OCI: After OL8 upgrade instances becomes inaccessible in network and need a reboot

APPLIES TO:

Oracle Cloud Infrastructure

Linux x86-64

SYMPTOMS

After upgrading the instance from OL7 to OL8, the instance is not accessible via ssh frequently and it must be rebooted to restore the connection.

CAUSE

Instances IP address loss.

$ ip a 
2: ens3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc fq_codel state UP group default qlen 1000 
    link/ether 02:00:17:25:ef:2d brd ff:ff:ff:ff:ff:ff 
    altname enp0s3 
    inet6 fe80::17ff:fe25:ef2d/64 scope link  
       valid_lft forever preferred_lft forever 

$ systemctl status newtork 
network.service - LSB: Bring up/down networking 
   Loaded: loaded (/etc/rc.d/init.d/network; generated) 
   Active: failed (Result: exit-code) since Wed 2024-05-29 00:57:30 GMT; 5h 51min ago 
     Docs: man:systemd-sysv-generator(8) 
 
May 29 00:57:27 localhost dhclient[1371]: DHCPDISCOVER on ens3 to 255.255.255.255 port 67 interval 5 (xid=0xeffc2b3a) 
May 29 00:57:27 localhost dhclient[1371]: DHCPREQUEST on ens3 to 255.255.255.255 port 67 (xid=0xeffc2b3a) 
May 29 00:57:27 localhost dhclient[1371]: DHCPOFFER from 169.254.169.254 
May 29 00:57:27 localhost dhclient[1371]: DHCPACK from 169.254.169.254 (xid=0xeffc2b3a) 
May 29 00:57:30 localhost dhclient[1371]: DHCPDECLINE on ens3 to 255.255.255.255 port 67 (xid=0xeffc2b3a) 
May 29