#+TITLE: Deploying demo jira instance
#+AUTHOR: James Blair <jablair@redhat.com>
#+DATE: <2023-03-10 Fri 10:15>

This guide will outline the steps to follow to deploy a demo jira instance to an existing kubernetes cluster. For our purposes that cluster will be an existing [[https://aws.amazon.com/rosa/][ROSA]] cluster running in AWS ~ap-southeast-1~.


* Login to cluster

As mentioned above we have an existing OpenShift cluster to use for this demo, we will need to login to the cli to automate the remainder of the jira setup.

#+NAME: Login to openshift
#+begin_src tmate
oc login --kubeconfig ~/.kube/rosa --token=<token> --server=<server>
#+end_src


* Create kubernetes namespace

Our first step is to create a kubernetes [[https://kubernetes.io/docs/concepts/overview/working-with-objects/namespaces/][namespace]] for our jira deployment.

#+NAME: Create jira namespace
#+begin_src tmate
kubectl --kubeconfig ~/.kube/rosa create namespace jira
#+end_src


* Build and deploy jira

Once we have a namespace we can use a one line ~oc~ command to create a build process in OpenShift based on our github repository containing a Dockerfile.

This will build a container image within OpenShift and then create a Deployment of that image which will give us a single running jira pod.

Note: This deployment will not be backed by persistent storage but for demo purposes this is fine. Do not use this in production...

#+NAME: Build and deploy jira
#+begin_src tmate
# Initiate the build from github
oc --kubeconfig ~/.kube/rosa --namespace jira --name jira new-app https://github.com/jmhbnz/docker-atlassian-jira

# Watch the progress
oc --kubeconfig ~/.kube/rosa --namespace jira logs --follow buildconfig/jira
#+end_src


Once the container image has built successfully we can verify the jira instance is running by checking the pod status.

#+NAME: Check pod status
#+begin_src tmate
kubectl --kubeconfig ~/.kube/rosa --namespace jira get pods
#+end_src


* Expose jira deployment

With our jira instance now running within our cluster we can create a ~route~ to expose it outside the cluster.

#+NAME: Expose jira deployment
#+begin_src tmate
oc --kubeconfig ~/.kube/rosa --namespace jira expose service jira
#+end_src


With our route created lets retrieve that and perform the first time setup for jira. This is currently a manual process involving obtaining a trial license from [[https://my.atlassian.com/product][atlassian]].

#+NAME: Retrieve jira route
#+begin_src tmate
echo http://$(oc --kubeconfig ~/.kube/
rosa --namespace jira get route | grep apps.com | awk '{print $2}')
#+end_src