Plan your server to cloud migration
Documents to help you prepare to migrate your Atlassian server products.
Use Jira Site Import to migrate from server to cloud
For a test migration or UAT, we recommend that your test cloud site is not part of the organization that also hosts your prod site. The prod site should be hosted in a different organization. This is to ensure smooth migration of the relevant users and groups.
This guide shows you how to use Jira's native site backup and restore functionality (Site Import) to move from Jira Server to Jira Cloud. The process described in this article is recommended for migrating Jira Service Management (formerly Jira Service Desk) or a combination of Jira products including Jira Service Management.
If you're planning to migrate only Jira Software and/or Jira Core, we recommend using the Jira Cloud Migration Assistant, which you can use to migrate individual projects or whole sites from server to cloud.
Before using Jira Site Import, evaluate which migration method is best for you.
Migration planning
We recommend reviewing our guide to planning your migration before completing the steps on this page. Then come back and start at step 1. Or read on if you've already read the planning guide.
Step 1: Complete the pre-migration checklist
Before attempting a test or production migration, complete all of the steps for Jira Site Import in the pre-migration checklist. The checklist will help you prepare yourself and your data for migration, and ensure you avoid common sources of migration failure.
Step 2: Determine your user migration strategy
The best way to import your users and groups will depend on a few variables: what products you’re migrating, how your users and groups are currently managed in server, and if you plan to use Atlassian Access for SCIM (user provisioning). Take time to review the possible Jira user migration strategies and determine which you'll need to use.
Step 3: Back up your Jira Server database
This step will back up your Jira Server database in a portable XML format. If you have multiple Jira applications on the same site, for example Jira Software, Jira Core and Jira Service Management (formerly Jira Service Desk), this backup will contain all project types (Software, Business and Service) by default.
Log in as a user with the Jira System Administrators permission.
Choose > System.
Select Import & Export > Backup System to open the Jira backup page.
In the File name field, type a name for your backup file.
Click the Backup button and wait while your Jira data is backed up. Jira will save your backup as a zipped XML archive file within the export subdirectory of the Jira application home directory (jira-home for short). Check that Jira has the necessary file system permissions to write to this location.
When the backup is complete, you'll see a confirmation message. You can now retrieve the file from your server.
Step 4: Back up your Jira Server attachments, project avatars, and logos
If you want to migrate attachments, project avatars, or logos, you'll need to back them up as well. These are stored in your Jira application’s data directory, which is a sub-directory of the Jira application home directory.
To complete this step, you'll need to create a ZIP archive of the /data/attachments, /data/avatars/, and /logos/ directories. Make sure it uses the following structure:
1
2
3
4
5
6
7
8
9
10
11
12
attachments-1.zip
└─── data
├── attachments
│ ├── ProjectKey1
│ │ └── 10000
│ │ ├── IssueKey-1
│ │ ├── IssueKey-2
│ │ └── IssueKey-3
│ ├── ProjectKey2
│ └── ProjectKey3
└── avatars
└── logosThere’s no one specific way to back up the data directory, but Backing up data outlines a few methods to consider. An example command that can be run from a *nix server: 1
2
cd <Jira Home Dir>
zip -r data.zip logos/ data/avatars/ data/attachments/Once the ZIP file is created, you’ll need to transfer the file from your server to the local workstation you’re using. Some common methods of doing this include using SFTP, rsync, wget, and curl. |
Step 5: Back up your Jira cloud site
Importing your server backup to your Jira cloud site will overwrite all data in your cloud site. This data cannot be recovered. So, before importing your server backup into your Jira cloud site, you must back up your Jira cloud site. Make sure you download and keep a copy of the backup file. To back up your cloud site:
Log in to your Jira Cloud site with site-admin permissions.
Click Jira Settings > System.
In the Import and Export section, click Backup manager.
Under the Back up for cloud section, click Create backup for cloud. Make sure you include attachments, avatars, and logos in the backup by selecting the checkbox.
Step 6: Import your server database backup into Jira Cloud
(If not already) log in to your new Jira Cloud site with site-admin permissions.
Click Jira Settings > System.
In the Import and Export section, click Restore system.
Select Import data and select the ZIP file you downloaded in Step 2: Back up your Jira Server database.
We'll check your file for errors, and ask you to choose settings for how you want your data imported.
First, select your outgoing mail setting. Whichever option you choose, you can manually change after you migrate in the outgoing mail settings.
You can choose to Enable outgoing mail, which allows Jira to send automated emails for interactions – for example, new comments on issues or issue transitions.
If you choose to Disable outgoing mail, Jira won't send automated emails for interactions.
Next, select whether you want to overwrite or merge users from your server site with the users on your cloud site.
Choosing to overwrite will completely replace all users from Confluence and Jira with the users in your backup file.
Note: If you don’t see this option, you have the improved user management experience. If you need access to the Overwrite existing cloud users option or have any concerns, contact support.Choosing to merge users will also merge groups. This means that if we find a group on your server instance with the same name as a group on your cloud site (in either Jira or Confluence), we will merge the users from the server group into the cloud group. The server group users will receive the permissions of the cloud group. Make sure you check for possible permission escalation before choosing this option.
If you've migrated users using the Jira Cloud Migration Assistant, and you want to import all other project data, select I've migrated user data using the Jira Cloud Migration Assistant.
Click Run import. This will take you to a page where you can track your import progress. It may take a while for your import to finish, depending on the size of your backup.
When the import is complete, you'll be taken to a confirmation screen.
Turn off the incoming mail handlers in cloud during testing.
By default, the incoming mail handlers are turned on after an import. This can cause mail handlers in cloud to process emails instead of mail handlers in server. We strongly recommend turning these off during testing to prevent mails from incorrect processing.
You can do the following:
Click Jira Settings > System.
Select Global Mail Settings, and then turn off Email puller and Email processor.
Select Incoming Mail, and then delete all mail servers and handlers.
If you have Jira Service Management, go to Settings > Products > Email requests, and make sure the Email addresses section is empty.
From here, if you still need to import your media (attachments, logos, and project avatars) choose Import media. You'll need to return to the Import and Export page before managing application access. If you've already completed this step, or aren't planning to import your media, you can skip to step 6 by clicking the link to Grant application access.
Troubleshooting this step
Before you import, unzip the backup file and make sure it has a file structure similar to: 1
2
3
JIRA-backup-20161021
├── activeobjects.xml
├── entities.xmlIf you see this error: Import error There was an error importing file JIRA-backup-20161021.zip: Validation failed. The following issues were reported: The import archive doesn't contain entities.xml file. Then you may have zipped a folder containing the .xml files, instead of zipping the XML files directly. |
You may have to clean up invalid characters in your backup before importing into Jira Cloud. Take a moment to look over the documentation and integrate it into your migration plan. |
During the import process, you may see it stall on the following percentages/steps:
This is going to be normal and you just need to wait for them to complete. |
You may find you need to audit usernames or email addresses to check for errors before you can complete the import to cloud. The primary things to review are:
If we find invalid or duplicate emails, we’ll provide you an option to download a CSV file containing these emails. You’ll have to either: Enter the correct email address in the NewEmail column. Or If you don’t want to migrate the user, you’ll need to change FALSE to TRUE in the tombstone column. You’ll then need to upload the corrected file using the Upload CSV option and run the import again. |
At this step, you may see a list of URLs for Atlassian server sites. To make sure links point to the new Jira Cloud site instead of the old server site after you migrate:
|
Site import does not support migrating users managed by external directories. We recommend using the Jira Cloud Migration Assistant to migrate your data. If you can't use the Jira Cloud Migration Assistant to migrate all data, consider using it to migrate users only. For example, you can migrate all your Jira users with the migration assistant first and then import your projects through Jira Site import later. |
Step 7: Import your Jira Server media into Jira Cloud
(If not already) log in to your new Jira Cloud site with site-admin permissions.
Click Jira Settings > System.
In the Import and Export section, click Restore system.
Choose Import media and then find and select the media ZIP you created and downloaded in Step 3: Back up your Jira Server attachments, project avatars, and logos.
This will take you to a page where you can track your import progress. It may take a while for your import to finish, depending on the size of your file.
When the import is complete, you'll be taken to a confirmation screen.
From here, if you still need to import more media, repeat this step until all media files are imported. Note that you may want to test that all media was imported successfully in a new tab before importing the next batch of media.
Troubleshooting this step
Large attachment files may not import properly and can take a long time to complete. If your file is larger than 10 GB, you may need to split up the backup into 2-5 GB chunks instead and then import each one separately. The file structure should be: 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
attachments-1.zip
└─── data
├── attachments
│ ├── ProjectKey1
│ │ └── 10000
│ │ ├── IssueKey-1
│ │ ├── IssueKey-2
│ │ └── IssueKey-3
│ ├── ProjectKey2
│ └── ProjectKey3
└── avatars
└── logos
attachments-2.zip
└─── data
└── attachments
├── ProjectKey4
├── ProjectKey5
└── ProjectKey6
attachments-3.zip
└─── data
└── attachments
└── ProjectKey#See How to import only part of a Jira export into Jira Cloud for more instructions. |
During the attachment import step, there are a few distinct phases, including uploading, processing, and the actual import. Large attachment files may be interrupted during the processing phase. You can identify this if the page reloads suddenly and there's no further progress or message shown. If everything is okay, after the processing step you should see a message like: Attention This will import non-database content (such as attachments and avatars) into your Jira instance. You should only do this after a full database import because the full import will remove any existing data that you import using this page. This operation could take anywhere from several minutes to several hours depending on the size and number of attachments. Note that this operation cannot be undone. This indicates the processing is completed. From here, click the Proceed button to finish importing your attachments. If you notice that the import failed during the processing step, the solution is to divide the attachments into smaller ZIP files and import each file separately. |
Step 8: Review and grant access to groups imported from server
Once you've imported everything into your cloud site, you'll need to review the groups imported from server and decide which to grant access to.
As a security measure, the import process won't automatically apply the default application access settings or grant access to new users. You'll need to grant the application access to these users before they can log in.
Log in to your new Jira Cloud site and click Jira Settings > User management.
In the Site settings section, choose Product access.
Choose Review imported groups to review and confirm the default groups' access.
Note that during the import, the name of the default Jira Server group Jira-administrators changes to administrators in Jira Cloud.
Step 9: Re-add yourself to groups imported from server
As a security measure, the group settings for the person performing the migration are not imported. This means the user who executed the migration will need to be added back to their previous groups. You can do this two ways, either from the Groups section or the Users section at https://admin.atlassian.com.
From the Users section
If you have multiple cloud sites, you'll need to select the cloud site you've just migrated to.
In the User management section, first choose Users.
Find your profile.
Click the ... icon
Choose Add user to group.
Select the group(s) you need to be added to.
Click Add to Groups.
From the Groups section
Go to https://admin.atlassian.com.
If you have multiple cloud sites, you'll need to select the cloud site you've just migrated to.
In the User management section, first choose Groups.
Then, select the group you need to be added to.
Choose Add members, and add yourself to the group.
Repeat as needed for the rest of your groups.
Step 10: Complete post-migration checks
Depending on the type of migration, there may be some things you need to do once your migration is finished. For a full list of post-migration recommendations, refer to the migration planning guide.
Troubleshooting this step
There's a known bug affecting admins on the cloud site. After importing your data to cloud, the content associated with your account in server may be associated to "Admin". If you run into this bug, just contact our support team for help resolving the issue. |
Additional troubleshooting
If you've run into a problem during your migration, we're here to help. You can start by searching for known issues in our public issue tracker. There, you can find information about common issues we see with Jira migrations, including their status and suggested workarounds.
Some known issues include the following:
Unsupported Crowd error: See Importing from Local Jira to Jira Cloud throws Unsupported Crowd Directory Error.
500 errors in Jira: See Opening of Jira Issue Navigator Fails with Exception 500. Once data is fixed in Jira Server, you'll need to back it up again, then import it to your cloud site following the steps on this page.
More information and support
We have a number of channels available to help you with your migration.
For more migration planning information and FAQs, visit the Atlassian Cloud Migration Center.
Have a technical issue or need more support with strategy and best practices? Get in touch.
Looking for peer advice? Ask the Atlassian Community.
Want expert guidance? Work with an Atlassian Partner.
Last modified on Sep 6, 2021
Cached at 11:29 AM on Sep 20, 2021 |
Was this helpful?