Oracle Linux: After Upgrading from OL7 to OL8, Instance Becomes Unresponsive After 24 Hours

APPLIES TO

Linux OS - Version Oracle Linux 7.0 and later
Linux x86-64

SYMPTOMS

Instance becomes unreachable about 24 hours after boot (aligns with DHCP lease time = 86400 seconds).
After upgrade from Oracle Linux 7 to 8 (via Leapp), networking is non-functional; ens3 shows as “unmanaged.”

dhclient process is not running when the issue occurs.

# nmcli  device status : 

  DEVICE ens3 ethernet unmanaged
  No active connection for ens3

OCI DHCP options visible when working (e.g., server <IP_ADDRESS>, MTU 9000), but absent when broken.
Missing iSCSI node config file at /var/lib/iscsi/nodes/iqn.2015-02.oracle.boot:uefi/<IP_ADDRESS>,<Port>,1/default before discovery.