Migrating from OnDemand to a Crowd installed site
You can extract your user data from a Confluence or Jira application installed site instance. Use these queries to export your users and memberships to CSV, and then import them into Crowd.
Migrate Jira Cloud or Confluence Cloud
First you'll need to migrate Jira/Confluence Cloud to your server versions:
Exporting user data
If you are using Postgres as your database with Jira or Confluence, you can generate the required CSV files from either of those systems, with the following queries. If you are using a different database, you may need to tweak these.
// Run on JIRA/Confluence copy (SELECT user_name AS "Username", first_name AS "First Name", last_name AS "Last Name", email_address AS "Email Address", credential AS "Password" FROM cwd_user) to '/tmp/users.csv' csv header;
// Run on JIRA/Confluence copy (SELECT DISTINCT child_name AS "Username", parent_name AS "Groupname" FROM cwd_membership) to '/tmp/group_memberships.csv' csv header;
Retrieving the backup data
The files will be generated in /tmp, or in the folder specified in the query above.
What data is backed up
The backup includes the following data:
- User accounts
Importing the data into Crowd
- Follow the instructions in Importing Users from CSV Files
- Connect Crowd to your applications: Adding an Application
- Ensure that the Crowd Directory is placed above the Internal Directory in your User Directories in all downstream applications: Configuring User Directories. This is important because you will have the same users in both directories, and the users will authenticate against the highest directory in the hierarchy.