zshi-redhat

1. In above yaml files, apiVersion is set to app/v1 as below, I copied it from Dan's PR which I think is used in openshift. With kubernetes, apiVersion is set to extensions/v1beta1

apiVersion: apps/v1

2. the image names in above files will be replaced by env vars set in network operator manifests; example can be found here: https://github.com/openshift/cluster-network-operator/pull/54/commits/4211d3a41cf0b9cd526a1229c08453f18038a7d9#diff-1cf3d8447bdd10b4fd172cc55ca05700R37

{{.ContainerNetworkingCNIPluginsSupportedImage}} and {{.ContainerNetworkingCNIPluginsUnsupportedImage}}

zshi-redhat

[core@ip-10-0-15-231 cluster-network-operator]$ ./_output/linux/amd64/cluster-network-operator -v 6 --url-only-kubeconfig=/etc/kubernetes/kubeconfig
2019/01/11 07:17:45 Go Version: go1.10.3
2019/01/11 07:17:45 Go OS/Arch: linux/amd64
2019/01/11 07:17:45 operator-sdk Version: v0.1.0
2019/01/11 07:17:45 overriding kubernetes api to https://multus-test-cluster-api.devcluster.openshift.com:6443
INFO[0000] trying to become the leader                  
INFO[0000] Skipping leader election; not running in a cluster 
2019/01/11 07:17:46 Registering Components.
2019/01/11 07:17:46 Starting the Cmd.
2019/01/11 07:17:46 Reconciling NetworkConfig /default

zshi-redhat

apiVersion: v1
kind: ServiceAccount
metadata:
  namespace: sriov-device-plugin
  name: sriovdp-admission-controller-sa
---
apiVersion: v1
kind: Secret
metadata:
  name: sriovdp-admission-controller-sa-secret

zshi-redhat

apiVersion: "networkoperator.openshift.io/v1"
kind: "NetworkConfig"
metadata:
  name: "default"
spec:
  serviceNetwork: "172.30.0.0/16"
  clusterNetworks:
    - cidr: "10.128.0.0/14"
      hostSubnetLength: 9
  defaultNetwork:

zshi-redhat

apiVersion: "networkoperator.openshift.io/v1"
kind: "NetworkConfig"
metadata:
  name: "default"
spec:
  serviceNetwork: "172.30.0.0/16"
  clusterNetworks:
    - cidr: "10.128.0.0/14"
      hostSubnetLength: 9
  defaultNetwork:

zshi-redhat

apiVersion: v1
kind: Namespace
metadata:
  name: multus
  labels:
    name: multus
    openshift.io/run-level: "0"
  annotations:
    openshift.io/node-selector: "" #override default node selector
    openshift.io/description: "Multus network plugin components"

zshi-redhat

apiVersion: v1
kind: Pod
metadata:
  name: test1
spec:
  containers:
  - name: app1 
    image: centos/tools 
    imagePullPolicy: IfNotPresent
    command: [ "/bin/bash", "-c", "--" ]

zshi-redhat

#!/bin/bash
# scale up and down pod replicas
# sleep time is to ensure pods are created or deleted successfully

while true
do
	oc scale --replicas 1 deployment/multus-test-deployment >/dev/null
	sleep 20
	oc scale --replicas 6 deployment/multus-test-deployment >/dev/null
	sleep 20

zshi-redhat

[zshi@192 templates]$ oc describe pod multus-test-deployment-694d7d698c-bf52x
Name:               multus-test-deployment-694d7d698c-bf52x
Namespace:          default
Priority:           0
PriorityClassName:  <none>
Node:               ip-10-0-128-153.ec2.internal/10.0.128.153
Start Time:         Wed, 23 Jan 2019 12:23:41 +0800
Labels:             app=multus-test-app
                    pod-template-hash=2508382547
Annotations:        k8s.v1.cni.cncf.io/networks-status=[{

zshi-redhat

#!/bin/bash
set -x

## On the undercloud, don't forget to configure the DNS

yum -y install docker
systemctl enable docker
systemctl start docker

cat <<EOF > /etc/yum.repos.d/kubernetes.repo