Run Jira Data Center on Kubernetes Part One

In this post, I would like to share how I was able to run Jira Data Center on Kubernetes in an effect way.

First of all, you need to have a storage class that supports dynamic provisioning. If you don’t have one, you can create one by following this how to articles: Use Glusterfs for Dynamic Volume Provisioning in Kubernetes. In this example, I use Glusterfs for my persistent data.

Second of all, you need to have a database ready. If not, you can quickly setup one on Kubernetes as well. I have the example for creating Postgres:

	—
	kind: PersistentVolumeClaim
	apiVersion: v1
	metadata:
	name: jira-postgres-pvc
	labels:
	app: jira-postgres
	spec:
	accessModes:
	– ReadWriteOnce
	resources:
	requests:
	storage: 5Gi
	storageClassName: glusterfs

view raw k8s_jira_postgres_pv.yaml hosted with ❤ by GitHub

	—
	apiVersion: v1
	kind: ConfigMap
	metadata:
	name: jira-postgres-config
	labels:
	app: jira-postgres
	data:
	POSTGRES_DB: jira
	POSTGRES_USER: admin
	POSTGRES_PASSWORD: admin

view raw k8s_jira_postgres_configmap.yaml hosted with ❤ by GitHub

	—
	apiVersion: apps/v1
	kind: Deployment
	metadata:
	name: jira-postgres
	labels:
	app: jira-postgres
	spec:
	replicas: 1
	selector:
	matchLabels:
	app: jira-postgres
	template:
	metadata:
	labels:
	app: jira-postgres
	spec:
	containers:
	– name: postgres
	image: postgres:9.6
	imagePullPolicy: "IfNotPresent"
	ports:
	– containerPort: 5432
	envFrom:
	– configMapRef:
	name: jira-postgres-config
	volumeMounts:
	– mountPath: /var/lib/postgresql/data
	name: jira-postgres-pv
	volumes:
	– name: jira-postgres-pv
	persistentVolumeClaim:
	claimName: jira-postgres-pvc

view raw k8s_jira_postgres_deployment.yaml hosted with ❤ by GitHub

	—
	apiVersion: v1
	kind: Service
	metadata:
	name: jira-postgres
	labels:
	app: jira-postgres
	spec:
	type: ClusterIP
	ports:
	– port: 5432
	selector:
	app: jira-postgres

view raw k8s_jira_postgres_service.yaml hosted with ❤ by GitHub

By now, we should have all prerequisites (dynamic PV and database) ready. Lets start to build Jira Data Center.

Jira Data Center needs two home folders: One is local home folder, the other one is shared home folder. They both need to be stored on persistent volume, as pods are short living, but the data needs to stay permanently.

Step one: create a persistent volume for the shared home. As we use dynamic persistent volume provisioning, we only need to create a PVC (persistent volume claim), the PV (persistent volume) will be created on the fly.

	—
	kind: PersistentVolumeClaim
	apiVersion: v1
	metadata:
	name: jira-share-pvc
	labels:
	app: jira-share
	spec:
	accessModes:
	– ReadWriteMany
	storageClassName: glusterfs
	resources:
	requests:
	storage: 10Gi

view raw k8s_jira_sharedhome_pv.yaml hosted with ❤ by GitHub

Step two: create a config map which will be used to pass the environment variables into the containers. Adjust the settings accordingly. In the example, we use the official Jira Software docker images

	—
	apiVersion: v1
	kind: ConfigMap
	metadata:
	name: jira-config
	labels:
	app: jira
	data:
	JVM_MINIMUM_MEMORY: 2048m
	JVM_MAXIMUM_MEMORY: 2048m
	ATL_PROXY_NAME: jira-sandbox.mydomain.com
	ATL_PROXY_PORT: '32631'

view raw k8s_jira_configmap.yaml hosted with ❤ by GitHub

Step three: create statefulSet for Jira. As Jira cluster is a stateful application, here statefulSet fits better than deployment. e.g statefulSet supports ordinal startup and shutdown. The jira local home folder is created dynamically here, and it is tied to a specific pod based on the pod’s name. e.g jira-0, jira-1 …

	—
	apiVersion: apps/v1
	kind: StatefulSet
	metadata:
	name: jira
	spec:
	serviceName: jira
	replicas: 3
	selector:
	matchLabels:
	app: jira
	template:
	metadata:
	labels:
	app: jira
	spec:
	containers:
	– name: jira
	image: atlassian/jira-software:8.5
	readinessProbe:
	httpGet:
	path: /status
	port: 8080
	initialDelaySeconds: 120
	periodSeconds: 10
	livenessProbe:
	httpGet:
	path: /
	port: 8080
	initialDelaySeconds: 600
	periodSeconds: 10
	envFrom:
	– configMapRef:
	name: jira-config
	ports:
	– containerPort: 8080
	name: web
	volumeMounts:
	– name: local-home
	mountPath: /var/atlassian/application-data/jira
	– name: jira-share-pv
	mountPath: /var/atlassian/application-data/jira/shared-home
	volumes:
	– name: jira-share-pv
	persistentVolumeClaim:
	claimName: jira-share-pvc
	volumeClaimTemplates:
	– metadata:
	name: local-home
	spec:
	accessModes: [ "ReadWriteOnce" ]
	resources:
	requests:
	storage: 5Gi

view raw k8s_jira_statefulset.yaml hosted with ❤ by GitHub

Step four: create a service for Jira. You can either do NodePort or ClusterIP. If you have ingress controller, then ClusterIP should just be fine. In my case, I use traefik to route the traffics to Jira service if the hostname is jira-sande.mydomain.com.

	—
	apiVersion: v1
	kind: Service
	metadata:
	name: jira
	labels:
	app: jira
	spec:
	type: NodePort
	ports:
	– port: 8080
	targetPort: 8080
	selector:
	app: jira

view raw k8s_jira_service.yaml hosted with ❤ by GitHub

80% of the work should be done by now. The rest are just some configurations which I will show you in the part 2.