Install Istio in Dual-Stack mode

3 minute read page test

Prerequisites

Istio 1.17 or later.
Kubernetes 1.23 or later configured for dual-stack operations.

Installation steps

If you want to use kind for your test, you can set up a dual stack cluster with the following command:

$ kind create cluster --name istio-ds --config - <<EOF
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
networking:
  ipFamily: dual
EOF

To enable dual-stack for Istio, you will need to modify your IstioOperator or Helm values with the following configuration.

apiVersion: install.istio.io/v1alpha1
kind: IstioOperator
spec:
  meshConfig:
    defaultConfig:
      proxyMetadata:
        ISTIO_DUAL_STACK: "true"
  values:
    pilot:
      env:
        ISTIO_DUAL_STACK: "true"
    # The below values are optional and can be used based on your requirements
    gateways:
      istio-ingressgateway:
        ipFamilyPolicy: RequireDualStack
      istio-egressgateway:
        ipFamilyPolicy: RequireDualStack

meshConfig:
  defaultConfig:
    proxyMetadata:
      ISTIO_DUAL_STACK: "true"
values:
  pilot:
    env:
      ISTIO_DUAL_STACK: "true"
  # The below values are optional and can be used based on your requirements
  gateways:
    istio-ingressgateway:
      ipFamilyPolicy: RequireDualStack
    istio-egressgateway:
      ipFamilyPolicy: RequireDualStack

Verification

Create three namespaces:
- dual-stack: tcp-echo will listen on both an IPv4 and IPv6 address.
- ipv4: tcp-echo will listen on only an IPv4 address.
- ipv6: tcp-echo will listen on only an IPv6 address.
```
$ kubectl create namespace dual-stack
$ kubectl create namespace ipv4
$ kubectl create namespace ipv6
```

Enable sidecar injection on all of those namespaces as well as the default namespace:

$ kubectl label --overwrite namespace default istio-injection=enabled
$ kubectl label --overwrite namespace dual-stack istio-injection=enabled
$ kubectl label --overwrite namespace ipv4 istio-injection=enabled
$ kubectl label --overwrite namespace ipv6 istio-injection=enabled

Create tcp-echo deployments in the namespaces:

$ kubectl apply --namespace dual-stack -f @samples/tcp-echo/tcp-echo-dual-stack.yaml@
$ kubectl apply --namespace ipv4 -f @samples/tcp-echo/tcp-echo-ipv4.yaml@
$ kubectl apply --namespace ipv6 -f @samples/tcp-echo/tcp-echo-ipv6.yaml@

Deploy the curl sample app to use as a test source for sending requests.
Zip
```
$ kubectl apply -f @samples/curl/curl.yaml@
```

Verify the traffic reaches the dual-stack pods:

$ kubectl exec "$(kubectl get pod -l app=curl -o jsonpath='{.items[0].metadata.name}')" -- sh -c "echo dualstack | nc tcp-echo.dual-stack 9000"
hello dualstack

Verify the traffic reaches the IPv4 pods:

$ kubectl exec "$(kubectl get pod -l app=curl -o jsonpath='{.items[0].metadata.name}')" -- sh -c "echo ipv4 | nc tcp-echo.ipv4 9000"
hello ipv4

Verify the traffic reaches the IPv6 pods:

$ kubectl exec "$(kubectl get pod -l app=curl -o jsonpath='{.items[0].metadata.name}')" -- sh -c "echo ipv6 | nc tcp-echo.ipv6 9000"
hello ipv6

Verify the envoy listeners:

$ istioctl proxy-config listeners "$(kubectl get pod -n dual-stack -l app=tcp-echo -o jsonpath='{.items[0].metadata.name}')" -n dual-stack --port 9000 -ojson | jq '.[] | {name: .name, address: .address, additionalAddresses: .additionalAddresses}'

You will see listeners are now bound to multiple addresses, but only for dual stack services. Other services will only be listening on a single IP address.

"name": "fd00:10:96::f9fc_9000",
"address": {
    "socketAddress": {
        "address": "fd00:10:96::f9fc",
        "portValue": 9000
    }
},
"additionalAddresses": [
    {
        "address": {
            "socketAddress": {
                "address": "10.96.106.11",
                "portValue": 9000
            }
        }
    }
],

Verify virtual inbound addresses are configured to listen on both 0.0.0.0 and [::].

$ istioctl proxy-config listeners "$(kubectl get pod -n dual-stack -l app=tcp-echo -o jsonpath='{.items[0].metadata.name}')" -n dual-stack -o json | jq '.[] | select(.name=="virtualInbound") | {name: .name, address: .address, additionalAddresses: .additionalAddresses}'

"name": "virtualInbound",
"address": {
    "socketAddress": {
        "address": "0.0.0.0",
        "portValue": 15006
    }
},
"additionalAddresses": [
    {
        "address": {
            "socketAddress": {
                "address": "::",
                "portValue": 15006
            }
        }
    }
],

Verify envoy endpoints are configured to route to both IPv4 and IPv6:

$ istioctl proxy-config endpoints "$(kubectl get pod -l app=curl -o jsonpath='{.items[0].metadata.name}')" --port 9000
ENDPOINT                 STATUS      OUTLIER CHECK     CLUSTER
10.244.0.19:9000         HEALTHY     OK                outbound|9000||tcp-echo.ipv4.svc.cluster.local
10.244.0.26:9000         HEALTHY     OK                outbound|9000||tcp-echo.dual-stack.svc.cluster.local
fd00:10:244::1a:9000     HEALTHY     OK                outbound|9000||tcp-echo.dual-stack.svc.cluster.local
fd00:10:244::18:9000     HEALTHY     OK                outbound|9000||tcp-echo.ipv6.svc.cluster.local

Now you can experiment with dual-stack services in your environment!

Cleanup

Cleanup application namespaces and deployments

$ kubectl delete -f @samples/curl/curl.yaml@
$ kubectl delete ns dual-stack ipv4 ipv6