Migrating Bamboo to Kubernetes

This document explains the process of migrating an existing Bamboo installation to Kubernetes.

Pre-configured resources

• Rook ceph(Block, Filesystem)
• cert-manager
• Nginx ingress controller
• DB(PostgreSQL)

Helm Chart

Version: v3.17.3(GoVersion: go1.23.7)

Add the Helm chart repository

helm repo add atlassian-data-center \
  https://atlassian.github.io/data-center-helm-charts

Check Atlassian product version with chart version

helm search repo atlassian-data-center/bamboo

Get values.yaml command

helm show values altassian-data-center/bamboo --version 1.21.1 > values.yaml

Editing values.yaml

Modify according to your configuration.

Create namespace

kubectl create ns atlassian

Deploy helm chart

Bamboo version: 9.6.11

helm install bamboo \
 atlassian-data-center/bamboo \
 -f values.yaml \
 -n atlassian \
 --version 1.21.1

Edit Bamboo statefulset scale

kubectl scale statefulset --replicas 0 -n atlassian bamboo

Data Migration

Process for importing existing application data and using it in the existing environment

Migration strategy

1. Bamboo deploy(replicas=1 → replicas=0) with empty data(application-data, shared-home)
2. Deploying a BusyBox pod with Bamboo's local-home and shared-home mounted (I used a Debian image instead of BusyBox, as I’m more familiar with it)
3. Get the application-data directory from the as-is Bamboo instance and rsync it into the Bamboo volume using a BusyBox pod
4. Get the shared-home directory from the as-is Bamboo instance and rsync it into the Bamboo volume using a BusyBox pod
5. Rescale Bamboo statefulset(replicas=0 → replicas=1)
6. Check Bamboo data
7. Rescale Bamboo statefulset(final replicasCount)

Why Multiple Bamboo Pods Don’t Appear in the Cluster List?

Atlassian explains it as follows:

At present Bamboo Data Center utilizes an active-passive clustering model. This architecture is not ideal where K8s deployments are concerned.
At present, Bamboo server cluster sizes comprising only 1 pod is the only supported topology for now.

Therefore, in this post, I’ll explain how to manually add cluster nodes.
(Even if you add cluster nodes manually, the settings are preserved as long as you apply a retain policy to the PVC)

How to Configure the N‑th Bamboo Node in the Cluster (Excluding the First Node)

1. On Node 1: Execute commands to copy the contents of the bamboo.cfg.xml file
2. On Node N: Execute commands to paste the copied contents into the bamboo.cfg.xml file
3. Restart the Pod: To apply the changes, delete the N‑th node’s pod using kubectl delete pod

Upgrade

Upgrade strategy

1. replicas 0(shutdown all Bamboo pod)
2. Set image.tag
3. Helm chart upgrade
4. Bamboo statefulset scaling

Shutdown all Bamboo pod

kubectl scale statefulset --replicas 0 -n atlassian bamboo

Set image.tag

If you previously set image.tag but the APP VERSION matches the version you want to upgrade to, proceed with this step.

image:

  # -- The Docker Crowd Docker image to use
  # https://hub.docker.com/r/atlassian/crowd
  #
  repository: atlassian/crowd

  # -- Image pull policy
  #
  pullPolicy: IfNotPresent

  # -- The docker image tag to be used. Defaults to appVersion in Chart.yaml
  #
  tag: "10.2.7"

Helm chart

Bamboo version: 9.6.11 → 10.2.7

helm upgrade bamboo \
 atlassian-data-center/bamboo \
 -f values.yaml \
 -n atlassian

Bamboo Agent

In a Kubernetes environment, when registering a Remote Agent via JMS, you need to configure both the Broker URL and the Broker client URL to use SSL.

Reference

Products upgrade - Atlassian DC Helm Charts

logo

Limitations - Atlassian DC Helm Charts

logo