DC to Cloud migration

This page explains how to migrate your Assets and Inventory Plugin (AIP) data from Jira Data Center to Jira Cloud.

On this page:

AIP currently does not provide a JCMA (Jira Cloud Migration Assistant) path. Instead, you can migrate by:

Before you start

Before starting the migration, review the DC to Cloud feature comparison page to understand what is and isn’t supported in Jira Cloud. This helps you:
- confirm which asset attributes can be migrated as-is,
- identify unsupported attribute types early, and
- adjust your Cloud configuration or use cases before exporting data.
Admin permissions for AIP in both Jira Data Center and Jira Cloud.
Access to your Data Center asset repository.
A Cloud environment where you can create asset structure (collections, types, attributes).

Asset type collections, asset types, and attributes are not created automatically during import.
You must create them in Jira Cloud before importing assets.
Not all Data Center attribute types are supported in Jira Cloud. Check the feature comparison page and:
- create the closest supported attribute type in Cloud, or
- adjust your use case before exporting data.
System fields (global attributes) such as Location are not supported in Cloud.
- Use a workaround (for example, a Select list attribute).
- This may require changes both in Data Center and in the exported CSV.
Asset links (references between assets) are not migrated automatically via CSV import.
Asset custom fields and their values from Jira work items are not migrated as part of this process.
Plan your scope:
- If you want to migrate everything, make sure your Asset Navigator search includes all assets.
- If you only want a subset, filter first and export only what you need.
Currently, you can import up to 1,000 assets per CSV file in Jira Cloud. If you have more assets:
- split your export into multiple files, or
- filter your Asset Navigator search accordingly.

If you prefer a visual walkthrough, watch the screen recording before you start.

Log in to your Jira Cloud instance and create the same structure and required fields you have in Data Center:

Asset type collections
Asset types (make sure they are in the correct collections)
Attributes (make sure each attribute is created under the correct asset type)
- Keep attribute names and types consistent with Data Center.
- If your attributes include references, statuses, or users, make sure the Cloud-side configurations are ready so the import can match values correctly.

Before exporting data, double-check that:

Missing or mismatched attributes will cause the import to fail.

In Jira Data Center:

In the top navigation bar, go to Asset Navigator.
Refine your search to include the assets you want to migrate.
Open the Export menu and select Export for Cloud.
Select the delimiter, then click Download.
Save the generated file. This file is Cloud-compatible and is meant for On-Premise → Cloud import.

Important export considerations

After downloading the CSV, open it in a plain text editor (for example, TextEdit or Notepad) and check the separators used in the file:
- Field separator: separates columns (for example, , or ;)
- Decimal separator: used in number values (for example, . or ,)
- Multi-value separator: separates multiple values in the same field
These values must match the separators you configure during the import in Jira Cloud. Although the default field separator is usually a comma (,), some operating systems or spreadsheet tools save CSV files using a semicolon (;). If the separators don’t match, the import may fail or data may be mapped to the wrong fields.
If your assets contain reference attributes (links to other assets), you must decide how referenced assets are identified:
- By ID – matches assets using their unique internal ID
- By name – matches assets using the asset name
Use the same method consistently during export and import. If asset names are not unique in your Data Center instance, exporting and importing references by ID is strongly recommended. Otherwise, the import may link assets incorrectly or fail to resolve references.

For detailed instructions and explanations of import options and field mappings, you can refer to this page.

You now have a Cloud-compatible CSV export from Data Center.

In Jira Cloud:

Open AIP and go to the Import flow (CSV import tool).
Upload the CSV you exported from Data Center.
Make sure your import settings match your file format (for example):
- Character Set: UTF-8
- Separators: field / decimal / multi-value (match what you used when exporting)
- Date/time format: must match Jira Cloud expectations

Reference, user, and group attributes

Referenced assets must already exist in Cloud at import time.
You cannot set a reference to an asset that hasn’t been imported yet.
Configure the Reference Asset Fetch Method during import to match your export (ID or name).
If your attributes use Jira users or groups:
- those users and groups must already exist in Jira Cloud,
- migrate your Jira environment first, then migrate AIP data.

Continue through the import steps (Upload → Review → Import).
After the import finishes, review the summary to confirm:
- how many assets were imported successfully
- which assets failed and why

For detailed instructions and explanations of import options and field mappings, you can refer to this page.

Your assets are now available in AIP for Jira Cloud.

Keep in mind that the following are not migrated automatically:

This usually means the asset type attributes were not created (or don’t match) in Cloud.

Go back to Step 1 and confirm:
- asset type collection exists
- asset type exists under the correct collection
- all required attributes exist and match the CSV columns

This can happen because of special characters or whitespace in group names.

Spreadsheet editors (especially Excel) can change date formats depending on OS settings.

Confirm your CSV uses date/time formats accepted by Jira Cloud.
Avoid re-saving the exported CSV through tools that auto-convert dates unless needed.