[Bug]: Upgrade to 0.2.3 tries to reingest default docs and gets stuck at onboarding; Then, cannot ingest knowledge after factory reset

[aimurphy]

OpenRAG Version

0.2.2 > 0.2.3

Deployment Method

Other

Operating System

unknown

Python Version

unknown

Affected Area

Ingestion (document processing, upload, Docling)

Bug Description

Submitted on behalf of a user:

User performed an update and container restart using the TUI.
After the update, the OpenRAG web UI got stuck at onboarding because it was trying to reingest the default documents that are already present in the OpenSearch index.

Steps to Reproduce

1. Install 0.2.2 and make sure the default docs are ingested.
2. Upgrade to 0.2.3 using the TUI.
3. Start and launch OpenRAG.
4. Proceed through onboarding.
5. Once you reach embedding provider setup, it fails when it tries to reingest the default documents.

Additional test, after the above: Do a factory reset.
Now you should be able to get through Onboarding, however, the ability to ingest knowledge is bugged, and inspecting the knowledge base shows that the default docs (in ~/.openrag/documents/) weren't reingested after the reset, even though the index was wiped out.

Expected Behavior

An upgrade should either delete and reingest the default docs, or it should skip if they already exist.

Actual Behavior

The user got stuck at onboarding because it couldn't reingest the default docs

Relevant Logs

Screenshots

will ask user to share the screenshot

Additional Context

No response

Checklist

• I have searched existing issues to ensure this bug hasn't been reported before.
• I have provided all the requested information.

[aimurphy]

This could also be addressed by changing the way the default docs are handled in #575

[GolexPrime]

Hi, here is the screenshot you asked for. I updated from (i believe) 0.2.2 to 0.2.3.

[GolexPrime]

Tried to delete ~/.openrag/documents/* but to no avail. Now it complains about a .DS_Store file (hate those) that it cannot read. I removed any such file from ~/.openrag but the one that causes the issue still remains. Can I configure OpenRAG to ignore those or can I just delete the file myself (not sure where it is located though 🤔)

[GolexPrime]

updated to 0.2.3.0.1 and did a factory reset. Now the init runs through, but i cannot add any knowledge (and the starting documents are missing, even though they exist in ~/.openrag/documents/)

Also interesting: When accessed through the api, openrag now produces a 500. it seems the backend receives an empty response from langflow

2026-01-28T12:39:29.345415Z /app/src/agent.py:267 User prompt received prompt=Test
INFO:     10.89.0.5:57836 - "GET /tasks HTTP/1.1" 200 OK
2026-01-28T12:39:30.903653Z /app/src/agent.py:308 Exception in non-streaming response error='NoneType' object is not iterable
Traceback (most recent call last):
  File "/app/src/agent.py", line 288, in async_response
    output_text = getattr(response, "output_text", None)
  File "/app/.venv/lib/python3.13/site-packages/openai/types/responses/response.py", line 247, in output_text
    for output in self.output:
                  ^^^^^^^^^^^
TypeError: 'NoneType' object is not iterable
INFO:     10.89.0.5:48030 - "POST /v1/chat HTTP/1.1" 500 Internal Server Error
ERROR:    Exception in ASGI application
Traceback (most recent call last):
  File "/app/.venv/lib/python3.13/site-packages/uvicorn/protocols/http/httptools_impl.py", line 409, in run_asgi
    result = await app(  # type: ignore[func-returns-value]
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        self.scope, self.receive, self.send
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/app/.venv/lib/python3.13/site-packages/uvicorn/middleware/proxy_headers.py", line 60, in _call_
    return await self.app(scope, receive, send)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/app/.venv/lib/python3.13/site-packages/starlette/applications.py", line 113, in _call_
    await self.middleware_stack(scope, receive, send)
  File "/app/.venv/lib/python3.13/site-packages/starlette/middleware/errors.py", line 186, in _call_
    raise exc
  File "/app/.venv/lib/python3.13/site-packages/starlette/middleware/errors.py", line 164, in _call_
    await self.app(scope, receive, _send)
  File "/app/.venv/lib/python3.13/site-packages/starlette/middleware/exceptions.py", line 63, in _call_
    await wrap_app_handling_exceptions(self.app, conn)(scope, receive, send)
  File "/app/.venv/lib/python3.13/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    raise exc
  File "/app/.venv/lib/python3.13/site-packages/starlette/_exception_handler.py", line 42, in wrapped_app
    await app(scope, receive, sender)
  File "/app/.venv/lib/python3.13/site-packages/starlette/routing.py", line 716, in _call_
    await self.middleware_stack(scope, receive, send)
  File "/app/.venv/lib/python3.13/site-packages/starlette/routing.py", line 736, in app
    await route.handle(scope, receive, send)
  File "/app/.venv/lib/python3.13/site-packages/starlette/routing.py", line 290, in handle
    await self.app(scope, receive, send)
  File "/app/.venv/lib/python3.13/site-packages/starlette/routing.py", line 78, in app
    await wrap_app_handling_exceptions(app, request)(scope, receive, send)
  File "/app/.venv/lib/python3.13/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    raise exc
  File "/app/.venv/lib/python3.13/site-packages/starlette/_exception_handler.py", line 42, in wrapped_app
    await app(scope, receive, sender)
  File "/app/.venv/lib/python3.13/site-packages/starlette/routing.py", line 75, in app
    response = await f(request)
               ^^^^^^^^^^^^^^^^
  File "/app/src/api_key_middleware.py", line 86, in wrapper
    return await handler(request)
           ^^^^^^^^^^^^^^^^^^^^^^
  File "/app/src/api/v1/chat.py", line 132, in chat_create_endpoint
    result = await chat_service.langflow_chat(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<6 lines>...
    )
    ^
  File "/app/src/services/chat_service.py", line 159, in langflow_chat
    response_text, response_id = await async_langflow_chat(
                                 ^^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<7 lines>...
    )
    ^
  File "/app/src/agent.py", line 572, in async_langflow_chat
    response_text, response_id, response_obj = await async_response(
                                               ^^^^^^^^^^^^^^^^^^^^^
    ...<6 lines>...
    )
    ^
  File "/app/src/agent.py", line 288, in async_response
    output_text = getattr(response, "output_text", None)
  File "/app/.venv/lib/python3.13/site-packages/openai/types/responses/response.py", line 247, in output_text
    for output in self.output:
                  ^^^^^^^^^^^
TypeError: 'NoneType' object is not iterable

[aimurphy]

I think there are several bugs here:

• Cannot complete onboarding after upgrade
• Reset doesn't reingest /documents even though it destroys the OpenSearch index
• Upgrade to 0.2.3 followed by factory reset breaks the knowledge page (cannot ingest knowledge)

[aimurphy]

@GolexPrime - There is a fix in the next release that @lucaseduoli believes will fix/prevent this issue. @lucaseduoli said he will connect with you to test this when that version is ready.

[GolexPrime]

Did a quick round with @lucaseduoli
Turns out my Mac was down to about 50GB of free space. Ingestion did not work with that amount. After deleting about 100 GB worth of trash it started working again. Thanks to @lucaseduoli for the quick help (a dedicated error for this issue would be sweet though).

[prasanthcaibmcom]

@aimurphy @lucaseduoli What is the next step?

[aimurphy]

@prasanthcaibmcom - I added troubleshooting to the docs for this, but maybe @lucaseduoli can see if we can add a better error message in the app itself?

[lucaseduoli]

Yes!
I will put this issue in the GA milestone.