How to best preserve existing Confluence, when integrating with new Jira, AD users managed by crowd

[Steinar Bang]

I am trying to set up crowd 2.12.2 so that it manages users both for a brand new Jira 7.4, and an existing Confluence 5.10.2 installation.

The Confluence installation have up until now managed its own users.

I would like to preserve as much of the existing Confluence strucure as possible, while at the same time integrating the Confluence installation with Jira.

Is there a good way to map existing Confluence users to AD users?

Failing that: is there a good way to transfer ownership of pages and spaces from the Confluence users to the equivalent AD users?

Thanks!

[AnnWorley]

An application link that integrates JIRA and Confluence will not change your Confluence structure, so we are good there. It is possible to use JIRA for user management for Confluence, but you mention you will be using Crowd.

The AD directory will be created in Crowd, then, if the AD users have the same usernames, Confluence will "see" them as the same users, so the permissions in Confluence will be fine. If they have a different username in AD, you can use aliasing within Crowd to continue logging into Confluence with the same usernames: Specifying a User's Aliases

If you don't need aliasing or Single Sign-on you may want to connect JIRA and Confluence each directly to AD as in these docs:

JIRA - Connecting to an LDAP directory

Confluence - Connecting to an LDAP directory

I look forward to any follow up questions.

[Steinar Bang]

Hi and thanks for your answer

The Confluence installation had 41 active users with usernames matching their AD usernames, and 101 active users with usernames not matching their AD usernames.

Based on Effective memberships with multiple directories we made an initial plan to:

1. Manage Jira with a delegated directory, with locally managed groups, using lazy synchronization of users, with initial membership in the group jira-users
2. Rename the usernames of the 101 users that have a different username than their AD user, to get a username that matches their AD user
3. Import the Confluence users and groups into a crowd directory
4. Set up Confluence to be managed by both the same delegated directory as Jira, as well as the imported Confluence directory, aggregating users (and maybe groups)

Hopefully this will give us a Confluence and Jira working nicely together, while at the same time preserving the history and ownership of the existing Confluence pages...?

Initially we will start by managing just Jira with crowd using the delegated directory, and once that's up and running we will decide where we go from there.

[Steinar Bang]

An additional bit of information: Single SignOn is a requirement of whatever solution we end up with.