C7 unresponsive (again)

It seems my problem with my C7 going paws up continues, and I'm stumped as to what could be causing it.

The thing just goes offline every now and again. It won't come up in the app on my phone, and I can't connect on my browser (not found), However, pinging the device does get a return. I have to go down to the office room where I have it, and power cycle it.

I've checked the logs, and the only entry other than regular status updates from my smart thermostat, is that a new driver has been detected for my Inovelli dimmer.

Any ideas? At this point, I'm thinking of trying to have it automatically reboot once a week or so, to see if that can clear out whatever gremlin seems to be getting into it on the regular. :slight_smile:

Thanks for any help you may be able to offer!

It is best to continue detailing your problems on the same thread as before, if you started a thread already, otherwise you'll end up getting the same questions every time. What platform version are you running, what shows in logs, what do you mean exactly by "going offline" (is the Diagnostic Tool reachable, are your automations working when the hub is offline, etc).

I don't mind answering questions again, if a new topic means the possibility of fresh eyes seeing this. No one who has previously responded to other topics has had a suggestion that worked. (I'm not faulting them at all. They were kind enough to offer ideas, and I am thankful for that.)

In order:

  1. I am running platform version

  2. What shows in logs was already stated: I get regular status updates (temps/on off state) from my smart thermostat, and I also had a notice of a new driver available for my Inovelli dimmer

  3. "going offline" means it is (not) responding, as if powered down. There is no connection. The app cannot connect to it, and neither can a web browser. There is no connecting to the diagnostic tool, as there is not connecting, period......except by pinging it from a command line on my computer. That gets a response. I have no automations that run automatically (time based, etc).

I could not find another topic you mentioned this, so I have no idea what was already asked so sorry if this was already discussed.

How many days between unresponsiveness? Depending on what and how many devices and apps you are running the hub could be running low on memory and when it reaches a low value it stops working.

I have a rule set up to monitor for low memory and reboot and notify me when it happens.

I am not sure the precise intervals, as I have not noted dates. I will have to do that. It has happened weeks apart, though- and generally happens in the overnight hours.

I don't think I'm taxing memory, as I have fewer than a dozen devices (only 7 physical), and eleven apps. But this begs the question- is there any low memory indicator that would show up in logs and such? It seem odd that the only indicator of low memory would be the unit becoming unresponsive.

I might have to set up a rule similar to yours, if for no other reason than to see if that's what's causing it.

Check this post: ‼ READ FIRST - Before Posting in Get Help

Do a local backup (download the backup) then restore right away (which automatically does a soft reset in the process). There is a chance you are encountering a rapid memory decline, which that will fix.

It is very odd that you can ping the hub but not connect to the diagnostic tool. I have never heard of that. Usually if the diag tool wont come up the hub is dead to the world and wont respond to anything. You are trying to access the diag tool on port 8081? Hubitat Diagnostic Tool | Hubitat Documentation

I've already done the local backup and soft reset one time, and it didn't seem to solve the problem. I guess I can do it again.

Next time it happens, I will try the diagnostic again. In my half awake stage when I discovered it this morning, I may have messed up that process.

Start here and get this working.

Then create a rule something like this.

of course change the IP address to your IP address. I am assuming you are familiar with RM, right? I also use the 15 minute average memory value in Hub Info to compare to. I let my hubs to get down to 90.0 but others use higher numbers.