Confluence
Confluence is a collaborative work-management tool designed for sharing, storing, and working on project planning, software development, and product management. Amazon Kendra supports both Confluence Server/Data Center and Confluence Cloud. You can use Amazon Kendra to index the following Confluence entities:
-
Spaces – Top-level designated areas for organizing related content. Each space serves as a container, capable of holding multiple pages, blogs, and attachments.
-
Pages – Individual documents within a space where users create and manage content. Pages can contain text, images, tables, and multimedia elements, and can have nested sub-pages. Each page is considered a single document.
-
Blogs – Content similar to pages, typically used for updates or announcements. Each blog post is considered as a single document.
-
Comments – Allows users to give feedback or engage in discussions on specific content within pages or blog posts.
-
Attachments – Files uploaded to pages or blog posts in Confluence, such as images, documents, or other file types.
By default, Amazon Kendra doesn't index Confluence archives and personal spaces. You can choose to index them when you create the data source. If you don't want Amazon Kendra to index a space, mark it private in Confluence.
You can connect Amazon Kendra to your Confluence data source using either the Amazon Kendra console
Amazon Kendra has two versions of the Confluence connector. The following features are supported.
Confluence connector V2.0 / TemplateConfiguration API
-
Field mappings
-
User access control
-
Inclusion/exclusion patterns
-
Full and incremental content syncs
-
Virtual private cloud (VPC)
Confluence connector V1.0 / ConfluenceConfiguration API (no longer supported)
-
Field mappings
-
User access control
-
Inclusion/exclusion filters
-
(Confluence Server only) Virtual private cloud (VPC)
Note
Confluence connector V1.0 / ConfluenceConfiguration API ended in 2023. We recommend migrating to or using Confluence connector V2.0 / TemplateConfiguration API.
For troubleshooting your Amazon Kendra Confluence data source connector, see Troubleshooting data sources.
ACLs in Confluence Connector
Connectors support crawling Access Control Lists (ACLs) and identifying information where applicable based on the data source. If you index documents without ACLs, all documents are considered public. Indexing documents with ACLs ensures data security.
The Amazon Kendra Confluence connector scans spaces to collect pages and blog posts along with their ACLs. If there is no restriction applied on a page or blog, the connector inherits permissions from its space. If specific user or group restriction is applied on a page, only those users will be able to access that page. If page is nested, the nested page inherits the permissions of parent page if no restrictions are applied. A similar permissions model applies to blogs; however, Confluence does not support nested blogs.
In addition, Amazon Kendra Confluence connector crawls user principal information (local user alias, local group and federated group identity configurations) and its permissions for each configured space.
Note
The Confluence Cloud connector does not support crawling macros, whiteboards, or databases.
The Amazon Kendra Confluence connector updates ACL changes each time it crawls your data source content. To ensure the correct users have access to the correct content, regularly re-sync your data source to capture any ACL updates.
