Some time ago our customer asked my team to help them with the migration from server to cloud. The situation was special, because our customer was separating from larger group of companies and they didn't have admin rights in Jira and Confluence.
Prep phase
After some discussions we were promised to get the copy of the production Jira and Confluence, where all the data, which were not connected to our customer, were deleted. So, first of all we had to cooperate with our customer and collect a list of items, which should be migrated, so that we can obtain them in the cleaned copy of the production systems. After we had reviewed all the data, we were able to define precise list of Jira projects, boards, dashboards, filters and Confluence spaces for the migration.
It was possible to get admin rights in the cleaned copy of the production systems, so we continued with the next phase of analysis - how the customer works with the systems, which applications are they really using and how to migrate everything to cloud.
We found out the Advanced Roadmaps together with a cloud automation functionality can replace all previously installed applications for automations and planning. Situation with integrations was easy, because they were only using the Microsoft Teams together with Jira.
As a next step we had to specify, which users belonged to our customer (only approx. 300 of 2000).
We have also prepared a new list of project roles and user groups and a new system of permissions and notifications, because the original setup didn't work for our customer and they wanted to change it.
We were also able to get rid of tens of custom fields and keep only minimum of them.
Test phase
Test migration was necessary not only for users, so that they were able to try to work in cloud Jira and Confluence, but also for us to be able to create very detailed runbook, so that we would be able to transform the cleaned copy of production systems to a Jira and Confluence migratable to the cloud and suitable for our customer.
Migrate phase
We've maximally used the prepared runbook. It took us three days, but we succeeded. Before we started, we have asked the admins of the original Jira and Confluence to set the projects and spaces as read only, so that no changes were made during and after the migration. Later (after a few weeks) the original data was deleted.
After migrate phase
We are happy that our customer is able to continue to expand, a lot of new functionality and processes have been implemented since the migration (test management, asset management...). They have their own Jira and Confluence in the cloud and they can do whatever they like in the future.
2 comments