Iād like to take noneāretries and resiliency should be sufficient... thatās the basis of my discontent. However, Iām running the C8 100ma short of its recommended power (thanks to @aaiyar for pointing that out to me earlier in this thread) so who knows if thatās a contributor. Clearly, the do nothing approach isnāt working, though, so at this point and once Iām back home, Iāll correct the power situation and kick off a clean restart via the Diagnostic tool (port 8081) and take it from there.
If directed at me: as I mentioned, conditioned power backup is already in place albeit requiring a bit of re-wiring now Iām aware of my power miscalculation. Easy fix once Iām home.
In addition, any DHCP client worth its weight will assign an IPv4 APIPA address after an initial retry/timeout period (169.254.0.0/16) but continue to endlessly attempt re-discovery. My suspicion is that itās not doing that right now but Iām inferring thatāIāll know more once Iām home.
To kinda close this out because it's working: got back home; C8 responding to pings; tried admin interface from Edge browser and worked fine; logs indicate no comms with Amazon Echo skill; dashboard unavailable; remote admin unavailable; iPhone app unavailable even locally connected??; performed a graceful shutdown; powered back on and all working as normal. I still don't understand.
PS: it turns out my recollection of its power supply was wrong, my Flic hub is indeed powered by the laptop's USB 3.1 port whereas the Hubitat is powered by a USB hub providing 5V at 2.1A. Nonetheless, that was luck on my part so its power requirements remain a useful lesson but didn't contribute to this still-not-understood issue.
This is usually an indicator of DNS failure, so the hub cannot connect to Hubitat's cloud server(s). That latter connection is needed for the Echo skill to work, as well as remote admin access.
Both my hubs are configured with static IPs that are outside the range assigned by the DHCPd on my LAN. This is the DNS configuration that I use "192.168.7.1, 8.8.8.8, 1.1.1.1, 9.9.9.9". The private IP address points to an instance of DNSMasq exposed to VLAN the hubs are on.
The Windows DNS service is working perfectly for my client laptop and every other device in the house (80+); it may have been down at the point the Hubitat was first ungracefully rebooted after power was restored. I initially suspected an aggressive DNS negative cache but I (hard/cold) rebooted it two more times afterwards and it still didn't resolve the issue. The moment I gracefully shut it down and subsequently cold boot it, everything works again. Any idea why?
I have the C8's Ethernet interface configured with a static address of 172.31.0.201/16 and its Wi-Fi interface (which doesn't seem to support static addresses) configured with the same address through a DHCP reservation. The Ethernet interface is the active one so there are no ARP issues or the like. I did it this way so the address remains consistent for those few occasions where I've needed to move the hub away from an available Ethernet port.
@All those that have helped thus far: thanks! I fear we may have got off on the wrong foot-- apologies if any/all of that is on me.
I know you folks here probably hear this a lot but the infrastructure stuff that underlies what we're discussing here is what I do for a living--if you've heard of Active Directory or Windows Server (specifically, its security aspects) or Hyper-V Shielded VMs or Azure Automanage, I designed tons/much/pieces/chunks/components of these products and/or services as their owner in Microsoft's AD/Windows Server/Hyper-V/Azure product groups for almost 15 years (limeypride==Dean Wells). Anyhoo, hopeful that that might lend a bit of credibility to my explanation of the symptoms I'm experiencing (and can repro).
Agreed, While power outages are infrequent where I live due to buried power cables, there can be power voltage fluctuations. I have several Uninterruptible Power Supplies around my home. Most of these devices cost as much or more than my Hubitat, but I want protection for my cable modem, my router, my Hubitat, my Hue and Lutron bridges, my Internet switches, my network attached storage and my computers.
The nice thing about a UPS is that it provides high level surge protection from power spikes, protection from voltage variation high and low, as well as protection from power outages for a long enough period to allow a graceful device shutdown.
This winter, we had a power outage due to an ice storm that affected large portions of my community. There were several power surges before the power went out. I lost several of my Zigbee smart plugs that were plugged directly into the wall outlets. They were inexpensive and easily replaced. Fortunately, none of the more expensive equipment plugged into UPS units were affected. A UPS is one of the best insurance policies you can purchase for your electronics.
