Who faced such a problem (bug) of MySQL replication?

Question

MySQL

The replica works, everything is fine. The network is configured by default

network:
  version: 2
  renderer: networkd
  ethernets:
    enp1s0f0:
      dhcp4: yes

Further. The provider makes it possible to bind additional [Failover] IPs to the server and provides documentation. In accordance with it, I change on a replica

network:
  version: 2
  renderer: networkd
  ethernets:
    enp1s0:
      addresses:
       - адрес-сервера/24
       - дополнительный-ip1/32
       - дополнительный-ip2/32
      dhcp4: no
      gateway4: 88.999.99.1
      nameservers:
        addresses:
         - 88.999.99.2
         - 88.999.99.3

The server is up and running at these additional addresses. But, the replica stops working - and writes nonsense not related to reality

Fatal error: The slave I/O thread stops because master and slave have equal MySQL server ids;

Master_Server_Id: 5

The master ID - 1, the replica 2 - writes that they are the same. And I did not touch the configs at all. Even if I change it to 5 (and the master has 1), it still writes that the same and Master_Server_Id: 5 on the replica, although master 1
Restarting the slave does not help.
---
Next, I simply return the default network settings, and the replica works again.

Answer 1

So who has two addresses - the master or the replica?
If the master - replica uses one of them?