For the 3rd time in as many weeks, I got up this morning an my hub had lost connectivity, and nothing worked. I checked my router and the hubs static reservation was not listed in the active clients. I could not get the 8081 diagnostic tool to show up at all. I had to unplug the hub. The hub has since rebooted and is working. The last time it did this I was directed to do a soft reset and restore, which I did. I watched the freememory and yesterday it was at 381000 which I believe is OK. Below is the log. It looks like at 2:11am the unit just stopped, till I rebooted at 7:30am.
I don't know what is going on here but I am getting fed up with this failure. Can someone help.
The hub is directly plugged via ethernet to a Unifi UDM Pro on a static IP. It has been working on this system for over a year and just started losing connectivity within the last month.
Sounds like the hub is having issues communicating with the router, look in Logs | Location events and see if there are any messages that look like the hub trying to re-establish a DHCP connection. Under Networking there is also a DHCP auto reconnect time that may be useful.
I assumed when I saw this you meant a DHCP reservation rather than a static address assigned at the hub but just want to confirm.
FYI I have a similar setup except my HEs are plugged into downstream switches, not the UDM Pro itself. I'm on the latest software and have no issues but since this problem started fairly recently I wonder if it corresponds to a UDM software release? A long shot but had to ask. FYI here's what my port profile looks like, in case you want to compare.
I have screenshots of my UDMP and my US-16-150W switch and insights on the Hubitat C7 from last night when it stopped. You can see that the activity dropped after 2am.
If this was a network failure I'd expect to see log entries on the HE side after 2:11am. The hub would perhaps be inaccessible but all of the automations (less IP ones) would still run and log entries would still be generated. To me this looks more like a locked up hub than a network connectivity failure but I am by no means an expert. I'll be curious to see what @bobbyD can find in the engineering logs. Hopefully they'll give us something to go on.
Yes, I agree. The fact that the logs show nothing funky before 2:11am leads me to believe that the hub has indeed stopped functioning until the reboot. I have been checking the freememory often and it looks OK.
I don't think it has to do with the Database cleanup/backup time. It has stopped at 3 different times in the last month. BobbyD has looked the logs over and recommended setting the Ethernet speed to Auto Negotiated, which I have already done. I will be waiting to see what happens now.
Have you logged the temperatures reached by the hub? I don't know about where you are but here we have had some exceptionally hot weather and some of our machines are struggling. I've had a fan blowing on my hubs and router on the hottest days
My hubs do a backup at 02:15 nightly.
Could it be to do with this.
Maybe disable auto backup or move to a different time to see if the issue moves time wise or stops.
Just something to rule out maybe.
Nighly channel optimization maximizes wifi channel efficiency and minimize interference across multiple APs. Since OP's HE is using a hard-wired connection to the UDM's built-in switch, this should make no difference, unless he's trying to communicate with a secondary device that IS using wifi. But even then I would expect the impact to be limited to the remote device falling off the network. HE connectivity should be unaffected.
I have about 50 wifi devices and I use channel optimization... have had no issues, but of course everyone's experience is different.
I personally do not use channel optimization. I set my AP's using the signal mapper in the wifiman app. I have had no issues with wifi signal problems or interference.
