How to solve the error "Index chunk variable is required" when using the QA knowledge base pipeline

[xipingL]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

Content

• version: Self-Host 1.9.2
• config:

FILES_URL=http://localhost:5001
INTERNAL_FILES_URL=http://127.0.0.1:5001

• content:

When I used the knowledge pipeline and selected Q&A Processor to generate the Simple Q&A knowledge base, I encountered an error message "Index chunk variable is required".
Q&A Processor version: 0.0.13

• log:

2026-06-15 09:38:05,887 WARNING [log.py:232]
[on_tool_execution]

2026-06-15 09:38:05,887 WARNING [log.py:232]  Tool: qa_chunk

2026-06-15 09:38:05,887 WARNING [log.py:232]  Outputs: {"type":"text","message":{"text":"Get CSV file failed: [Errno 111] Connection refused"},"meta":null}

2026-06-15 09:38:05,991 ERROR [node.py:162]  Node knowledgeBase failed to run
Traceback (most recent call last):
  File "E:\dev\dify\api\core\workflow\nodes\base\node.py", line 144, in run
    result = self._run()
             ^^^^^^^^^^^
  File "E:\dev\dify\api\core\workflow\nodes\knowledge_index\knowledge_index_node.py", line 84, in _run
    raise KnowledgeIndexNodeError("Index chunk variable is required.")
core.workflow.nodes.knowledge_index.exc.KnowledgeIndexNodeError: Index chunk variable is required.
2026-06-15 09:38:06,14 ERROR [error_handler.py:101]  Node knowledgeBase failed with ABORT strategy: Index chunk variable is required.
2026-06-15 09:38:06,50 INFO [priority_rag_pipeline_run_task.py:63]  tenant_id: 74bb0622-debf-44f7-b0d3-ef35acec6427 , Rag pipeline run completed. Latency: 0.39469200000166893s
2026-06-15 09:38:06,62 INFO [trace.py:128]  Task tasks.rag_pipeline.priority_rag_pipeline_run_task.priority_rag_pipeline_run_task[b266a418-e6e8-45bf-949f-9ad4e6a49c89] succeeded in 0.4059999999590218s: None
2026-06-15 09:38:06,103 ERROR [workflow_entry.py:111]  Unknown Error when workflow entry running
Traceback (most recent call last):
  File "E:\dev\dify\api\core\workflow\workflow_entry.py", line 107, in run
    yield from generator
  File "E:\dev\dify\api\core\workflow\graph_engine\graph_engine.py", line 268, in run
    raise self._graph_execution.error
RuntimeError: Index chunk variable is required.

Thanks!

[dosubot[bot]]

The "Index chunk variable is required" error is a downstream symptom of the real problem: the qa_chunk tool is failing to fetch the CSV file because of a network connectivity issue.

Your logs confirm this — notice the line before the error:

Tool: qa_chunk
Outputs: {"type":"text","message":{"text":"Get CSV file failed: [Errno 111] Connection refused"}}

Because qa_chunk fails to produce any output, the downstream knowledgeBase node has no chunk variable to read, which triggers the "Index chunk variable is required" error.

Root cause: Your INTERNAL_FILES_URL is set to http://127.0.0.1:5001. In a Docker environment, 127.0.0.1 refers to the container's own loopback — not to other containers. When the plugin daemon tries to fetch the file at that address, it's connecting to itself, not to the API/file service container [1] [2].

Fix: Update INTERNAL_FILES_URL to use the Docker Compose service name instead of localhost:

INTERNAL_FILES_URL=http://api:5001

(where api is the service name of your Dify API container in docker-compose.yaml)

If you're running Dify without Docker (bare-metal/local development on Windows based on your path E:\dev\dify\api\...), then http://127.0.0.1:5001 should work for the API itself — but the plugin daemon may still be running in a separate process or container that can't reach that address. Make sure the plugin daemon process can reach port 5001 on the same machine, and that the file service is actually running and listening on that port when the pipeline executes.

To reply, just mention @dosu.

---

Share context across your team and agents. Try Dosu.