Unable to add a new worker node to a cluster installed using Assisted Installer. The node being added to the cluster is an BareMetal Node(Dell PowerEdge R730) with bonded interfaces. The discovery iso gets created with the supplied network config yaml. Booting the node with this iso does allow ssh and we can see the ignition is writing to disk. Once the node boots from the disk, it constantly tries to resolve the API over IPv6 and sits there forever. 

    4.12.46 / 4.14.13

 Install a 3 master 3 worker cluster on Baremetal Platform using AI and try to add a new BareMetal node to the cluster. 

1. Generate the ISO with bond+vlan configurations with 2 interfaces. 
2. Boot the node with generated discovery ISO. Networks starts as configured.
3. Once the node boots from the disk, node constantly tries to resolve the API using IPv6. There are no IPv4 interfaces.

    The node constantly tries to resolve the API and does not fetch the worker ignition data and hence does not get added to the cluster. 

    The Node to be added to the cluster.

- Tried to boot a node with Live ISO and apply the same networking configurations via nmtui and the node is able to reach the API and curl as well. 
- Tried to burn the RHCOS iso with ignition and networking data, but the same issue persists.  
- As per our findings, the the IP is lost after disk write. We can see the first boot does copy the networking details to `/run/NetworkManager/system-connections/`.
- Slack thread raised for the issue - 

Thread - https://redhat-internal.slack.com/archives/CUPJTHQ5P/p1707138547559769

- Drive link with must gather, console logs and error details - 

Link - https://drive.google.com/drive/folders/1l_0bxFCsIdyADYiCxuHXvAXjL-QZ8uU7?usp=sharing