This tutorial describes how to setup ExternalDNS for usage in conjunction with a Headless service.
The main use cases that inspired this feature is the necessity for fixed addressable hostnames with services, such as Kafka when trying to access them from outside the cluster. In this scenario, quite often, only the Node IP addresses are actually routable and as in systems like Kafka more direct connections are preferable.
We will go through a small example of deploying a simple Kafka with use of a headless service.
A simple deploy could look like this:
apiVersion: apps/v1
kind: Deployment
metadata:
name: exeternal-dns
spec:
strategy:
type: Recreate
selector:
matchLabels:
app: external-dns
template:
metadata:
labels:
app: external-dns
spec:
containers:
- name: external-dns
image: registry.opensource.zalan.do/teapot/external-dns:latest
args:
- --log-level=debug
- --source=service
- --source=ingress
- --namespace=dev
- --domain-filter=example.org.
- --provider=aws
- --registry=txt
- --txt-owner-id=dev.example.org
apiVersion: v1
kind: ServiceAccount
metadata:
name: external-dns
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: ClusterRole
metadata:
name: external-dns
rules:
- apiGroups: [""]
resources: ["services"]
verbs: ["get","watch","list"]
- apiGroups: [""]
resources: ["pods"]
verbs: ["get","watch","list"]
- apiGroups: ["extensions"]
resources: ["ingresses"]
verbs: ["get","watch","list"]
- apiGroups: [""]
resources: ["nodes"]
verbs: ["list"]
---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: ClusterRoleBinding
metadata:
name: external-dns-viewer
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: external-dns
subjects:
- kind: ServiceAccount
name: external-dns
namespace: default
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: exeternal-dns
spec:
strategy:
type: Recreate
selector:
matchLabels:
app: external-dns
template:
metadata:
labels:
app: external-dns
spec:
serviceAccountName: external-dns
containers:
- name: external-dns
image: registry.opensource.zalan.do/teapot/external-dns:latest
args:
- --log-level=debug
- --source=service
- --source=ingress
- --namespace=dev
- --domain-filter=example.org.
- --provider=aws
- --registry=txt
- --txt-owner-id=dev.example.org
First lets deploy a Kafka Stateful set, a simple example(a lot of stuff is missing) with a headless service called kafka-hsvc
apiVersion: apps/v1beta1
kind: StatefulSet
metadata:
name: kafka
spec:
serviceName: ksvc
replicas: 3
template:
metadata:
labels:
component: kafka
spec:
containers:
- name: kafka
image: confluent/kafka
ports:
- containerPort: 9092
hostPort: 9092
name: external
command:
- bash
- -c
- " export DOMAIN=$(hostname -d) && \
export KAFKA_BROKER_ID=$(echo $HOSTNAME|rev|cut -d '-' -f 1|rev) && \
export KAFKA_ZOOKEEPER_CONNECT=$ZK_CSVC_SERVICE_HOST:$ZK_CSVC_SERVICE_PORT && \
export KAFKA_ADVERTISED_LISTENERS=PLAINTEXT://$HOSTNAME.example.org:9092 && \
/etc/confluent/docker/run"
volumeMounts:
- name: datadir
mountPath: /var/lib/kafka
volumeClaimTemplates:
- metadata:
name: datadir
annotations:
volume.beta.kubernetes.io/storage-class: st1
spec:
accessModes: [ "ReadWriteOnce" ]
resources:
requests:
storage: 500Gi
Very important here, is to set the hostport
(only works if the PodSecurityPolicy allows it)! and in case your app requires an actual hostname inside the container, unlike Kafka, which can advertise on another address, you have to set the hostname yourself.
Now we need to define a headless service to use to expose the Kafka pods. There are generally two approaches to use expose the nodeport of a Headless service:
- Add
--fqdn-template={{name}}.example.org
- Use a full annotation
If you go with #1, you just need to define the headless service, here is an example of the case #2:
apiVersion: v1
kind: Service
metadata:
name: ksvc
annotations:
external-dns.alpha.kubernetes.io/hostname: example.org
spec:
ports:
- port: 9092
name: external
clusterIP: None
selector:
component: kafka
This will create 3 dns records:
kafka-0.example.org
kafka-1.example.org
kafka-2.example.org
If you set --fqdn-template={{name}}.example.org
you can ommit the annotation.
Generally it is a better approach to use --fqdn-template={{name}}.example.org
, because then
you would get the service name inside the generated A records:
kafka-0.ksvc.example.org
kafka-1.ksvc.example.org
kafka-2.ksvc.example.org