Unwanted DHCP setting on server

I'm configuring a new Debian 11 server running on a private network. It is determinedly taking a DHCP address in addition to the static IP I am manually configuring. I don't understand where it gets the DHCP address and how I can turn that off.

Out of the box it grabs a DHCP address (I guess it has an auto-configure for the network?); since I want it running on a manually assigned static IP, I modified /etc/network/interfaces to configure the static IP (in addition to the lo lines):

auto enp0s25
allow-hotplug enp0s25
iface enp0s25 inet static
    address 10.0.0.16/24
    gateway 10.0.0.1
    dns-nameservers 10.0.0.1

and I am rewarded for my effort by having two IPs configured for this device

user@server:~$ ip a
1: lo: ...
2: enp0s25: <BROADCAST,MULTICAST,DYNAMIC,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether d0:50:99:5a:08:98 brd ff:ff:ff:ff:ff:ff
    inet 10.0.0.16/24 brd 10.0.0.255 scope global enp0s25
       valid_lft forever preferred_lft forever
    inet 10.0.0.190/24 brd 10.0.0.255 scope global secondary enp0s25
       valid_lft forever preferred_lft forever
    inet6 fe80::d250:99ff:fe5a:898/64 scope link 
       valid_lft forever preferred_lft forever

(note this configuration is stably reproducible on reboot)

As I understand it there should be three ways of configuring the network:

• networking - I'm using that, because that's what I'm most used to
• NetworkManager - it's not installed, and the service isn't running
• systemd - systemd-networkd claims to be disabled

So what other config(s) am I not finding? And how do I make sense of this mess?

There are other screwy things about the networking (ifdown/ifup not working properly, restarting networking throwing errors), but I'm guessing/hoping its related to unclear control of the interface.

EDIT: This is the syslog output relevant to enp0s25:

Dec 16 19:37:57 angelo systemd[1]: Started ifup for enp0s25.
Dec 16 19:37:57 angelo kernel: [    1.222564] e1000e 0000:00:19.0 enp0s25: renamed from eth0
Dec 16 19:37:57 angelo kernel: [    2.613911] e1000e 0000:00:19.0 enp0s25: NIC Link is Down
Dec 16 19:37:57 angelo sh[413]: ifup: failed to bring up enp0s25
Dec 16 19:37:57 angelo systemd[1]: [email protected]: Main process exited, code=exited, status=1/FAILURE
Dec 16 19:37:57 angelo systemd[1]: [email protected]: Failed with result 'exit-code'.
Dec 16 19:37:57 angelo connmand[467]: enp0s25 {create} index 2 type 1 <ETHER>
Dec 16 19:37:57 angelo connmand[467]: enp0s25 {update} flags 36866 <DOWN>
Dec 16 19:37:57 angelo connmand[467]: enp0s25 {newlink} index 2 address D0:50:99:5A:08:98 mtu 1500
Dec 16 19:37:57 angelo connmand[467]: enp0s25 {newlink} index 2 operstate 2 <DOWN>
Dec 16 19:37:57 angelo connmand[467]: Adding interface enp0s25 [ ethernet ]
Dec 16 19:37:57 angelo connmand[467]: enp0s25 {update} flags 36867 <UP>
Dec 16 19:37:57 angelo connmand[467]: enp0s25 {newlink} index 2 address D0:50:99:5A:08:98 mtu 1500
Dec 16 19:37:57 angelo connmand[467]: enp0s25 {newlink} index 2 operstate 2 <DOWN>
Dec 16 19:37:58 angelo avahi-daemon[464]: Joining mDNS multicast group on interface enp0s25.IPv4 with address 10.0.0.16.
Dec 16 19:37:58 angelo connmand[467]: enp0s25 {add} address 10.0.0.16/24 label enp0s25 family 2
Dec 16 19:37:58 angelo avahi-daemon[464]: New relevant interface enp0s25.IPv4 for mDNS.
Dec 16 19:37:58 angelo connmand[467]: enp0s25 {add} route 10.0.0.0 gw 0.0.0.0 scope 253 <LINK>
Dec 16 19:37:58 angelo avahi-daemon[464]: Registering new address record for 10.0.0.16 on enp0s25.IPv4.
Dec 16 19:37:58 angelo connmand[467]: enp0s25 {add} route 0.0.0.0 gw 10.0.0.1 scope 0 <UNIVERSE>
Dec 16 19:38:00 angelo connmand[467]: enp0s25 {add} route fe80:: gw :: scope 0 <UNIVERSE>
Dec 16 19:38:00 angelo connmand[467]: enp0s25 {update} flags 102467 <UP,RUNNING,LOWER_UP>
Dec 16 19:38:00 angelo connmand[467]: enp0s25 {newlink} index 2 address D0:50:99:5A:08:98 mtu 1500
Dec 16 19:38:00 angelo connmand[467]: enp0s25 {newlink} index 2 operstate 6 <UP>
Dec 16 19:38:00 angelo kernel: [    5.902126] e1000e 0000:00:19.0 enp0s25: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx
Dec 16 19:38:00 angelo kernel: [    5.902187] IPv6: ADDRCONF(NETDEV_CHANGE): enp0s25: link becomes ready
Dec 16 19:38:01 angelo avahi-daemon[464]: Joining mDNS multicast group on interface enp0s25.IPv6 with address fe80::d250:99ff:fe5a:898.
Dec 16 19:38:01 angelo avahi-daemon[464]: New relevant interface enp0s25.IPv6 for mDNS.
Dec 16 19:38:01 angelo avahi-daemon[464]: Registering new address record for fe80::d250:99ff:fe5a:898 on enp0s25.*.
Dec 16 19:38:05 angelo avahi-daemon[464]: Registering new address record for 10.0.0.190 on enp0s25.IPv4.
Dec 16 19:38:05 angelo connmand[467]: enp0s25 {add} address 10.0.0.190/24 label enp0s25 family 2
Dec 16 19:38:05 angelo connmand[467]: enp0s25 {add} route 10.0.0.1 gw 0.0.0.0 scope 253 <LINK>
Dec 16 19:38:05 angelo connmand[467]: enp0s25 {add} route 82.165.8.211 gw 10.0.0.1 scope 0 <UNIVERSE>
Dec 16 19:38:09 angelo connmand[467]: enp0s25 {del} route 82.165.8.211 gw 10.0.0.1 scope 0 <UNIVERSE>
Dec 16 19:38:42 angelo connmand[467]: enp0s25 {del} route 0.0.0.0 gw 10.0.0.1 scope 0 <UNIVERSE>

You said it was a private network so I'm going to assume you have access to your router. Changing the DHCP settings on your router to give it a static IP address will not result in needless IP associations.

1. Type ifconfig and look for the gateway ip address
2. Login to your router in a browser using the gateway address
3. In DHCP settings on your router, you can assign a static IP association there. It should not give you an additonal IP address.

I uninstalled connman to resolve this. Not my preferred choice, but I couldn't find a better way by modifying config files to make it stop doing its auto-DHCP nonsense. Note: uninstalling connman does take down the network connections of the server, so reboot required

Some links that convinced me this was the best shot:

• https://www.linuxquestions.org/questions/linux-networking-3/ip-configured-as-static-but-connmand-set-it-to-dhcp-4175602199/
• https://www.reddit.com/r/pihole/comments/por9uo/network_after_upgrade_to_bullseye/

Would still welcome a less nuclear, more directed solution.

I assume you correctly checked anything else such as network-manager not running as you mentioned. But even then, if you configure the /etc/network/interfaces file as I show below the system should automatically use that instead of things like network-manager.

I have always found that the following consistently provides me with a static ip address even on networks with a dhcp server running. Of course you want to make sure the ip address is outside of the block the dhcp server hands out, otherwise you get conflicts. I used a https://www.rfc-editor.org/rfc/rfc5737.html network block.

auto enp0s25
iface enp0s25 inet static
        address 192.0.2.10
        netmask 255.255.255.0
        broadcast 192.0.2.255
        network 192.0.2.0
        gateway 192.0.2.1

I prefer to not use allow-hotplug because (man interfaces):

Lines beginning with "allow-" are used to identify interfaces that should 
be brought up automatically by various subsystems. This may be done using    
a command  such as  "ifup  --allow=hotplug eth0 eth1", which will only 
bring up eth0 or eth1 if it is listed in an "allow-hotplug" line.

It is possible the interface had already a dhcp assigned ip address before you changed it to be static. In that case it may have 2 ip address until you reboot.

As an aside, this probably does not apply to debian systems, but just in case, on ubuntu 20.04 systems I find this when wanting to configure the interfaces file:

# ifupdown has been replaced by netplan(5) on this system.  See
# /etc/netplan for current configuration.
# To re-enable ifupdown on this system, you can run:
#    sudo apt install ifupdown