Setting Up Kubernetes Networking with Weave Net

In order for Weave Net to work, you need to make sure IP forwarding is enabled on the worker nodes. Enable it by running the following on both workers:

sudo sysctl net.ipv4.conf.all.forwarding=1

echo "net.ipv4.conf.all.forwarding=1" | sudo tee -a /etc/sysctl.conf

Do the following on the controller server:

Log in to the controller server in a new terminal window, and then do the following:

1. Get the configuration from Weaveworks like this:

   wget https://github.com/weaveworks/weave/releases/download/v2.6.0/weave-daemonset-k8s-1.11.yaml
   

2. Edit the configuration file with vim.

   vim weave-daemonset-k8s-1.11.yaml
   

3. Add the following lines (press Escape and then enter i for insert mode):

   - name: IPALLOC_RANGE
     value: 10.200.0.0/16
   

4. The edited code snippet should then look like this:

       spec:
         containers:
           - name: weave
             command:
               - /home/weave/launch.sh
             env:
               - name: IPALLOC_RANGE
                 value: 10.200.0.0/16
               - name: HOSTNAME
   
   

5. Save the file by pressing Escape and then entering :wq! .

6. Apply the configuration with:

   kubectl apply -f ./weave-daemonset-k8s-1.11.yaml
   

7. Verify that everything is working:

kubectl get pods -n kube-system

This should return two weave-net pods and look something like this:

NAME              READY     STATUS    RESTARTS   AGE
weave-net-m69xq   2/2       Running   0          11s
weave-net-vmb2n   2/2       Running   0          11s

1. Spin up some Pods to test the networking functionality by first creating an Nginx deployment with two replicas:

cat << EOF | kubectl apply --kubeconfig admin.kubeconfig -f -
   apiVersion: apps/v1
   kind: Deployment
   metadata:
     name: nginx
   spec:
     selector:
       matchLabels:
         run: nginx
     replicas: 2
     template:
       metadata:
         labels:
           run: nginx
       spec:
         containers:
         - name: my-nginx
           image: nginx
           ports:
           - containerPort: 80
EOF

1. Next, create a service for that deployment so that we can test connectivity to services as well:

kubectl expose deployment/nginx

1. Start up another Pod. We will use this Pod to test our networking. We will test whether we can connect to the other Pods and services from this Pod.

kubectl run busybox --image=radial/busyboxplus:curl --command -- sleep 3600

POD_NAME=$(kubectl get pods -l run=busybox -o jsonpath="{.items[0].metadata.name}")

1. Get the IP addresses of our two nginx pods:

kubectl get ep nginx

There should be two IP addresses listed under `ENDPOINTS`. For example:

NAME      ENDPOINTS                       AGE
nginx     10.200.0.2:80,10.200.128.1:80   50m

1. Make sure the busybox Pod can connect to the nginx Pods on both of those IP addresses:

kubectl exec $POD_NAME -- curl <first nginx pod IP address>

kubectl exec $POD_NAME -- curl <second nginx pod IP address>

Both commands should return some HTML with the title "Welcome to Nginx!" This means that we can successfully connect to other pods.

5. Now let's verify that we can connect to services.

kubectl get svc

This should display the IP address for our Nginx service. For example, in this case, the IP is 10.32.0.54:

NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.32.0.1    <none>        443/TCP   1h
nginx        ClusterIP   10.32.0.54   <none>        80/TCP    53m

6. Check that we can access the service from the busybox Pod:

kubectl exec $POD_NAME -- curl <nginx service IP address>

This should also return HTML with the title "Welcome to nginx!"

Getting this response means that we have successfully reached the Nginx service from inside a Pod and that our networking configuration is working!