Bug 1822945

Description of problem:
Follow https://docs.openshift.com/container-platform/3.11/admin_guide/managing_networking.html#admin-guide-deploying-an-egress-router-pod to deploy egress router pod in v4.5, but egress router pod is stuck in Init:CrashLoopBackOff state

Version-Release number of selected component (if applicable):
4.5.0-0.nightly-2020-04-08-134816

How reproducible:
Always

Steps to Reproduce:
[weliang@weliang FILE]$ cat egressrouterpod.yaml
apiVersion: v1
kind: Pod
metadata:
  name: egress-redirect-pod
  labels:
    name: egress-redirect-pod
  annotations:
    pod.network.openshift.io/assign-macvlan: "true"
spec:
  initContainers:
  - name: egress-router
    image: registry.redhat.io/openshift4/ose-egress-router 
    imagePullPolicy:  IfNotPresent
    securityContext:
      privileged: true
    env:
    - name: EGRESS_SOURCE
      value: 139.178.76.12
    - name: EGRESS_GATEWAY
      value: 139.178.76.1
    - name: EGRESS_DESTINATION 
      value:  172.217.7.206
    - name: EGRESS_ROUTER_MODE
      value: init
  containers:
  - name: egressrouter-redirect
    image: registry.redhat.io/openshift4/ose-egress-router
    imagePullPolicy:  IfNotPresent
[weliang@weliang FILE]$  oc create -f egressrouterpod.yaml
[weliang@weliang FILE]$ oc get pod
NAME                  READY   STATUS                  RESTARTS   AGE
egress-redirect-pod   0/1     Init:CrashLoopBackOff   6          7m41s
test-pod-1            1/1     Running                 0          18m
test-pod-2            1/1     Running                 0          18m
test-pod-3            1/1     Running                 0          18m

[weliang@weliang FILE]$ oc logs egress-redirect-pod
Error from server (BadRequest): container "egressrouter-redirect" in pod "egress-redirect-pod" is waiting to start: PodInitializing
[weliang@weliang FILE]$ oc describe pods egress-redirect-pod
Name:         egress-redirect-pod
Namespace:    test
Priority:     0
Node:         compute-0/139.178.76.11
Start Time:   Fri, 10 Apr 2020 10:52:19 -0400
Labels:       name=egress-redirect-pod
Annotations:  k8s.v1.cni.cncf.io/network-status:
                [{
                    "name": "openshift-sdn",
                    "interface": "eth0",
                    "ips": [
                        "10.131.0.41"
                    ],
                    "default": true,
                    "dns": {}
                }]
              k8s.v1.cni.cncf.io/networks-status:
                [{
                    "name": "openshift-sdn",
                    "interface": "eth0",
                    "ips": [
                        "10.131.0.41"
                    ],
                    "default": true,
                    "dns": {}
                }]
              openshift.io/scc: node-exporter
              pod.network.openshift.io/assign-macvlan: true
Status:       Pending
IP:           10.131.0.41
IPs:
  IP:  10.131.0.41
Init Containers:
  egress-router:
    Container ID:   cri-o://3934dcea39c81f2e0b70d7aca0909bbdbb911ee8a172a86e8fb2283bbff2c728
    Image:          registry.redhat.io/openshift4/ose-egress-router
    Image ID:       registry.redhat.io/openshift4/ose-egress-router@sha256:1e7abd047edcd20034f1abd4526240956564cb252f941de8806e40c59a953eb6
    Port:           <none>
    Host Port:      <none>
    State:          Waiting
      Reason:       CrashLoopBackOff
    Last State:     Terminated
      Reason:       Error
      Exit Code:    3
      Started:      Fri, 10 Apr 2020 10:53:56 -0400
      Finished:     Fri, 10 Apr 2020 10:53:56 -0400
    Ready:          False
    Restart Count:  4
    Environment:
      EGRESS_SOURCE:       139.178.76.12
      EGRESS_GATEWAY:      139.178.76.1
      EGRESS_DESTINATION:  172.217.7.206
      EGRESS_ROUTER_MODE:  init
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-c9hpg (ro)
Containers:
  egressrouter-redirect:
    Container ID:   
    Image:          registry.redhat.io/openshift4/ose-egress-router
    Image ID:       
    Port:           <none>
    Host Port:      <none>
    State:          Waiting
      Reason:       PodInitializing
    Ready:          False
    Restart Count:  0
    Environment:    <none>
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-c9hpg (ro)
Conditions:
  Type              Status
  Initialized       False 
  Ready             False 
  ContainersReady   False 
  PodScheduled      True 
Volumes:
  default-token-c9hpg:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-c9hpg
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type     Reason          Age                 From                Message
  ----     ------          ----                ----                -------
  Normal   Scheduled       <unknown>           default-scheduler   Successfully assigned test/egress-redirect-pod to compute-0
  Normal   AddedInterface  113s                multus              Add eth0 [10.131.0.41/23]
  Normal   Pulling         113s                kubelet, compute-0  Pulling image "registry.redhat.io/openshift4/ose-egress-router"
  Normal   Pulled          108s                kubelet, compute-0  Successfully pulled image "registry.redhat.io/openshift4/ose-egress-router"
  Normal   Created         18s (x5 over 108s)  kubelet, compute-0  Created container egress-router
  Normal   Started         18s (x5 over 108s)  kubelet, compute-0  Started container egress-router
  Normal   Pulled          18s (x4 over 107s)  kubelet, compute-0  Container image "registry.redhat.io/openshift4/ose-egress-router" already present on machine
  Warning  BackOff         18s (x9 over 106s)  kubelet, compute-0  Back-off restarting failed container
[weliang@weliang FILE]$ 

Actual results:
egress-redirect-pod   0/1     Init:CrashLoopBackOff

Expected results:
egress-redirect-pod   1/1     Running

Additional info: