The assisted installer will always fail to create bond with active backup using nmstate yaml and the errors are : 

~~~ 
Jul 26 07:11:47 <hostname> bootkube.sh[8366]: Unable to reach API_URL's https endpoint at https://xx.xx.32.40:6443/version
Jul 26 07:11:47 <hostname> bootkube.sh[8366]: Checking validity of <hostname> of type API_INT_URL 
Jul 26 07:11:47 <hostname> bootkube.sh[8366]: Successfully resolved API_INT_URL <hostname> 
Jul 26 07:11:47 <hostname> bootkube.sh[8366]: Unable to reach API_INT_URL's https endpoint at https://xx.xx.32.40:6443/versionJul 26 07:12:23 <hostname> bootkube.sh[12960]: Still waiting for the Kubernetes API: 
Get "https://localhost:6443/readyz": dial tcp [::1]:6443: connect: connection refusedJul 26 07:15:15 <hostname> bootkube.sh[15706]: The connection to the server <hostname>:6443 was refused - did you specify the right host or port? 
Jul 26 07:15:15 <hostname> bootkube.sh[15706]: The connection to the server <hostname>:6443 was refused - did you specify the right host or port? 
 ~~~ 

Where, <hostname> is the actual hostname of the node. 

Adding sosreport and nmstate yaml file here : https://drive.google.com/drive/u/0/folders/19dNzKUPIMmnUls2pT_stuJxr2Dxdi5eb

4.12 
Dell 16g Poweredge R660

Always at customer side

1. Open Assisted installer UI (console.redhat.com -> assisted installer) 
2. Add the network configs as below for host1  

-----------
interfaces:
- name: bond99
  type: bond
  state: up
  ipv4:
    address:
    - ip: xx.xx.32.40
      prefix-length: 24
    enabled: true
  link-aggregation:
    mode: active-backup
    options:
      miimon: '140'
    port:
    - eno12399
    - eno12409
dns-resolver:
  config:
    search:
    - xxxx
    server:
    - xx.xx.xx.xx
routes:
  config:
    - destination: 0.0.0.0/0
      metric: 150
      next-hop-address: xx.xx.xx.xx
      next-hop-interface: bond99
      table-id: 254    
-----------

3. Enter the mac addresses of interfaces in the fields. 
4. Generate the iso and boot the node. The node will not be able to ping/ssh. This happen everytime and reproducible.
5. As there was no way to check (due to ssh not working) what is happening on the node, we reset root password and can see that ip address was present on bond, still ping/ssh does not work.
6. After multiple reboots, customer was able to ssh/ping and provided sosreport and we could see above mentioned error in the journal logs in sosreport.  
 

Fails to install. Seems there is some issue with networking.

Able to proceed with installation without above mentioned issues

- The installation works with round robbin bond mode in 4.12. 
- Also, the installation works with active-backup 4.10. 
- Active-backup bond with 4.12 is failing.