MemPalace/mempalace · error · RuntimeError

Drawer write was acknowledged but the new ID is not readable

Error message

Drawer write was acknowledged but the new ID is not readable. The palace index may be stale; run reconnect or repair.

What it means

Error "Drawer write was acknowledged but the new ID is not readable. The palace index may be stale; run reconnect or repair." thrown in MemPalace/mempalace.

Source

Thrown at mempalace/mcp_server.py:3002

        idempotency_probe_ids = [drawer_id, f"{drawer_id}_chunk_{last_chunk_idx:06d}"]
    try:
        existing = col.get(ids=idempotency_probe_ids, include=[])
        if _get_result_ids(existing):
            return {"success": True, "reason": "already_exists", "drawer_id": drawer_id}
    except Exception as e:
        logger.warning("Idempotency pre-check failed for %s", idempotency_probe_ids, exc_info=True)
        return {"success": False, "error": f"Idempotency check failed before write: {e}"}

    try:
        if len(content) <= chunk_size:
            col.upsert(
                ids=[drawer_id],
                documents=[content],
                metadatas=[{**base_meta, "chunk_index": 0}],
            )
            inserted = col.get(ids=[drawer_id], include=[])
            if not _get_result_ids(inserted):
                raise RuntimeError(
                    "Drawer write was acknowledged but the new ID is not readable. "
                    "The palace index may be stale; run reconnect or repair."
                )
            _metadata_cache = None
            logger.info(f"Filed drawer: {drawer_id} -> {wing}/{room}")
            return {
                "success": True,
                "drawer_id": drawer_id,
                "wing": wing,
                "room": room,
                "chunks": 1,
            }

        # Oversized content: split into bounded per-chunk drawers so the
        # embedding model never sees a document above ``chunk_size``.
        # Single batched ``upsert`` so the embedding pass either commits
        # every chunk or none — no half-written palace if the embedding
        # model fails mid-loop (#1539).

View on GitHub (pinned to 06cb6987f0)

Solutions

1. Run the reconnect or repair tool to rebuild the stale index

When it happens

Trigger: Thrown at mempalace/mcp_server.py:3002 when the library encounters an invalid state.

Common situations: Write succeeded but immediate read-back failed due to a stale palace index.

AI-assisted analysis of MemPalace/mempalace@06cb6987f0 (2026-08-15). Data as JSON: /api/errors/18397d12bcad80df. Report an issue: GitHub.