iflytek/astron-agent · error · CustomException

GetFileContentFailed

GetFileContentFailed

Error message

Xinghuo knowledge base failed to get document chunk content data

What it means

get_chunks() polls for up to 70 iterations (4s apart, ~4.7 minutes) waiting for the chunks endpoint to return data. If after all retries `data` is still empty/None, it raises CustomException with CodeEnum.GetFileContentFailed. It indicates the Xinghuo service never produced retrievable chunk content within the polling window.

Solutions

1. Check get_file_status(): if still splitting, wait and call get_chunks() again later.
2. Re-trigger the split for the document; if it completes faster the next time, this was a transient stall.
3. For very large files, increase max_retries or the sleep interval, or split the source document into smaller files.
4. Verify the file_id is correct — polling a wrong/nonexistent id yields empty responses until the budget runs out.

Example fix

# before
chunks = await get_chunks(file_id=file_id)
# after
try:
    chunks = await get_chunks(file_id=file_id)
except CustomException:
    await asyncio.sleep(60)
    chunks = await get_chunks(file_id=file_id)  # retry once after long processing

status = await get_file_status(file_id=file_id)
if not status:
    raise ValueError(f"no status found for file {file_id}; verify the id")

def chunks_ready(response: object) -> bool:
    return isinstance(response, list) and len(response) > 0

try:
    chunks = await get_chunks(file_id=file_id)
except CustomException as e:
    if "chunk content" in str(e):
        await asyncio.sleep(120)  # long job may still be processing
        chunks = await get_chunks(file_id=file_id)
    else:
        raise

When it happens

Trigger: Polling a file whose status stays "spliting"/"ocring" beyond 70 retries (very large document, stuck job), or a file whose chunks endpoint keeps returning an empty response despite a non-failed status.

Common situations: Very large PDFs whose OCR exceeds ~5 minutes; jobs silently stuck server-side; file_id pointing to a file with no chunks (e.g. empty document); slow Xinghuo service under load so the polling budget expires.

Understand the failure class

Background: Request timed out: what client-side request timeouts mean across libraries (Request timed out, TIMED_OUT, APITimeoutError) — this error's family across 39 libraries.

Related errors

• Document splitting failed after retries
• Document splitting failed
• {desc from XINGHUO-RAG response}
• Failed to 【XINGHUO-RAG】; code
• RAGFLOW_RAGError

AI-assisted analysis of iflytek/astron-agent@5e758547a8 (2026-09-12). Data as JSON: /api/errors/bc3f97db80a0ee01. Report an issue: GitHub.

Appendix: source

Thrown at core/knowledge/infra/xinghuo/xinghuo.py:209

        if response:
            # Response could be a dict or list, handle both cases
            if isinstance(response, list):
                data = response
            else:
                # If response is a dict, wrap it in a list to match expected type
                data = [response] if response else []
            break

        logger.info(
            f"File: {file_id} - Retry {retry_count + 1}, document chunk content not obtained, continuing to retry..."
        )
        retry_count += 1
        if retry_count < max_retries:
            await asyncio.sleep(4)

    if not data:
        raise CustomException(
            CodeEnum.GetFileContentFailed,
            "Xinghuo knowledge base failed to get document chunk content data",
        )

    # Ensure data is properly typed as List[Dict[str, Any]]
    return data if isinstance(data, list) else []


async def new_topk_search(
    query: str,
    doc_ids: Optional[List[str]] = None,
    top_n: Optional[int] = 5,
    **kwargs: Any,
) -> Dict[str, Any]:
    """
    Use new retrieval interface for hybrid search.

    Args:

View on GitHub (pinned to 5e758547a8)