Data Source Connection Feature: How to Use It? Error When Adding Confluence/Notion and No Files Visible

[wikj163]

Hi everyone,
I’m trying to use the new Data Source feature in RAGFlow (v0.22.0+), but I’m running into issues when adding Confluence or Notion as a data source.
Here’s what I did and what went wrong:

I successfully connected Confluence using my API Token and Base URL (connection shows “Connected” in the UI).
I also tried Notion with the Internal Integration Token.
After connecting, I clicked “Sync Now” and waited for a few minutes, but:
No files/pages from Confluence/Notion appear in the data source list or in any Dataset.
When I create a new Dataset and link the data source, the files are still not visible.

When I try to add or configure Confluence/Notion data source, the page sometimes throws an error (TypeError or page crash), and the form doesn’t respond properly.

My questions:

Is the Data Source feature fully usable in the current version? Are there any special requirements or known bugs?
What are the exact steps to make Confluence or Notion sync work (e.g., do I need to fill Space Key/Database ID, specific permissions, etc.)?
Has anyone successfully synced files from Confluence/Notion? Could you share your configuration or screenshots?

Environment:

RAGFlow version: v0.21.1-15-g56def59c (custom build with some cherry-picks to v0.22.0)
Deployment: Source code + local development (npm run dev + launch_backend_service.sh)
Browser: Latest Chrome

Any help, documentation links, or troubleshooting tips would be greatly appreciated! Thank you.
Best regards,

[dosubot[bot]]

The Data Source feature for Confluence and Notion is usable in v0.22.0+, but there are some important requirements and known issues to watch for:

Configuration & Sync Steps:
After connecting a data source, you must link it to a dataset and manually start synchronization from the dataset page—this step is essential and not automatic. For Confluence, required fields are Name, Username, Access Token, Wiki Base URL (no trailing slash), and Cloud toggle. Space Key, Page ID, and Index Recursively are optional; Space Key is only required if you select "index_mode: space" in the UI. For Notion, you need the Internal Integration Token (as a Bearer token) and optionally a root_page_id to limit indexing. The Notion integration must be shared with every page/database you want to index—if not, those items won't appear in RAGFlow. Credentials must have read access to all content you want to sync.
Confluence config details
Notion config details

Known Issues & Troubleshooting:

• "Connected" status in the UI only means credentials were saved, not that sync succeeded.

• If no files appear after sync, check:

  • Data source is linked to a dataset and sync was started
  • Credentials and base URL are correct
  • Permissions allow reading all spaces/pages (Confluence) or shared with integration (Notion)
  • Backend services (Redis, MySQL/PostgreSQL, Elasticsearch, MinIO) are running and healthy
  • Database schema includes the sync_logs table
  • For Docker deployments, UI sync may not work; run python rag/svr/sync_data_source.py inside the container as a workaround
  • Remove any trailing slash from the Confluence Wiki Base URL
  • For Confluence Server, set "Cloud" toggle to False
  • Parsing and indexing may take time; documents show "RUNNING" status until complete
    Troubleshooting checklist

• The "must be called from async context" error is likely due to a frontend async bug or incorrect usage of asyncio.run(main) instead of asyncio.run(main()) in the sync script. This was fixed in recent code updates—make sure your build includes this fix if you see this error.
  Async context error details

• Confluence connector only supports Basic Auth (username + token/password); Bearer token/SAML/SSO is not supported yet.
  Auth limitation

Recent Changes:
Recent PRs added UI fields for Space Key, Page ID, Index Recursively, and Index Method (everything, space, page). The UI now conditionally requires Space Key or Page ID based on the selected index mode.
PR: Space Key field
PR: More attributes
PR: Index Method UI

Successful Syncs:
There are user reports and screenshots confirming successful Confluence syncs when the above steps and requirements are met.
Example and screenshots

If you still see no files after following these steps, double-check permissions and backend health, and check the sync logs for error messages. Let me know if you want help interpreting specific error logs or configuration!

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

   

[Magicbook1108]

Upgrade to 22.0+, then follow what dosu said and try agin.