Configuring Passthrough Networking for a LoadBalancer Service

Background

A Kubernetes cluster can publish applications running on a group of pods as Services, which provide unified layer-4 access entries. For a Loadbalancer Service, kube-proxy configures the LoadbalanceIP in status of the Service to the local forwarding rule of the node by default. When a pod accesses the load balancer from within the cluster, the traffic is forwarded within the cluster instead of being forwarded by the load balancer.

kube-proxy is responsible for intra-cluster forwarding. kube-proxy has two forwarding modes: iptables and IPVS. iptables is a simple polling forwarding mode. IPVS has multiple forwarding modes but it requires modifying the startup parameters of kube-proxy. Compared with iptables and IPVS, load balancers provide more flexible forwarding policies as well as health check capabilities.

Solution

CCE supports passthrough networking. You can configure the annotation of kubernetes.io/elb.pass-through for the Loadbalancer Service. Intra-cluster access to the Service load balancer address is then forwarded to backend pods by the load balancer.

Figure 1 Passthrough networking illustration

• CCE clusters

  When a LoadBalancer Service is accessed within the cluster, the access is forwarded to the backend pods using iptables/IPVS by default.

  When a LoadBalancer Service (configured with elb.pass-through) is accessed within the cluster, the access is first forwarded to the load balancer, then the nodes, and finally to the backend pods using iptables/IPVS.

• CCE Turbo clusters

  When a client accesses a LoadBalancer Service from within the cluster, passthrough is used by default. In this case, the client directly accesses the load balancer private network IP address and then access a container through the load balancer.

Constraints

• In a CCE standard cluster, after passthrough networking is configured for a dedicated load balancer, the private IP address of the load balancer cannot be accessed from the node where the workload pod resides or other containers on the same node as the workload.
• Passthrough networking is not supported for clusters of v1.15 or earlier.
• In IPVS network mode, the passthrough settings of Services connected to the same load balancer must be the same.
• If node-level (local) service affinity is used, kubernetes.io/elb.pass-through is automatically set to onlyLocal to enable pass-through.

Procedure

This section describes how to create a Deployment using an Nginx image and create a Service with passthrough networking enabled.

1. Use the Nginx image to create a Deployment.

   apiVersion: apps/v1     
   kind: Deployment         
   metadata:
     name: nginx            
   spec:
     replicas: 2                     
     selector:              
       matchLabels:
         app: nginx
     template:              
       metadata:
         labels:
           app: nginx
       spec:
         containers:
         - image: nginx:latest
           name: container-0
           resources:
             limits:
               cpu: 100m
               memory: 200Mi
             requests:
               cpu: 100m
               memory: 200Mi
         imagePullSecrets:
         - name: default-secret

2. For a LoadBalance Service, set kubernetes.io/elb.pass-through to true. In this example, a shared load balancer named james is automatically created.
   For details about how to create a LoadBalancer Service, see LoadBalancer.

   apiVersion: v1 
   kind: Service 
   metadata: 
     annotations:   
       kubernetes.io/elb.pass-through: "true"
       kubernetes.io/elb.class: union
       kubernetes.io/elb.autocreate: '{"type":"public","bandwidth_name":"cce-bandwidth","bandwidth_chargemode":"bandwidth","bandwidth_size":5,"bandwidth_sharetype":"PER","eip_type":"5_bgp","name":"james"}'
     labels: 
       app: nginx 
     name: nginx 
   spec: 
     externalTrafficPolicy: Local
     ports: 
     - name: service0 
       port: 80
       protocol: TCP 
       targetPort: 80
     selector: 
       app: nginx 
     type: LoadBalancer

Verification

Check the ELB load balancer corresponding to the created Service. The load balancer name is james. The number of ELB connections is 0, as shown in the following figure.

Use kubectl to connect to the cluster, go to an Nginx container, and access the ELB address. The access is successful.

# kubectl get pod
NAME                     READY   STATUS    RESTARTS   AGE
nginx-7c4c5cc6b5-vpncx   1/1     Running   0          9m47s
nginx-7c4c5cc6b5-xj5wl   1/1     Running   0          9m47s
# kubectl exec -it nginx-7c4c5cc6b5-vpncx -- /bin/sh
# curl 120.46.141.192
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
<style>
    body {
        width: 35em;
        margin: 0 auto;
        font-family: Tahoma, Verdana, Arial, sans-serif;
    }
</style>
</head>
<body>
<h1>Welcome to nginx!</h1>
<p>If you see this page, the nginx web server is successfully installed and
working. Further configuration is required.</p>

<p>For online documentation and support please refer to
<a href="http://nginx.org/">nginx.org</a>.<br/>
Commercial support is available at
<a href="http://nginx.com/">nginx.com</a>.</p>

<p><em>Thank you for using nginx.</em></p>
</body>
</html>

Wait for a period of time and view the ELB monitoring data. A new access connection is created for the ELB, indicating that the access passes through the ELB load balancer as expected.