This is not expected but ultimately what when wrong is related to the OS, not openHAB itself. When you say “complete image” do you mean the openHABian image?
Are you wired or wireless?
Many routers require you to apply the changes once you make a setting change like statically assigning an IP to a MAC address. Are you certain your change on the router actually applied?
I used the openHaBian image, connected by LAN
And yes - after reboot the system accepted the new “67” address as per the changed dhcp. Actually this is a managed cisco switch with a dhcp server - it is not the router.
In my experience, a restart of the machine, as OP said they did, will requery DHCP for an address. It almost has to because if it’s off it has no idea whether it’s been moved to a new network or not.
With dhcp the client checks for renewal at half the lease time period.
If you had a 2 hour lease time it would check at 1 hour and then renew for the 2 hours.
That’s how I understand it.
To get a different IP you would be best to flush the dhcp on the client or reboot it otherwise you get the old address again.
So, All of the debate of how DHCP leases work set aside.
If you have configured a reserved DHCP assigned IP to a MAC then as soon as the OS restarts the adaptor will pick up the newly assigned reserved IP address and apply that to the adaptor.
The OP clearly said that occurred
So the question would be more on why was the reserved IP assignment not fully propagating and this would be more in how his cisco device running the dhcp server is configured.
without seeing what was fully applied to his network adaptor it is a guessing game.
as to what all the DHCP server presented after configuring for a dhcp reserved IP address and the adaptor applied those values..
Things like DNS and mDNS are a few other items that would cache may come into play if OP is using a local DNS forwarder in his cisco device that may also be a factor.
Either way this appears to all be outside of the openhabian and OpenHab applications realm of control.
.More than likely he has something configured unique to his network to cache and it was still referring to old 114 ip and that had not fully updated after his changes. It is highly possible had he let the system sit and soak for a while or possible rebooted a second time after a hour or 2 this would have cleared up and worked fine.
EDIT:
However, one other thing that comes to mind that could be OpenHab related based on his described behaviors would be the various suggestion finder options repolling his network due to the IP change. He might try turning those off and see if it acts more normal after the reassignment of the new IP.
Hello to all
I found the reason - kind of..The Cisco had its own dhcp server running but for another VLAN (address range/pool). When I deactivated that dhcp server and now that the router was the (only) dhcp server, the setup actually worked.
I tried a homeassistant install and this one had the same problems - could not connect to the internet (IPv4 was up and running, internal ping was possible, but was stuck) and then I went through the network settings again.
Actually now I run the OH again with the cisco dhcp up and running - just that it is not enough to set the static route and keep uplink and DNS in “auto” in the cisco - you HAVE to provide these settings.
Still - what I do not like: If you install the OH first with dhcp (dynamic address) to learn the MAC address, you cannot then change to another static route via dhcp during a reboot. The log viewer seems to remember the first IP address even if the core system runs with the new static one and some “funny” thing in 3-rd party add-ons pop up. You better install a simple debian lite first on the raspbi to check the MAC, then set the static route as per need, THEN burn the OH image and install so that the static address is never changed again.
Thanks for all the help!
