Skip to content
Permalink
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
131 lines (94 sloc) 6.29 KB

Atlassian Bamboo and Bitbucket images for GKE clusters

Atlassian is an enterprise software company that develops products are widely used by enterprise customers Atlassian products cover a wide range of products, but here we're gonna work with:

  • Bitbucket: a web based source code and project management solution
  • Bamboo: a continuous integration server which integrates with Bitbucket.

Requisites

You'll need an Atlassian account. If you need an account, go here. The account is required to obtain an evaluation license. If you don't own a license, you can request an evaluation license from the set up page of each product directly.

An GKE cluster able to run both software. We recommend a cluster with at least one n1-standard-2 machine. You can launch your cluster with this command: machine. You can launch your cluster with this command: gcloud alpha container clusters create $NAME --num-nodes=1 --enable-autoscaling --min-nodes=1 --max-nodes=3 --no-enable-cloud-logging -m n1-standard-2

See Atlassian requirements about that:

Configure kubecfg with GKE

To be able to deploy your workloads on GKE you will need to grant access to your cluster. To do so, we're going to create a Service Account, with the role type Container Engine Developer . A service account is very similar to an IAM group on AWS. You can create it in: https://console.cloud.google.com/iam-admin/serviceaccounts/

Depending of your situation, you may need to enable the Container Engine API on your GCP project, too.

Do not forget to click on Furnish a new key as JSON file and store it securely. The contents of the file will going to be used as a secret. To do so, you need to encode as base64 and paste in the file.

Obtain the secret template:

wget https://raw.githubusercontent.com/jbianquetti-nami/atlassian-k8s/master/bamboo/secret.yaml

Then edit the file to adapt it to your environment:

apiVersion: v1
kind: Secret
metadata:
  name: gcloud-svc-account
data:
  gcloud-cluster: --- Your GKE cluster name, BASE64 ENCODED ---
  gcloud-cluster-zone: --- Your GKE cluster zone , BASE64 ENCODED --- 
  gcloud-svc-account: --- CONTENTS OF THE JSON SVC ACCOUNT, BASE64 ENCODED --

Use a command like this to base64 encode your data:

echo -n somevalue | base64 

The secret will be use to create a config file to be used by kubecfg. By this way you're authorized to run kubectl/kubecfg commands

Create the secret in your cluster:

kubectl create -f secret.yaml

Install the Atlassian software

We're going to expose the services as service type LoadBalancer

To run Bitbucket, execute:

# Create the server
$ kubectl create -f https://raw.githubusercontent.com/jbianquetti-nami/atlassian-k8s/master/bitbucket/deployment.yaml
# Expose the web app on port 80 and git on port 7990
$ kubectl create -f  https://raw.githubusercontent.com/jbianquetti-nami/atlassian-k8s/master/bitbucket/svc.yml

To run Bamboo, execute:

# Create the server 
$ kubectl create -f https://raw.githubusercontent.com/jbianquetti-nami/atlassian-k8s/master/bamboo/deployment.yaml
# Expose the web app on port 80
$ kubectl create -f https://raw.githubusercontent.com/jbianquetti-nami/atlassian-k8s/master/bamboo/svc.yml

You'll need a pair of minutes while the pods and load balancer gets up and running.

Configure Bitbucket

# Use this command to find the Bitbucket public URL 
$ echo "http://$(kubectl get svc atlassian-bitbucket  -o jsonpath={.status.loadBalancer.ingress[].ip}).xip.io"

Point your browser to the addresses, this will correspond to Bitbucket service.

  • Click 'Next' and select 'I have a Bitbucket license key'. Introduce your Bitbucket license key.
  • Click 'Next' and introduce the username, Full name, email address and password to create an admin user.
  • Click 'Go to Bitbuket'. Once done, you'll be redirected to the Login page.
  • Introduce your admin login.
  • Click on 'Projects' and create a project called 'Kubecfg demo' , use 'a kubecfg demo' as 'Description' and click on 'Create project'
  • Click on 'Import repository' and click on the 'Git' icon and introduce 'https://github.com/jbianquetti-nami/kubecfg-nonroot-nginx' as URL. Click and 'Import repository'
  • Clone the repository to your elsewhere in your disk.

Configure Bamboo

# Use this command to find the Bamboo public URL 
$ echo "http://$(kubectl get svc atlassian-bamboo -o jsonpath={.status.loadBalancer.ingress[].ip}).xip.io"
  • Introduce your Bamboo license key and click on 'Custom install'.
  • Introduce the same user you've created for Bitbucket.
  • Wait a bit while Bamboo is being configured. Keep sure that your cluster meets the requirements of Bamboo to be installed. Otherwhise your install process can be stalled

Bamboo image includes a kubectl proxy listen on port 8001 as a sidecar container. This means that you can interact with your cluster using kubecfg commands.

Useful tips for debugging

If you face some issues trying to reproduce this steps, here's a few tips to look:

  • Bamboo requires a good amount of resources to run (see #Requisites section), so if you find that your Bamboo install process is stalled, try to use a n1-standard-4 machine
  • Check that your Atlassian licenses are still valid. They are valid for 30 days.
  • Check that your GKE service account is proper configured, and counts with at least the Container Engine Developer role access level.
  • Remember that values in a Kubernetes secret are Base64 enconded!
  • If you see this message invalid configuration: default cluster has no server defined means that your service account data is not correct. To figure out which is the problem, execute:
# Access to the Bamboo pod 
$ kubectl exec -it atlassian-bamboo-xxxxxx-xxxx  -c bamboo  bash
# Now, in the pod execute:
export CLOUDSDK_CONFIG=/var/atlassian/bamboo/gcloud
gcloud auth activate-service-account --key-file /var/lib/kubernetes/gcloud-svc-account/gcloud-svc-account
gcloud container clusters get-credentials `cat /var/lib/kubernetes/gcloud-svc-account/gcloud-cluster` --zone `cat /var/lib/kubernetes/gcloud-svc-account/gcloud-cluster-zone `
You can’t perform that action at this time.