BerriAI/litellm · error · HTTPException

f"Maximum {MAX_BATCH_SIZE} users can be updated at once. Fou

Error message

f"Maximum {MAX_BATCH_SIZE} users can be updated at once. Found {len(all_users_in_db)} users."

What it means

Error "f"Maximum {MAX_BATCH_SIZE} users can be updated at once. Found {len(all_users_in_db)} users."" thrown in BerriAI/litellm.

Source

Thrown at litellm/proxy/management_endpoints/internal_user_endpoints.py:1772

        # Only proxy admins can update all users at once
        if user_api_key_dict.user_role != LitellmUserRoles.PROXY_ADMIN.value:
            raise HTTPException(
                status_code=403,
                detail="Only proxy admins can update all users at once.",
            )
        # Optimized path for updating all users directly in database
        all_users_in_db: Final = await _user_table(prisma_client).find_many(order={"created_at": "desc"})

        if not all_users_in_db:
            raise HTTPException(
                status_code=400,
                detail={"error": "No users found to update"},
            )

        # Limit batch size to prevent overwhelming the system
        MAX_BATCH_SIZE = 500  # Increased limit for all-users operations
        if len(all_users_in_db) > MAX_BATCH_SIZE:
            raise HTTPException(
                status_code=400,
                detail={
                    "error": f"Maximum {MAX_BATCH_SIZE} users can be updated at once. Found {len(all_users_in_db)} users."
                },
            )

        # Apply update transformations (reuse existing logic)
        data_json: Final[dict] = data.user_updates.model_dump(exclude_unset=True)
        non_default_values: Final = _update_internal_user_params(data_json=data_json, data=data.user_updates)

        # Remove user identification fields since we're updating by user_id
        non_default_values.pop("user_id", None)
        non_default_values.pop("user_email", None)

        successful_updates = 0
        failed_updates: Final = 0
        results: Final[list[UserUpdateResult]] = []

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Split the update into batches of at most MAX_BATCH_SIZE users per request.
2. For very large installations, update users in groups filtered by team or organization instead of all at once.

When it happens

Trigger: Thrown at litellm/proxy/management_endpoints/internal_user_endpoints.py:1772 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/373975ac06b03e31. Report an issue: GitHub.