A little over a week ago my hub started dying every day at midnight. I now have to remove and reinsert the power cord every morning. Looking at add a smart plug (with automation outside of HE) that will cycle power everyday for the short term until I figure out what I want to do. Here is some background on my setup and what I have tried.
Hardware version: Rev C-7
Platform version: 2.3.4.116
Device is setup with a static IP address. Address is also reserved in DHCP so there is no IP address conflict.
Have performed a backup, soft reset, and then a restore
DNS servers are set to 8.8.8.8, 8.8.4.4, 1.1.1.1, 1.0.0.1, 9.9.9.9
From the network Test page I am able to ping yahoo.com and espn.com. I am not able to ping google.com. I can ping goole.com from all other devices on my network. Not sure if this is by design or an HE bug.
There are no errors in the log when the device dies or in the hours leading up to the failure.
Hubitat Package Manager is checking for updates when the failure occurs.
Not sure what to do at this point. Without SSH or console access to troubleshoot, the only other option I can think of is to do a factory reset and set it back up from scratch. To go that route, I would do things one day at a time meaning on day 1 just get it setup on the network. Day two add the first app or device. Each day add an app or a device until I hopefully find the app or device that is casing the issue (if it is even an app or a deivce). This will take a couple of weeks to accomplish as I am sure the last one I add will be the one causing the issue.
Hoping someone else has an idea of something else I can try or some way to troubleshot this other than the method described above.
Please send me a private message and we could take a look at your hub's engineering logs to see if something jumps out to explain the problem. Generally, the symptoms you are experiecing are often related to a conflicting IP address within your network, but it can also be a hardware issue.
Also, we've had similar incidents in networks that have jumbo frames enabled.
I did enable jumbo frames on my network about the same time that this started happening. I moved the HE to an old 10/100 MB hub I had lying around and it did not die last night.
Sounds like jumbo frames may be your issue but FYI the above will likely cause other issues, like database corruption (which you would have corrected with the backup/restore).
Should you need to reboot regularly (hopefully not) you can safely reboot the hub using @thebearmay's hub info driver or the endpoint Send POST to: http://your-hubs-ip-here:8080/hub/reboot. There's also a rebooter available in HPM that allows you to reboot on a schedule.
When it dies, it is not on the network, so the only option is to do a hard power cycle unless there is some way to get console access I am unaware of to get to the WebUI to do a graceful reboot.
