When we try to create new nodes by scaling up a MachineSet using a 4.3 base image, the nodes cannot join the cluster and we can see that they report a failure like this one:

:/# systemctl --failed
  UNIT               LOAD   ACTIVE SUB    DESCRIPTION                          
● rhcos-fips.service loaded failed failed Check for FIPS mode                  
 
LOAD   = Reflects whether the unit definition was properly loaded.
ACTIVE = The high-level unit activation state, i.e. generalization of SUB.
SUB    = The low-level unit activation state, values depend on unit type.
 
1 loaded units listed. Pass --all to see loaded but inactive units, too.
To show all installed unit files use 'systemctl list-unit-files'.

# journalctl -u rhcos-fips.service 
-- Logs begin at Fri 2025-03-14 11:44:32 UTC, end at Fri 2025-03-14 11:46:53 UTC. --
Mar 14 11:44:37 ip-10-0-25-89.us-east-2.compute.internal systemd[1]: Starting Check for FIPS mode...
Mar 14 11:44:37 ip-10-0-25-89.us-east-2.compute.internal rhcos-fips[914]: Missing /run/ignition.json
Mar 14 11:44:37 ip-10-0-25-89.us-east-2.compute.internal systemd[1]: rhcos-fips.service: Main process exited, code=exited, status=1/FAILURE
Mar 14 11:44:37 ip-10-0-25-89.us-east-2.compute.internal systemd[1]: rhcos-fips.service: Failed with result 'exit-code'.
Mar 14 11:44:37 ip-10-0-25-89.us-east-2.compute.internal systemd[1]: Failed to start Check for FIPS mode.
Mar 14 11:44:37 ip-10-0-25-89.us-east-2.compute.internal systemd[1]: rhcos-fips.service: Triggering OnFailure= dependencies.
    

IPI on AWS. Version  

$ oc get clusterversion
NAME      VERSION                              AVAILABLE   PROGRESSING   SINCE   STATUS
version   4.19.0-0.nightly-2025-03-13-133005   True        False         7h26m   Cluster version is 4.19.0-0.nightly-2025-03-13-133005


    

Always
    

    1. Configure a machineset so that it uses a 4.3 base image.
    2. Scale up the machineset to create a new node
 
Detailed steps can be found here: https://polarion.engineering.redhat.com/polarion/redirect/project/OSE/workitem?id=OCP-77051


    

The nodes cannot join the cluster. When we access the console we find this error:

/# systemctl --failed
  UNIT               LOAD   ACTIVE SUB    DESCRIPTION                          
● rhcos-fips.service loaded failed failed Check for FIPS mode                  

    

The nodes should join without problems
    

We were not able to reproduce this issue in FIPS clusters. We were not able to reproduce it in 4.18 either.