Migrating Jira to Kubernetes

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

2025년 6월 17일 — Kyungmin Park

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

💡

Jira 9.12.15 to 10.3.6

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/jira

Get `values.yaml` command

helm show values altassian-data-center/jira --version 1.22.1 > values.yaml

Editing `values.yaml`

Modify according to your configuration.

Create namespace

kubectl create ns atlassian

Deploy helm chart

Jira version: 9.12.15

helm install jira \
 atlassian-data-center/jira \
 -f values.yaml \
 -n atlassian \
 --version 1.22.1

Check deployment

💡

The purpose of this initial startup is to intentionally enter an invalid database URL in order to create an empty Persistent Volume (PV) for Jira data(application-data, shared-home) migration.

Data Migration

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

Migration strategy

Jira deploy(replicas=1) with empty data(application-data, shared-home)
Get application-data directory from as-is jira instance(Rsync in to-be jira pod)
Get shared-home directory from as-is jira instance(Rsync in to-be jira pod)
Jira pod delete for restart
Check jira data
Scale the Jira pod from 1 to 2

Upgrade

Upgrade strategy

replicas 0(shutdown all Jira pod)
Helm chart upgrade

Shutdown all Jira pod

kubectl scale statefulset --replicas 0 -n atlassian jira

For a Rolling upgrade, adjusting the replicas is not necessary.

Helm chart

Jira version: 9.12.15 → 10.3.6

helm upgrade \
 jira atlassian-data-center/jira \
 --version 2.0.1 \
 -f values.yaml \
 -n atlassian

🗒️

During helm upgrade, pods are created based on the replicaCount in values.yaml, and the first pod performs the upgrade task.
The previous --replicas 0 setting is ignored.

Notes

Database unsupported collation

When PostgreSQL(15.7) was first deployed via Docker, the collation was set to en_US.utf8 by default.

Shutdown Jira

kubectl scale statefulset --replicas 0 -n atlassian jira

Restarting PostgreSQL to Reset Connections

Delete all pod

kubectl delete pod -n atlassian jira-db-0

Execute db pod

kubectl exec -it -n atlassian jira-db-0 – bash

Export the existing database

pg_dump -d <OLD-DATABASE> -Fc -f old.dump

(Optional) Drop the existing database

DROP DATABASE <OLD-DATABASE>;

Create new database

CREATE DATABASE <NEW-DATABASE>      
  LC_COLLATE='C.UTF-8'
  LC_CTYPE='C.UTF-8'
  ENCODING='UTF8'
  TEMPLATE=template0;

Restore the dump file to the new database

pg_restore -d <NEW-DATABASE> -Fc old.dump

Grant database privileges to an existing user

GRANT ALL PRIVILEGES ON DATABASE <NEW-DATABASE> TO <EXIST-USER>;

Start Jira

Switch to existing replicas

kubectl scale statefulset --replicas 2 -n atlassian jira