The manifests within this repo can be used to deploy a basic Liferay stack onto an Azure Kubernetes cluster. The image used in the Elasticsearch manifest uses a build that was created by Liferay and includes the necessary plugins required to run Elasticsearch with Liferay.
This readme assumes that access to an AKS cluster is in place and that the AzureCLI tools have been installed and all noted commands are run locally. First connect to the Azure Portal using the following:
az login
Once authenticated you will need to connect to the AKS cluster. The required command can be found on the Connect tab of the AKS cluster and will follow a syntax like this:
az aks get-credentials --resource-group resourcegroup --name clustername
To deploy any of the manifests included in this repo into AKS first clone the repo to your local PC then run the following command:
kubectl apply -f X_deploy.yaml
Where X is the application manifest you wish to deploy. To delete any of the items defined within the manifest run the following:
kubectl delete -f X_deploy.yaml
To create a route to the Liferay instance first configure an ingress controller for the AKS cluster. There are many options available for setting up an ingress controller but in my case I deployed the Kubernetes NGINX ingress controller. This can be done by deploying the raw manifest content from the related git repo (https://github.com/kubernetes/ingress-nginx) using the following command:
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.1.3/deploy/static/provider/cloud/deploy.yaml
Once the ingress controller is in place then deploy the ingress manifest to create the route:
kubectl apply -f ingress.yaml
Once the route has been completed you can check on it's status with the following command:
kubectl get ingress
The output should look like this:
| NAME | CLASS | HOSTS | ADDRESS | PORTS | AGE |
|---|---|---|---|---|---|
| nginx-ingress | * | 52.139.7.62 | 80 | 7m30s |
Once the Address has been defined you should be able to access Liferay by putting that IP into your browser.