If your Jira runs on your own servers today, its clock is ticking. Atlassian has announced the end of Jira Data Center: it stopped selling Data Center to new customers on 30 March 2026, and Data Center reaches its end of life on 28 March 2029 (Atlassian's announcement). Server editions ended earlier. After 2029, the Jira you can buy runs in Atlassian's cloud.

For many teams that is fine. For others it is not an option at all.

Who this leaves without a home

  • Teams whose data must stay inside. Regulated industries, public sector, defence suppliers, and anyone with a contract or a law that says where project data may live.
  • Networks with no way out. Labs, factories, and offices that run air-gapped or behind strict egress rules.
  • Teams that want their own AI. If issue text and documentation can only go to a model you run, a hosted tracker makes that hard.
  • Teams that simply prefer to own their tools. Their own backups, their own upgrade schedule, their own servers.

Three years sounds like plenty. It is not much once you count an evaluation, a trial migration, a pilot team, training and the switch-over, all around normal work. The teams that move calmly are the ones that start early.

Specivo: a self-hosted Jira alternative

Specivo is a project tracker and knowledge base that you run yourself, designed to stay on your infrastructure:

  • Runs where you put it. Install it with Docker on a server you control, including fully air-gapped networks.
  • Issues and wiki in one place. Projects, sprints, versions, time tracking, custom fields, and a wiki with history, without a second product to connect.
  • Your AI, your choice. Connect a cloud model, a local model on your own hardware, or none at all. AI agents work through the same permissions as people.
  • Your upgrade schedule. You pin the version and upgrade when it suits you.

Your history comes with you

Moving trackers is only worth it if years of decisions come along. The Jira migration tool reads your projects through Jira's public REST API, from Cloud or Data Center, and writes a Specivo import bundle. It needs read access only and changes nothing in Jira.

Issues come across with their comments, change history, attachments and inline images, links between issues, parents and sub-tasks, worklogs, versions, components, sprints, custom fields, labels and watchers. Whatever cannot be carried over is listed in a report, so nothing disappears silently. The full list is in the documentation.

A move you can rehearse

You do not have to switch in one weekend and hope:

  1. Export one project or the whole site while Jira keeps running.
  2. Load it into a test instance with a dry run first, and look around with your team.
  3. Repeat as often as you like. Projects can move one at a time, and links between them still resolve.
  4. Switch when you are ready, with a final export that picks up what changed.

The Jira migration tool documentation walks through every step, the permissions the Jira account needs, and how each kind of data is mapped.

Requirements

  • Specivo 0.11.0 or later, installed with Docker.
  • Jira Cloud, or Jira Data Center reachable from where you run the tool.
  • An account that can read the projects you want to move.

The migration tool is released together with Specivo 0.11.0.

Start before the deadline does

Install Specivo on a spare server, export one real project, and see how your own history looks on the other side. If you are also moving from Redmine, there is a Redmine migration path too.

Jira and Atlassian are trademarks of Atlassian Pty Ltd. Specivo is not affiliated with or endorsed by Atlassian. Dates are as published by Atlassian at the time of writing; check Atlassian's page for the current ones.