This tutorial describes how to setup ExternalDNS for usage within a GKE (Google Kuberentes Engine) cluster. Make sure to use >=0.11.0 version of ExternalDNS for this tutorial
If you prefer to try-out ExternalDNS in one of the existing environments you can skip this step
The following instructions use access scopes to provide ExternalDNS with the permissions it needs to manage DNS records within a single project, the organizing entity to allocate resources.
Note that since these permissions are associated with the instance, all pods in the cluster will also have these permissions. As such, this approach is not suitable for anything but testing environments.
This solution will only work when both CloudDNS and GKE are provisioned in the same project. If the CloudDNS zone is in a different project, this solution will not work.
Setup your environment to work with Google Cloud Platform. Fill in your variables as needed, e.g. target project.
# set variables to the appropriate desired values
PROJECT_ID="my-external-dns-test"
REGION="europe-west1"
ZONE="europe-west1-d"
ClOUD_BILLING_ACCOUNT="<my-cloud-billing-account>"
# set default settings for project
gcloud config set project $PROJECT_ID
gcloud config set compute/region $REGION
gcloud config set compute/zone $ZONE
# enable billing and APIs if not done already
gcloud beta billing projects link $PROJECT_ID \
--billing-account $BILLING_ACCOUNT
gcloud services enable "dns.googleapis.com"
gcloud services enable "container.googleapis.com"
gcloud container clusters create $GKE_CLUSTER_NAME \
--num-nodes 1 \
--scopes "https://www.googleapis.com/auth/ndev.clouddns.readwrite"
WARNING: Note that this cluster will use the default compute engine GSA that contians the overly permissive project editor (roles/editor
) role. So essentially, anything on the cluster could potentially grant escalated privileges. Also, as mentioned earlier, the access scope ndev.clouddns.readwrite
will allow anything running on the cluster to have read/write permissions on all Cloud DNS zones within the same project.
Create a DNS zone which will contain the managed DNS records. If using your own domain that was registered with a third-party domain registrar, you should point your domain's name servers to the values under the nameServers
key. Please consult your registrar's documentation on how to do that. This tutorial will use example domain of example.com
.
gcloud dns managed-zones create "example-com" --dns-name "example.com." \
--description "Automatically managed zone by kubernetes.io/external-dns"
Make a note of the nameservers that were assigned to your new zone.
gcloud dns record-sets list \
--zone "example-com" --name "example.com." --type NS
Outputs:
NAME TYPE TTL DATA
example.com. NS 21600 ns-cloud-e1.googledomains.com.,ns-cloud-e2.googledomains.com.,ns-cloud-e3.googledomains.com.,ns-cloud-e4.googledomains.com.
In this case it's ns-cloud-{e1-e4}.googledomains.com.
but your's could slightly differ, e.g. {a1-a4}
, {b1-b4}
etc.
More often, following best practices in regards to security and operations, Cloud DNS zones will be managed in a separate project from the Kubernetes cluster. This section shows how setup ExternalDNS to access Cloud DNS from a different project. These steps will also work for single project scenarios as well.
ExternalDNS will need permissions to make changes to the Cloud DNS zone. There are three ways to configure the access needed:
Below are examples on how you can configure Cloud DNS and GKE in separate projects, and then use one of the three methods to grant access to ExternalDNS. Replace the environment variables to values that make sense in your environment.
For this process, create projects with the appropriate APIs enabled.
# set variables to appropriate desired values
GKE_PROJECT_ID="my-workload-project"
DNS_PROJECT_ID="my-cloud-dns-project"
ClOUD_BILLING_ACCOUNT="<my-cloud-billing-account>"
# enable billing and APIs for DNS project if not done already
gcloud config set project $DNS_PROJECT_ID
gcloud beta billing projects link $CLOUD_DNS_PROJECT \
--billing-account $ClOUD_BILLING_ACCOUNT
gcloud services enable "dns.googleapis.com"
# enable billing and APIs for GKE project if not done already
gcloud config set project $GKE_PROJECT_ID
gcloud beta billing projects link $CLOUD_DNS_PROJECT \
--billing-account $ClOUD_BILLING_ACCOUNT
gcloud services enable "container.googleapis.com"
Create a Cloud DNS zone in the designated DNS project.
gcloud dns managed-zones create "example-com" --project $DNS_PROJECT_ID \
--description "example.com" --dns-name="example.com." --visibility=public
If using your own domain that was registered with a third-party domain registrar, you should point your domain's name servers to the values under the nameServers
key. Please consult your registrar's documentation on how to do that. The example domain of example.com
will be used for this tutorial.
Create a GSA (Google Service Account) and grant it the minimal set of privileges required for GKE nodes:
GKE_CLUSTER_NAME="my-external-dns-cluster"
GKE_REGION="us-central1"
GKE_SA_NAME="worker-nodes-sa"
GKE_SA_EMAIL="$GKE_SA_NAME@${GKE_PROJECT_ID}.iam.gserviceaccount.com"
ROLES=(
roles/logging.logWriter
roles/monitoring.metricWriter
roles/monitoring.viewer
roles/stackdriver.resourceMetadata.writer
)
gcloud iam service-accounts create $GKE_SA_NAME \
--display-name $GKE_SA_NAME --project $GKE_PROJECT_ID
# assign google service account to roles in GKE project
for ROLE in ${ROLES[*]}; do
gcloud projects add-iam-policy-binding $GKE_PROJECT_ID \
--member "serviceAccount:$GKE_SA_EMAIL" \
--role $ROLE
done
Create a cluster using this service account and enable workload identity:
gcloud container clusters create $GKE_CLUSTER_NAME \
--project $GKE_PROJECT_ID --region $GKE_REGION --num-nodes 1 \
--service-account "$GKE_SA_EMAIL" \
--workload-pool "$GKE_PROJECT_ID.svc.id.goog"
In this method, the GSA (Google Service Account) that is associated with GKE worker nodes will be configured to have access to Cloud DNS.
WARNING: This will grant access to modify the Cloud DNS zone records for all containers running on cluster, not just ExternalDNS, so use this option with caution. This is not recommended for production environments.
GKE_SA_EMAIL="$GKE_SA_NAME@${GKE_PROJECT_ID}.iam.gserviceaccount.com"
# assign google service account to dns.admin role in the cloud dns project
gcloud projects add-iam-policy-binding $DNS_PROJECT_ID \
--member serviceAccount:$GKE_SA_EMAIL \
--role roles/dns.admin
After this, follow the steps in Deploy ExternalDNS. Make sure to set the --google-project
flag to match the Cloud DNS project name.
In this scenario, a new GSA (Google Service Account) is created that has access to the CloudDNS zone. The credentials for this GSA are saved and installed as a Kubernetes secret that will be used by ExternalDNS.
This allows only containers that have access to the secret, such as ExternalDNS to update records on the Cloud DNS Zone.
DNS_SA_NAME="external-dns-sa"
DNS_SA_EMAIL="$DNS_SA_NAME@${GKE_PROJECT_ID}.iam.gserviceaccount.com"
# create GSA used to access the Cloud DNS zone
gcloud iam service-accounts create $DNS_SA_NAME --display-name $DNS_SA_NAME
# assign google service account to dns.admin role in cloud-dns project
gcloud projects add-iam-policy-binding $DNS_PROJECT_ID \
--member serviceAccount:$DNS_SA_EMAIL --role "roles/dns.admin"
Generate static credentials from the ExternalDNS GSA.
# download static credentials
gcloud iam service-accounts keys create /local/path/to/credentials.json \
--iam-account $DNS_SA_EMAIL
Create a Kubernetes secret with the credentials in the same namespace of ExternalDNS.
kubectl create secret generic "external-dns" --namespace ${EXTERNALDNS_NS:-"default"} \
--from-file /local/path/to/credentials.json
After this, follow the steps in Deploy ExternalDNS. Make sure to set the --google-project
flag to match Cloud DNS project name. Make sure to uncomment out the section that mounts the secret to the ExternalDNS pods.
Workload Identity allows workloads in your GKE cluster to impersonate GSA (Google Service Accounts) using KSA (Kubernetes Service Accounts) configured during deployemnt. These are the steps to use this feature with ExternalDNS.
DNS_SA_NAME="external-dns-sa"
DNS_SA_EMAIL="$DNS_SA_NAME@${GKE_PROJECT_ID}.iam.gserviceaccount.com"
gcloud iam service-accounts create $DNS_SA_NAME --display-name $DNS_SA_NAME
gcloud projects add-iam-policy-binding $DNS_PROJECT_ID \
--member serviceAccount:$DNS_SA_EMAIL --role "roles/dns.admin"
Add an IAM policy binding bewtween the workload identity GSA and ExternalDNS GSA. This will link the ExternalDNS KSA to ExternalDNS GSA.
gcloud iam service-accounts add-iam-policy-binding $DNS_SA_EMAIL \
--role "roles/iam.workloadIdentityUser" \
--member "serviceAccount:$GKE_PROJECT_ID.svc.id.goog[${EXTERNALDNS_NS:-"default"}/external-dns]"
Deploy ExternalDNS with the following steps below, documented under Deploy ExternalDNS. Set the --google-project
flag to the Cloud DNS project name.
Add the proper workload identity annotation to the ExternalDNS KSA.
kubectl annotate serviceaccount "external-dns" \
--namespace ${EXTERNALDNS_NS:-"default"} \
"iam.gke.io/gcp-service-account=$DNS_SA_EMAIL"
Update the Pod spec to schedule the workloads on nodes that use Workload Identity and to use the annotated Kubernetes service account.
kubectl patch deployment "external-dns" \
--namespace ${EXTERNALDNS_NS:-"default"} \
--patch \
'{"spec": {"template": {"spec": {"nodeSelector": {"iam.gke.io/gke-metadata-server-enabled": "true"}}}}}'
After all of these steps you may see several messages with googleapi: Error 403: Forbidden, forbidden
. After several minutes when the token is refreshed, these error messages will go away, and you should see info messages, such as: All records are already up to date
.
Then apply the following manifests file to deploy ExternalDNS.
apiVersion: v1
kind: ServiceAccount
metadata:
name: external-dns
labels:
app.kubernetes.io/name: external-dns
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: external-dns
labels:
app.kubernetes.io/name: external-dns
rules:
- apiGroups: [""]
resources: ["services","endpoints","pods","nodes"]
verbs: ["get","watch","list"]
- apiGroups: ["extensions","networking.k8s.io"]
resources: ["ingresses"]
verbs: ["get","watch","list"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
name: external-dns-viewer
labels:
app.kubernetes.io/name: external-dns
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: external-dns
subjects:
- kind: ServiceAccount
name: external-dns
namespace: default # change if namespace is not 'default'
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: external-dns
labels:
app.kubernetes.io/name: external-dns
spec:
strategy:
type: Recreate
selector:
matchLabels:
app.kubernetes.io/name: external-dns
template:
metadata:
labels:
app.kubernetes.io/name: external-dns
spec:
serviceAccountName: external-dns
containers:
- name: external-dns
image: registry.k8s.io/external-dns/external-dns:v0.13.5
args:
- --source=service
- --source=ingress
- --domain-filter=example.com # will make ExternalDNS see only the hosted zones matching provided domain, omit to process all available hosted zones
- --provider=google
- --log-format=json # google cloud logs parses severity of the "text" log format incorrectly
# - --google-project=my-cloud-dns-project # Use this to specify a project different from the one external-dns is running inside
- --google-zone-visibility=public # Use this to filter to only zones with this visibility. Set to either 'public' or 'private'. Omitting will match public and private zones
- --policy=upsert-only # would prevent ExternalDNS from deleting any records, omit to enable full synchronization
- --registry=txt
- --txt-owner-id=my-identifier
# # uncomment below if static credentials are used
# env:
# - name: GOOGLE_APPLICATION_CREDENTIALS
# value: /etc/secrets/service-account/credentials.json
# volumeMounts:
# - name: google-service-account
# mountPath: /etc/secrets/service-account/
# volumes:
# - name: google-service-account
# secret:
# secretName: external-dns
Create the deployment for ExternalDNS:
kubectl create --namespace "default" --filename externaldns.yaml
The following will deploy a small nginx server that will be used to demonstrate that ExternalDNS is working.
Create the following sample application to test that ExternalDNS works. This example will provision a L4 load balancer.
apiVersion: v1
kind: Service
metadata:
name: nginx
annotations:
# change nginx.example.com to match an appropriate value
external-dns.alpha.kubernetes.io/hostname: nginx.example.com
spec:
type: LoadBalancer
ports:
- port: 80
targetPort: 80
selector:
app: nginx
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx
spec:
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- image: nginx
name: nginx
ports:
- containerPort: 80
Create the deployment and service objects:
kubectl create --namespace "default" --filename nginx.yaml
After roughly two minutes check that a corresponding DNS record for your service was created.
gcloud dns record-sets list --zone "example-com" --name "nginx.example.com."
Example output:
NAME TYPE TTL DATA
nginx.example.com. A 300 104.155.60.49
nginx.example.com. TXT 300 "heritage=external-dns,external-dns/owner=my-identifier"
Note created TXT
record alongside A
record. TXT
record signifies that the corresponding A
record is managed by ExternalDNS. This makes ExternalDNS safe for running in environments where there are other records managed via other means.
Let's check that we can resolve this DNS name. We'll ask the nameservers assigned to your zone first.
dig +short @ns-cloud-e1.googledomains.com. nginx.example.com.
104.155.60.49
Given you hooked up your DNS zone with its parent zone you can use curl
to access your site.
curl nginx.example.com
Let's check that Ingress works as well. Create the following Ingress.
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: nginx
spec:
rules:
- host: server.example.com
http:
paths:
- path: /
pathType: Prefix
backend:
service:
name: nginx
port:
number: 80
Create the ingress objects with:
kubectl create --namespace "default" --filename ingress.yaml
Note that this will ingress object will use the default ingress controller that comes with GKE to create a L7 load balancer in addition to the L4 load balancer previously with the service object. To use only the L7 load balancer, update the service manafest to change the Service type to NodePort
and remove the ExternalDNS annotation.
After roughly two minutes check that a corresponding DNS record for your Ingress was created.
gcloud dns record-sets list \
--zone "example-com" \
--name "server.example.com." \
Output:
NAME TYPE TTL DATA
server.example.com. A 300 130.211.46.224
server.example.com. TXT 300 "heritage=external-dns,external-dns/owner=my-identifier"
Let's check that we can resolve this DNS name as well.
dig +short @ns-cloud-e1.googledomains.com. server.example.com.
130.211.46.224
Try with curl
as well.
curl server.example.com
Make sure to delete all Service and Ingress objects before terminating the cluster so all load balancers get cleaned up correctly.
kubectl delete service nginx
kubectl delete ingress nginx
Give ExternalDNS some time to clean up the DNS records for you. Then delete the managed zone and cluster.
gcloud dns managed-zones delete "example-com"
gcloud container clusters delete "external-dns"