Guides Overview

There are several guides available. To follow the guides included in this section, you will need at least two Kubernetes clusters with Service Mesh Hub installed on one. You can easily set up two such clusters using Kind as detailed below. Some of the guides also make use of the Bookinfo sample application. You can install the application by following the steps in the Bookinfo deployment section.

To become familiar with Service Mesh Hub, we recommend following the guides in the order that they appear here, starting from the top and going down.


There are three pre-requisites to following these guides:

  1. Install kubectl
  2. Install meshctl
  3. Have multiple Kubernetes clusters ready to use, accessible in different kubeconfig contexts. If you don't have access to multiple Kubernetes clusters, see the section below, Using Kind to use Kubernetes in Docker (Kind) to spin up two clusters in containers.

Assumptions made in the guides

We will assume in this guide that you have access to two clusters and the following two contexts are available in your kubeconfig file.

Your actual context names will likely be different.

We assume you've installed Service Mesh Hub into the cluster represented by the context management-cluster-context.

Two registered clusters

We also assume you've registered both of those clusters with Service Mesh Hub:

meshctl cluster register \
  --remote-cluster-name management-cluster \
  --remote-context management-cluster-context
meshctl cluster register \
  --remote-cluster-name remote-cluster \
  --remote-context remote-cluster-context

At this point we have two clusters, management-cluster-context and remote-cluster-context both registered with Service Mesh Hub which happens to be installed on the management-cluster-context cluster.

Using Kind

Kubernetes in Docker makes it easy to stand up Kubernetes clusters on your local desktop for testing and development. To assist with learning about Service Mesh Hub, you can use Kind to deploy two Kubernetes clusters. You can install Kind by following the installation steps in their Quick Start guide.

Once you have Kind installed, you can create the two clusters in question by running the following:

meshctl demo init

The command will do the following:

You should now be ready to start going through the guides.

Bookinfo deployment

For some parts of the guide, you'll want to have the Bookinfo demo deployed to two clusters.

You'll want to first have Istio installed for multi-cluster before installing the Bookinfo demo.

The core components, including reviews-v1 and reviews-v2, are deployed to the management plane cluster, while reviews-v3 is deployed on the remote cluster.

Deploy part of the bookinfo application to the management-cluster-context cluster:

Be sure to switch the kubeconfig context to the management-cluster-context

# Set the proper context value for each cluster

kubectl config use-context $MGMT_CONTEXT

kubectl create ns bookinfo
kubectl label namespace bookinfo istio-injection=enabled
# we deploy everything except reviews-v3 to the management plane cluster
kubectl apply -n bookinfo -f -l 'app,version notin (v3)'
kubectl apply -n bookinfo -f -l 'account'

Now deploy only reviews-v3 to your remote-cluster-context cluster:

kubectl config use-context $REMOTE_CONTEXT

kubectl create ns bookinfo
kubectl label namespace bookinfo istio-injection=enabled
kubectl apply -n bookinfo -f -l 'app,version in (v3)' 
kubectl apply -n bookinfo -f -l 'service=reviews' 
kubectl apply -n bookinfo -f -l 'account=reviews' 
kubectl apply -n bookinfo -f -l 'app=ratings' 
kubectl apply -n bookinfo -f -l 'account=ratings' 

Now you have Bookinfo demo set up for the rest of the guides. From here you can federate your two clusters, or start configuring multi-cluster access and traffic policy.