»Single Consul Datacenter in Multiple Kubernetes Clusters
Requires consul-helm v0.32.1 or higher.
This page describes how to deploy a single Consul datacenter in multiple Kubernetes clusters, with both servers and clients running in one cluster, and only clients running in the rest of the clusters. In this example, we will use two Kubernetes clusters, but this approach could be extended to using more than two.
Note: This deployment topology requires that your Kubernetes clusters have a flat network for both pods and nodes, so that pods or nodes from one cluster can connect to pods or nodes in another.
»Deploying Consul servers and clients in the first cluster
First, we will deploy the Consul servers with Consul clients in the first cluster. For that, we will use the following Helm configuration:
global: datacenter: dc1 tls: enabled: true enableAutoEncrypt: true acls: manageSystemACLs: true gossipEncryption: secretName: consul-gossip-encryption-key secretKey: keyconnectInject: enabled: truecontroller: enabled: trueui: service: type: NodePort
Note that we are deploying in a secure configuration, with gossip encryption, TLS for all components, and ACLs. We are enabling the Consul Service Mesh and the controller for CRDs so that we can use them to later verify that our services can connect with each other across clusters.
We're also setting UI's service type to be
This is needed so that we can connect to servers from another cluster without using the pod IPs of the servers,
which are likely going to change.
To deploy, first we need to generate the Gossip encryption key and save it as a Kubernetes secret.
$ kubectl create secret generic consul-gossip-encryption-key --from-literal=key=$(consul keygen)
Now we can install our Consul cluster with Helm:
$ helm install cluster1 -f cluster1-config.yaml hashicorp/consul
Note: The Helm release name must be unique for each Kubernetes cluster. That is because the Helm chart will use the Helm release name as a prefix for the ACL resources that it creates, such as tokens and auth methods. If the names of the Helm releases are the same, the Helm installation in subsequent clusters will clobber existing ACL resources.
Once the installation finishes and all components are running and ready, we need to extract the gossip encryption key we've created, the CA certificate and the ACL bootstrap token generated during installation, so that we can apply them to our second Kubernetes cluster.
kubectl get secret consul-gossip-encryption-key cluster1-consul-ca-cert cluster1-consul-bootstrap-acl-token -o yaml > cluster1-credentials.yaml
»Deploying Consul clients in the second cluster
Note: If multiple Kubernetes clusters will be joined to the Consul Datacenter, then the following instructions will need to be repeated for each additional Kubernetes cluster.
Now we can switch to the second Kubernetes cluster where we will deploy only the Consul clients that will join the first Consul cluster.
First, we need to apply credentials we've extracted from the first cluster to the second cluster:
$ kubectl apply -f cluster1-credentials.yaml
To deploy in the second cluster, we will use the following Helm configuration:
global: enabled: false datacenter: dc1 acls: manageSystemACLs: true bootstrapToken: secretName: cluster1-consul-bootstrap-acl-token secretKey: token gossipEncryption: secretName: consul-gossip-encryption-key secretKey: key tls: enabled: true enableAutoEncrypt: true caCert: secretName: cluster1-consul-ca-cert secretKey: tls.crtexternalServers: enabled: true # This should be any node IP of the first k8s cluster hosts: ["10.0.0.4"] # The node port of the UI's NodePort service httpsPort: 31557 tlsServerName: server.dc1.consul # The address of the kube API server of this Kubernetes cluster k8sAuthMethodHost: https://kubernetes.example.com:443client: enabled: true join: ["provider=k8s kubeconfig=/consul/userconfig/cluster1-kubeconfig/kubeconfig label_selector=\"app=consul,component=server\""] extraVolumes: - type: secret name: cluster1-kubeconfig load: falseconnectInject: enabled: true
Note that we're referencing secrets from the first cluster in ACL, gossip, and TLS configuration.
Next, we need to set up the
refer to the IP and port of the UI's NodePort service deployed in the first cluster.
externalServers.hosts to any Node IP of the first cluster,
which you can see by running
kubectl get nodes -o wide.
externalServers.httpsPort to the
nodePort of the
In our example, the port is
$ kubectl get service cluster1-consul-ui --context cluster1NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGEcluster1-consul-ui NodePort 10.0.240.80 <none> 443:31557/TCP 40h
We set the
server.dc1.consul. This the DNS SAN
(Subject Alternative Name) that is present in the Consul server's certificate.
We need to set it because we're connecting to the Consul servers over the node IP,
but that IP isn't present in the server's certificate.
To make sure that the hostname verification succeeds during the TLS handshake, we need to set the TLS
server name to a DNS name that is present in the certificate.
Next, we need to set
externalServers.k8sAuthMethodHost to the address of the second Kubernetes API server.
This should be the address that is reachable from the first cluster, and so it cannot be the internal DNS
available in each Kubernetes cluster. Consul needs it so that
consul login with the Kubernetes auth method will work
from the second cluster.
More specifically, the Consul server will need to perform the verification of the Kubernetes service account
consul login is called, and to verify service accounts from the second cluster it needs to
reach the Kubernetes API in that cluster.
The easiest way to get it is to set it from your
kubeconfig by running
kubectl config view and grabbing
the value of
cluster.server for the second cluster.
Lastly, we need to set up the clients so that they can discover the servers in the first cluster.
For this, we will use Consul's cloud auto-join feature
for the Kubernetes provider.
To use it we need to provide a way for the Consul clients to reach the first Kubernetes cluster.
To do that, we need to save the
kubeconfig for the first cluster as a Kubernetes secret in the second cluster
and reference it in the
clients.join value. Note that we're making that secret available to the client pods
by setting it in
Note: The kubeconfig you're providing to the client should have minimal permissions. The cloud auto-join provider will only need permission to read pods. Please see Kubernetes Cloud auto-join for more details.
Now we're ready to install!
helm install cluster2 -f cluster2-config.yaml hashicorp/consul
»Verifying the Consul Service Mesh works
When Transparent proxy is enabled, services in one Kubernetes cluster that need to communicate with a service in another Kubernetes cluster must have a explicit upstream configured through the "consul.hashicorp.com/connect-service-upstreams" annotation.
Now that we have our Consul cluster in multiple k8s clusters up and running, we will deploy two services and verify that they can connect to each other.
First, we'll deploy
static-server service in the first cluster:
---apiVersion: consul.hashicorp.com/v1alpha1kind: ServiceIntentionsmetadata: name: static-serverspec: destination: name: static-server sources: - name: static-client action: allow---apiVersion: v1kind: Servicemetadata: name: static-serverspec: type: ClusterIP selector: app: static-server ports: - protocol: TCP port: 80 targetPort: 8080---apiVersion: v1kind: ServiceAccountmetadata: name: static-server---apiVersion: apps/v1kind: Deploymentmetadata: name: static-serverspec: replicas: 1 selector: matchLabels: app: static-server template: metadata: name: static-server labels: app: static-server annotations: "consul.hashicorp.com/connect-inject": "true" spec: containers: - name: static-server image: hashicorp/http-echo:latest args: - -text="hello world" - -listen=:8080 ports: - containerPort: 8080 name: http serviceAccountName: static-server
Note that we're defining a Service intention so that our services are allowed to talk to each other.
Then we'll deploy
static-client in the second cluster with the following configuration:
apiVersion: v1kind: Servicemetadata: name: static-clientspec: selector: app: static-client ports: - port: 80---apiVersion: v1kind: ServiceAccountmetadata: name: static-client---apiVersion: apps/v1kind: Deploymentmetadata: name: static-clientspec: replicas: 1 selector: matchLabels: app: static-client template: metadata: name: static-client labels: app: static-client annotations: "consul.hashicorp.com/connect-inject": "true" "consul.hashicorp.com/connect-service-upstreams": "static-server:1234" spec: containers: - name: static-client image: curlimages/curl:latest command: [ "/bin/sh", "-c", "--" ] args: [ "while true; do sleep 30; done;" ] serviceAccountName: static-client
Once both services are up and running, we can connect to the
$ kubectl exec deploy/static-client -- curl -s localhost:1234"hello world"