens only once for rules with before sunrise triggers. For your sunset trigger, there should be a Subscription to the sunset event. Like this:
Are you sure there's not another process changing the state of the lights?
Pick one of the items you're turning on that shouldn't have anything but this rule changing it's state. Maybe Christmas Tree? Look at that device in the Devices section and scroll to the bottom. You'll see a 'In Use By' section that shows everything that's using that device. I've had a few rules I deleted that weren't in the Rule Machine list anymore but the rules were listed under a device. I could click them in the 'In Use By' and then delete them again which did remove them completely.
I have found the same problems lately. Hub has become unreliable. I hadn't changed anything and yet sometimes the rules that were running smoothly don't run for some reason. Like last night my night routine to turn the heat down and turn off the lights didn't run. At 2:30 I got up because I was so hot and noticed all the lights still on and the heat still up. This rule usually runs but for some reason it didn't. And this is happening more and more. Hub freezes. Can't get into it using port 8081 so I have to pull the plug. Then I have to rebuild the database and then it runs ok for about a week and then things start happening again.
Hi @pcgirl, I didn't know about rebuilding the database... maybe I should try that. I will search for instructions. thanks - and I hope we can get back to good on these things....
This happens to me about once every week or two on three of my 5 hubs. There are two characteristics that distinguish two of those hubs; first they are the older model C-4 hubs with the external radio stick, and the second is that both hubs are more heavily loaded (> 200 devices each). The third hub is a āsticklessā C-4 hub which is my HubConnect server with >540 devices, mostly virtual.
The hubs are crashing. No blue light, only solid red. If I reboot them once a week the problem almost completely goes away. Iām convinced itās a resource leak somewhere.
Well I only have maybe 40-50 devices, if that and it is the C-4 hub with the stick on the side. Most times I have a solid blue light but I can't access the hub. Sometimes it is red. I have sent support a request to look at it.
every 3 weeks for the last 3 months my hub becomes unresponsive.
I am not getting red device lights, but externally and locally the hub is unresponsive so I have to do the :8081 and reboot it / soft reset.
Support have said its an issue with telnet and to slowly remove my wemo / tp link devices to see if that fixes it. It also affects my yeelights that were randomly staying on when triggered with a motion rule.
I constantly chase support / Bobby every 3rd day, all they come back with is that engineering is looking into it. No suggestions, no information, just, we're looking into it.
It is frustrating cause everything about HE I really like, and when it works for long periods and you completely forget about it its bliss. But when it goes bad and you cant isolate the problem - thats the worst lol
