When a  service is created  the ip of the service is overriden by adding a different ip to the /custom/hosts file and  when i try to resolve the dns name currently it is being resolved to the overriden ip, but i expect it to get resolved to the same service ip based on core dns file

4.21

Always

1. Install latest 4.21 microshift
2.  create pod and service using the yaml file below
apiVersion: v1
kind: Service
metadata:
  name: myservice
  namespace: test
spec:
  selector:
    app: myservice
  ports:
  - protocol: TCP
    port: 80
    targetPort: 80
  type: ClusterIP
---
apiVersion: v1
kind: Pod
metadata:
  name: myservice-pod
  namespace: test
  labels:
    app: myservice
spec:
  containers:
  - name: simple
    image: busybox:1.36
    command: ['sh', '-c', 'echo "Service is running" && sleep 3600']
    ports:
    - containerPort: 80

3. create custom host file using the content below
cat custom-hosts-conflict 
# Conflicting DNS entry - this should take precedence if configured correctly
# Actual service IP is: $SERVICE_IP
192.168.1.100  myservice.test.svc.cluster.local

# Additional test entries
192.168.1.101  test1.local
192.168.1.102  test2.local

4. Now add the host file to /etc/microshift/config.yaml
[redhat@el96-src-rpm-install-ostree-host1 ~]$ sudo cat /etc/microshift/config.yaml
dns:
  hosts:
    status: Enabled
    file: /home/redhat/custom-hosts-conflict

5. Restart microshift using the command `systemctl restart microshift`
6. check if entries have been propagated using command `oc get configmap hosts-file -n openshift-dns -o yaml`
7. create another pod using the definition below.
[redhat@el96-src-rpm-install-ostree-host1 ~]$ cat /tmp/pod1.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: dns-test-pod
  namespace: test
spec:
  containers:
  - name: test
    image: busybox:1.36
    command: ['sh', '-c', 'sleep 3600']
  restartPolicy: Never
8. Now run command `oc exec -n test myservice-pod -- nslookup myservice.test.svc.cluster.local`

[redhat@el96-src-rpm-install-ostree-host1 ~]$ oc exec -n test myservice-pod -- nslookup myservice.test.svc.cluster.local
Server:		10.43.0.10
Address:	10.43.0.10:53

Name:	myservice.test.svc.cluster.local
Address: 192.168.1.100


I see that it is being resolved to the ip that is added in the custom host file.

Should be resolved to the actual service ip based on the dns-default configmap here.
[redhat@el96-src-rpm-install-ostree-host1 ~]$ oc get configmap dns-default -n openshift-dns -o jsonpath='{.data.Corefile}'
.:5353 {
    bufsize 1232
    errors
    log . {
        class error
    }
    health {
        lameduck 20s
    }
    ready
    kubernetes cluster.local in-addr.arpa ip6.arpa {
        pods insecure
        fallthrough in-addr.arpa ip6.arpa
    }
    prometheus 127.0.0.1:9153
    forward . /etc/resolv.conf {
        policy sequential
    }
    cache 900 {
        denial 9984 30
    }
    hosts /tmp/hosts/hosts {
        fallthrough
    }        

    reload
}
hostname.bind:5353 {
    chaos
}