draw.io is one of the most widely used tools in the Atlassian ecosystem, and if your teams rely on it, making sure those diagrams survive the migration intact takes a few specific steps. This guide covers exactly what those steps are.
Why draw.io Deserves Its Own Place in Your Migration Plan
If you’re planning an Atlassian Data Center to Cloud migration, your checklist is long: Confluence content, Jira project configurations, automations, integrations, custom scripts, and apps. draw.io is one app that can’t be treated as an afterthought.
Atlassian’s Cloud Migration Assistance (CMA) tooling handles a lot of the heavy lifting when you move from Data Center to Cloud. But apps don’t always migrate cleanly on their own, and draw.io is a case where the standard migration carries most of the data across while leaving a few things that need manual attention.
The good news: draw.io’s migration process is well-documented and the tooling is solid. Understanding the process ahead of time is the difference between a clean cutover and a post-migration cleanup scramble.
If you haven’t already done broader pre-migration preparation, it’s worth reviewing our data cleanup guide. Getting your Confluence spaces and Jira projects into good shape before migration reduces complexity across the board, including for draw.io.
What the DC End-of-Life Timeline Means for You
Atlassian has set a clear sunset roadmap for Data Center:
- As of June 2026: New license sales have already stopped for new customers
- March 30, 2028: New license sales stop for existing customers
- March 30, 2029: Final day of Data Center support
That last date might feel distant, but migrations at scale take time to plan and execute. Organizations with thousands of users, multiple Atlassian instances, or complex app dependencies typically start planning 12 to 18 months before their target go-live. If you’re in that window, draw.io migration planning should already be on the list.
For teams evaluating migration support options, Atlassian offers tiered assistance: self-service tooling for teams under 1,000 users, the FastShift Program for 1,000 to 5,000 users, and Solution Design Acceleration for organizations over 5,000. We’ve covered FastShift and alternatives, if you’re weighing your options.
What Actually Happens to Your draw.io Diagrams?
Your draw.io diagram data isn’t stored in a separate database, and it is never sent to draw.io’s servers when you save your diagram. It lives in two places:
- Attachments: the diagram files themselves, stored as Confluence page attachments or Jira issue attachments
- Macros: the references embedded in pages and issues that render those diagrams
The behavior during migration differs between Confluence and Jira, so it’s worth understanding each separately.
In Confluence: the standard Atlassian migration automatically carries both attachments and macros across. Your diagrams don’t disappear. The issue is that page IDs change in the Cloud instance. If any of your diagrams include links to other Confluence pages, or if you use draw.io’s embed macro to pull a diagram into multiple pages, those references need to be updated to reflect the new IDs.
In Jira: the challenge is structural. Data Center stores draw.io diagrams as versioned binary packages that include the full revision history of each diagram. Jira Cloud’s API cannot read that format, so a format conversion step is required before migration. This conversion is one-way with no undo, which raises the stakes for preparation.
draw.io Data Center to Cloud Migration in Confluence: Step by Step
The official documentation outlines two approaches depending on your environment.
Option 1: The Original Page IDs Method (Recommended)
In some migration scenarios, Atlassian’s tooling preserves the original page IDs from your DC instance. When that happens, internal links in diagrams resolve correctly in Cloud without any additional draw.io steps. This is the preferred approach, particularly when migrating from older DC or Server versions.
Option 2: The draw.io Migration Tool
If page ID preservation isn’t possible, draw.io provides a dedicated migration tool. Only use this after confirming with draw.io support that it’s the right fit for your setup.
- Stop the Synchrony service to pause collaborative editing
- Go to draw.io Configuration, open the Cloud Migration tab, and click Start
- Download the migration report and copy the
linkAdjustmentsconfiguration command - Run the standard Atlassian Confluence migration
- Verify space permissions in Cloud: admins need write access, users need read access
- Paste the
linkAdjustmentsconfiguration code into draw.io settings in your Cloud instance - Run re-indexing to validate all diagram links
Version requirements: Confluence DC 7.x with draw.io 11.3.8 or later, or DC 8.x with draw.io 12.2.0 or later. Administrator access to both instances is required.
Need help with your Confluence migration? Contact our team about how we can assist with draw.io and the full Atlassian migration.
draw.io Data Center to Cloud Migration in Jira: Step by Step
The Jira migration guide covers the conversion process, but one point comes first: this migration is irreversible. Always test on a full clone of your production environment before running it on live data.
- In your Jira DC instance, go to Settings > Apps > draw.io Configuration and open the Server to Cloud migration tab
- Click Migrate to convert all diagrams to the Cloud-compatible format
- Download and save the migration report for troubleshooting reference
- Complete the standard Atlassian Jira migration to Cloud
- In your Jira Cloud instance, go to Settings > Apps > draw.io Configuration and click Start Indexing
On timing: run the Jira draw.io migration immediately before your cutover, not days in advance. Set your on-premises instance to read-only once the migration runs. Any new diagrams created in DC after the conversion won’t carry over in the Cloud-compatible format.
Before You Start: What to Get Right
A few things that consistently catch teams off guard during draw.io migration:
- Check version compatibility first: confirm draw.io meets minimum version requirements before anything else
- Test on a production clone: non-negotiable for Jira, strongly recommended for Confluence
- Factor in the migration window: the Jira format conversion adds time to your cutover process
- Save both migration reports: the Confluence and Jira tools each generate reports that are invaluable for post-migration troubleshooting
- Coordinate timing on Jira: the draw.io migration tool should run as close to cutover as possible, not days before
If you’re also consolidating multiple Atlassian instances as part of this migration, read our consolidation guide. draw.io’s migration steps apply per instance, so that context is directly relevant.
How Seibert Solutions Can Help
draw.io is one piece of a much larger migration. For IT leaders planning the move from Atlassian Data Center to Cloud, the app layer is often where migrations get complicated, and where the most time and cost can be saved with the right preparation.
Seibert Solutions is part of Seibert Group, the company behind draw.io. That means when draw.io questions go beyond what the documentation covers, our team has a direct line to the engineers who built it. Edge cases, unusual configurations, version-specific behavior: we can get answers other partners can’t.
But we’re here for the full migration, not just draw.io. As part of our cloud migration engagement, we run an app analysis across your entire Atlassian environment. That means reviewing every app currently running in Data Center and evaluating what you’re actually using. It’s common for organizations to carry licenses for apps they no longer actively use, overlapping tools that serve the same purpose, or apps with no Cloud equivalent that need a replacement strategy.
That analysis typically surfaces meaningful cost savings: licenses to drop, tools to consolidate, and Cloud pricing tiers that are a better fit for your actual usage pattern. For teams that have been on Data Center for several years, right-sizing the app layer often offsets a significant portion of the migration investment.
Whether you’re at the planning stage or mid-migration, we can step in and help.