What causes "retry_delay must be non-negative" in iflytek/astron-agent?

Clamp before calling: retry_delay = max(0.0, computed_delay)

iflytek/astron-agent · error · ValueError

retry_delay must be non-negative

Error message

retry_delay must be non-negative

What it means

get_document_chunks validates retry_delay and raises ValueError when it is negative. A negative inter-poll delay is nonsensical (it would mean sleeping backwards), so the function rejects it immediately alongside the max_retries check.

Solutions

Clamp before calling: retry_delay = max(0.0, computed_delay)
Fix the backoff/config computation yielding the negative delay
Validate timing config at startup and reject negatives
Use the documented default retry_delay of 3.0 seconds

Example fix

// before
chunks = await get_document_chunks(ds, doc_id, retry_delay=backoff)
// after
chunks = await get_document_chunks(ds, doc_id, retry_delay=max(0.0, backoff))

Defensive patterns

Strategy: validation

Validate before calling

def safe_retry_delay(v) -> float:
    d = float(v) if v is not None else 3.0
    if d < 0:
        raise ValueError("retry_delay must be >= 0")
    return d

chunks = await get_document_chunks(ds, doc_id, retry_delay=safe_retry_delay(cfg_delay))

Type guard

def is_valid_retry_delay(v) -> bool:
    return isinstance(v, (int, float)) and not isinstance(v, bool) and v >= 0

Try / catch

try:
    chunks = await get_document_chunks(ds, doc_id, retry_delay=d)
except ValueError as e:
    logger.error("bad retry_delay: %s", e)
    chunks = await get_document_chunks(ds, doc_id)  # default 3.0

Prevention

Clamp backoff computations with max(0.0, value)
Check sign of multipliers in backoff formulas
Validate timing-related config at load time
Prefer documented defaults over computed delays when unsure

When it happens

Trigger: Calling get_document_chunks(dataset_id, doc_id, retry_delay=-1.0) or any negative float, usually from miscomputed backoff arithmetic (e.g. delay = base * factor with a negative factor) or bad config.

Common situations: Backoff formulas with negative multipliers; config/env values entered as negative numbers; wiring an elapsed-time delta (which can be negative) into retry_delay.

Related errors

AI-assisted analysis of iflytek/astron-agent@5e758547a8 (2026-09-12). Data as JSON: /api/errors/202e51c75d569360. Report an issue: GitHub.

Appendix: source

Thrown at core/knowledge/infra/ragflow/ragflow_utils.py:367

        Get all chunks after parsing, retrying incomplete search snapshots.

        Each attempt delegates to the canonical fail-closed paginator. RAGFlow
        API errors and incomplete pagination therefore propagate instead of
        being misreported as a valid empty document.

        Args:
            dataset_id: Dataset ID
            doc_id: Document ID
            max_retries: Maximum incomplete-snapshot retries (default: 15)
            retry_delay: Delay between retries in seconds (default: 3.0)

        Returns:
            Complete chunk list, or an empty list after all empty retries.
        """
        if max_retries < 0:
            raise ValueError("max_retries must be non-negative")
        if retry_delay < 0:
            raise ValueError("retry_delay must be non-negative")

        doc_info = await get_document_info(dataset_id, doc_id)
        if doc_info is None:
            raise RuntimeError(
                f"RAGFlow document disappeared before chunk retrieval: doc={doc_id}"
            )

        expected_count = RagflowUtils._normalize_expected_chunk_count(
            doc_info.get("chunk_count")
        )

        last_visible_count = 0
        last_chunk_ids: Optional[tuple[str, ...]] = None
        stable_partial_reads = 0
        for attempt in range(max_retries + 1):
            chunks = await fetch_all_document_chunks(dataset_id, doc_id, page_size=100)
            last_visible_count = len(chunks)
            has_complete_snapshot = (

View on GitHub (pinned to 5e758547a8)