list_query_history
Retrieve and filter SQL query execution records from Databricks. Use parameters like warehouse, user, time range, and result limit to monitor and analyze query performance.
Instructions
List SQL query history
Args: warehouse_id: (Optional) Filter by specific warehouse user_id: (Optional) Filter by specific user start_time: (Optional) Start time in local format "YYYY-MM-DD HH:MM:SS" end_time: (Optional) End time in local format "YYYY-MM-DD HH:MM:SS" limit: Number of results to return
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| warehouse_id | No | ||
| user_id | No | ||
| start_time | No | ||
| end_time | No | ||
| limit | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- tools/compute.py:10-80 (handler)Implementation of the list_query_history MCP tool handler.
@mcp.tool def list_query_history( ctx: Context, warehouse_id: str = None, user_id: str = None, start_time: str = None, end_time: str = None, limit: int = 20 ) -> List[Dict[str, Any]]: """ List SQL query history Args: warehouse_id: (Optional) Filter by specific warehouse user_id: (Optional) Filter by specific user start_time: (Optional) Start time in local format "YYYY-MM-DD HH:MM:SS" end_time: (Optional) End time in local format "YYYY-MM-DD HH:MM:SS" limit: Number of results to return """ from databricks.sdk.service.sql import QueryFilter, TimeRange w = get_workspace_client() # Build filter filter_kwargs = {} if warehouse_id: filter_kwargs["warehouse_ids"] = [warehouse_id] if user_id: filter_kwargs["user_ids"] = [int(user_id)] # Time range (local time to Unix timestamp ms) if start_time or end_time: time_range_kwargs = {} if start_time: dt = datetime.strptime(start_time, "%Y-%m-%d %H:%M:%S") time_range_kwargs["start_time_ms"] = int(dt.timestamp() * 1000) if end_time: dt = datetime.strptime(end_time, "%Y-%m-%d %H:%M:%S") time_range_kwargs["end_time_ms"] = int(dt.timestamp() * 1000) filter_kwargs["query_start_time_range"] = TimeRange(**time_range_kwargs) if filter_kwargs: filter_by = QueryFilter(**filter_kwargs) response = w.query_history.list(filter_by=filter_by, max_results=limit) else: response = w.query_history.list(max_results=limit) queries_list = response.res if response and response.res else [] results = [] for q in queries_list: q_dict = q.as_dict() start_ms = q_dict.get("query_start_time_ms") start_time_local = None if start_ms: dt_local = datetime.fromtimestamp(start_ms / 1000) start_time_local = dt_local.strftime("%Y-%m-%d %H:%M:%S") results.append({ "query_id": q_dict.get("query_id"), "query_text": q_dict.get("query_text"), "status": q_dict.get("status"), "statement_type": q_dict.get("statement_type"), "user_name": q_dict.get("user_name"), "duration": q_dict.get("duration"), "start_time_ms": start_ms, "start_time_local": start_time_local }) return results