When a config drift happens in a system unit or in a dropin file, and we create the /run/machine-config-daemon-force to re-apply the configuration, the config drift is not fixed anymore.


    

4.22
    

Always
    

    1. Create a MC to deploy a dropin file

kind: List
apiVersion: v1
metadata: {}
items:
  - apiVersion: machineconfiguration.openshift.io/v1
    kind: MachineConfig
    metadata:
      labels:
        machineconfiguration.openshift.io/role: worker
      name: drifted-dropins-test
    spec:
      config:
        ignition:
          version: 3.5.0
        passwd:
          users: []
        storage:
          files: []
        systemd:
          units:
            - dropins:
                - contents: '[Service]

                    Environment="FAKE_OPTS=fake-value"'
                  name: 10-chrony-drop-test.conf
              enabled: true
              name: chronyd.service
      extensions: []
      kernelArguments: []
      osImageURL: ''

    2. "oc debug" a node and manually modify the 10-chrony-drop-test.conf file

    3. The MCP will become degraded reporting the config drif

    4. "oc debug" into the node again and create the /run/machine-config-daemon-force file to re-apply the config



    

The file causing the config drift degradation is not re-applied, and the degradation is not fixed.

    

When the configuration is re-applied once we create the /run/machine-config-daemon-force file, the config drfit degradation should be fixed.

    

It impacts these test cases:

- "[sig-mco] MCO config drift Author:sregidor-Longduration-NonPreRelease-High-47008-Config Drift. Dropin file. [Serial]"
- "[sig-mco] MCO config drift Author:sregidor-Longduration-NonPreRelease-High-47009-Config Drift. New Service Unit. [Serial]"

This issue can be verified by running them and checking that they pass.


Likely related to: https://github.com/openshift/machine-config-operator/pull/5582