Atlassian Is Finally Sunsetting Jira Data Center
Sep 15, 2025
It finally happened. Last week Atlassian officially announced the end of life for Jira Data Center. We all knew this was coming, despite Atlassian repeatedly saying it would keep supporting Data Center.
TL;DR Takeaways
- Atlassian is ending support for Jira Data Center on March 28, 2029.
- This is part of their push toward a “Cloud-first” strategy.
- Customers have to migrate to the Cloud or find an alternative.
- This is a controversial move, especially for large enterprises.
What’s Happening?
Atlassian is ending support for most Data Center products, not just Jira. The main exceptions are Bitbucket Data Center and Jira Align Data Center. Here are the key dates:
- March 30, 2026: End of new Data Center license sales for new customers.
- March 30, 2028: End of new Data Center license sales for existing customers.
- March 28, 2029: End of support for all Data Center products.
This is the clearest signal yet that Atlassian is all-in on Cloud.
Why Is This Happening?
Atlassian has been pushing Cloud for a while. It is clearly where the money is, and where they make the most profit. This is all part of their “Cloud-first” strategy. There are some technical reasons it makes sense too. In theory, it means they can:
- Ship faster: A single platform makes it easier to develop and deploy new features, especially things like Rovo and the other AI-powered tools.
- Improve security: Centralized infrastructure gives Atlassian more control over security standards.
- Simplify the user experience: Customers get a more consistent experience across products.
So from a business perspective, it makes sense. I am less convinced by the technical argument, but I am saying that from a place of very limited experience. I do not think supporting and porting features from Jira Cloud to Jira Data Center is some impossible technical hurdle.
My Thoughts
I have mixed feelings about this. On one hand, I get it. Atlassian wants to focus its resources on Cloud. Their Cloud products have come a long way, and the new AI features are impressive. Which makes sense from a financial perspective, if nothing else.
On the other hand, this is going to hurt for a lot of large enterprises, especially those in highly regulated industries like finance and government. They have legitimate reasons for wanting to keep their data on-premise. Moving to Cloud is not always a simple lift-and-shift.
Plus I’m just generally not a fan of any move that takes choice away from the customer. Rather than killing Data Center and forcing the issue, I think the strategy should have been to make Cloud good enough that customers wanted to move. Carrot, not stick.
I’ve worked with a lot of companies that have heavily customized their Jira Data Center instances. Migrating to the Cloud will be a massive undertaking for them, and some of their customizations might not even be possible in the Cloud environment.
I’ve also been in the game long enough to remember when Cloud did not scale at all. A big part of the business of the first company I worked for in the Atlassian space was migrating customers who had started on Cloud, outgrown it, and needed us to move them to On-Prem. Atlassian’s Cloud performance has improved a lot since then, but those memories die hard for a lot of old-timers in the ecosystem.
I am also kind of biased. I probably wouldn’t be where I am today if I wasn’t able to run my own little test instance of Jira Server to learn from. I think that’s how a lot of admins got their start.
I don’t want to sound like I am being too negative here though. Cloud has come a LONG way. And it really is the best option for like, 95% of their customers. What that means for the remaining 5%, we’ll see. There will certainly be a non-zero number of customers that use this as an excuse to migrate to something else, though the competitive options that allow you to self-host are still pretty limited. But I am sure the Big Brain Business people at Atlassian did the math and decided that small potential loss was worth the larger margins and potential gain of Cloud.
Industry Reactions
The Data Center EOL announcement has been met with a mix of resignation and frustration from the Atlassian community. A lot of people saw this coming after the end of Server support, but the official announcement still stings.
The general sentiment looks something like this:
- Forced march to Cloud: Many users, especially those in large, complex organizations, feel like they are being forced into Cloud whether they are ready or not. For companies with strict data residency requirements or deep integrations with on-premise systems, this is a major concern.
- Loss of control and customization: A common fear is losing the deep customization and control that Data Center provides. Many of the powerful Marketplace apps that admins rely on have limited functionality in Cloud, and the ability to directly access the database and file system is gone.
- The future of the Jira admin: There is also a discussion about the changing role of the Jira administrator. With Atlassian managing the infrastructure in Cloud, the role will likely shift from a technical, hands-on admin to a more strategic, business-focused role.
- Is it all about the money? Some in the community see this as a way for Atlassian to increase revenue through Cloud subscriptions, which can be more expensive for large organizations.
There is a lot of concern, but also a sense of inevitability. The long runway to 2029 gives companies time to plan, but for heavily customized Data Center customers, this is going to be a long and messy migration. (which, hey, if you need someone to help out with that I am available).