[Bug]: AzureCosmosDBNoSqlVectorSearch query error "One of the input values is invalid." #17384
Comments
nqtung
added
bug
|
Hey @nqtung! 👋 I'm here to assist you with any bugs, questions, or contributions you have. Let's squash those bugs together! The error message "CosmosHttpResponseError: (BadRequest) One of the input values is invalid" can be caused by several issues when querying Azure Cosmos DB using the LlamaIndex framework. Here are some common causes and solutions:
Make sure to validate and sanitize your data before performing operations, and ensure that all configurations match your Cosmos DB setup. To continue talking to Dosu, mention @dosu. Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other |
|
Pretty sure this is fixed, but you'll need to update to the latest version of llama index and the vector store integration. You are on a fairly old version |
|
I am using the latest version of llama_index_vector_stores_azurecosmosnosql-1.3.0, the issue still there. The root cause is building wrong CosmosNoSQL query, I already fixed this issue and tested in my project. You find the changed at https://github.com/run-llama/llama_index/pull/17385/files |
Bug Description
Got the error message when trying to query index from CosmosDBNoSQL by following the example at https://docs.llamaindex.ai/en/stable/examples/vector_stores/AzureCosmosDBNoSqlDemo/
The error message:
CosmosHttpResponseError: (BadRequest) One of the input values is invalid.
ActivityId: 3a2a7521-b72a-4723-bb4a-2fe8d49233c6, Windows/10.0.20348 cosmos-netstandard-sdk/3.18.0
Code: BadRequest
Message: One of the input values is invalid.\r
ActivityId: 3a2a7521-b72a-4723-bb4a-2fe8d49233c6, Windows/10.0.20348 cosmos-netstandard-sdk/3.18.0"
The package dependency
Version
0.10.68
Steps to Reproduce
query_engine = index.as_query_engine() response = query_engine.query("What did the author love working on?")Relevant Logs/Tracbacks
{ "name": "CosmosHttpResponseError", "message": "(BadRequest) One of the input values is invalid.\r ActivityId: 3a2a7521-b72a-4723-bb4a-2fe8d49233c6, Windows/10.0.20348 cosmos-netstandard-sdk/3.18.0 Code: BadRequest Message: One of the input values is invalid.\r ActivityId: 3a2a7521-b72a-4723-bb4a-2fe8d49233c6, Windows/10.0.20348 cosmos-netstandard-sdk/3.18.0", "stack": "--------------------------------------------------------------------------- CosmosHttpResponseError Traceback (most recent call last) Cell In[4], line 1 ----> 1 response = query_engine.query(\"What is a buffer?\") 2 print(textwrap.fill(str(response), 100)) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\instrumentation\\dispatcher.py:260, in Dispatcher.span.<locals>.wrapper(func, instance, args, kwargs) 252 self.span_enter( 253 id_=id_, 254 bound_args=bound_args, (...) 257 tags=tags, 258 ) 259 try: --> 260 result = func(*args, **kwargs) 261 except BaseException as e: 262 self.event(SpanDropEvent(span_id=id_, err_str=str(e))) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\base\\base_query_engine.py:52, in BaseQueryEngine.query(self, str_or_query_bundle) 50 if isinstance(str_or_query_bundle, str): 51 str_or_query_bundle = QueryBundle(str_or_query_bundle) ---> 52 query_result = self._query(str_or_query_bundle) 53 dispatcher.event( 54 QueryEndEvent(query=str_or_query_bundle, response=query_result) 55 ) 56 return query_result File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\instrumentation\\dispatcher.py:260, in Dispatcher.span.<locals>.wrapper(func, instance, args, kwargs) 252 self.span_enter( 253 id_=id_, 254 bound_args=bound_args, (...) 257 tags=tags, 258 ) 259 try: --> 260 result = func(*args, **kwargs) 261 except BaseException as e: 262 self.event(SpanDropEvent(span_id=id_, err_str=str(e))) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\query_engine\\retriever_query_engine.py:189, in RetrieverQueryEngine._query(self, query_bundle) 185 \"\"\"Answer a query.\"\"\" 186 with self.callback_manager.event( 187 CBEventType.QUERY, payload={EventPayload.QUERY_STR: query_bundle.query_str} 188 ) as query_event: --> 189 nodes = self.retrieve(query_bundle) 190 response = self._response_synthesizer.synthesize( 191 query=query_bundle, 192 nodes=nodes, 193 ) 194 query_event.on_end(payload={EventPayload.RESPONSE: response}) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\query_engine\\retriever_query_engine.py:144, in RetrieverQueryEngine.retrieve(self, query_bundle) 143 def retrieve(self, query_bundle: QueryBundle) -> List[NodeWithScore]: --> 144 nodes = self._retriever.retrieve(query_bundle) 145 return self._apply_node_postprocessors(nodes, query_bundle=query_bundle) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\instrumentation\\dispatcher.py:260, in Dispatcher.span.<locals>.wrapper(func, instance, args, kwargs) 252 self.span_enter( 253 id_=id_, 254 bound_args=bound_args, (...) 257 tags=tags, 258 ) 259 try: --> 260 result = func(*args, **kwargs) 261 except BaseException as e: 262 self.event(SpanDropEvent(span_id=id_, err_str=str(e))) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\base\\base_retriever.py:243, in BaseRetriever.retrieve(self, str_or_query_bundle) 238 with self.callback_manager.as_trace(\"query\"): 239 with self.callback_manager.event( 240 CBEventType.RETRIEVE, 241 payload={EventPayload.QUERY_STR: query_bundle.query_str}, 242 ) as retrieve_event: --> 243 nodes = self._retrieve(query_bundle) 244 nodes = self._handle_recursive_retrieval(query_bundle, nodes) 245 retrieve_event.on_end( 246 payload={EventPayload.NODES: nodes}, 247 ) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\instrumentation\\dispatcher.py:260, in Dispatcher.span.<locals>.wrapper(func, instance, args, kwargs) 252 self.span_enter( 253 id_=id_, 254 bound_args=bound_args, (...) 257 tags=tags, 258 ) 259 try: --> 260 result = func(*args, **kwargs) 261 except BaseException as e: 262 self.event(SpanDropEvent(span_id=id_, err_str=str(e))) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\indices\\vector_store\\retrievers\\retriever.py:101, in VectorIndexRetriever._retrieve(self, query_bundle) 95 if query_bundle.embedding is None and len(query_bundle.embedding_strs) > 0: 96 query_bundle.embedding = ( 97 self._embed_model.get_agg_embedding_from_queries( 98 query_bundle.embedding_strs 99 ) 100 ) --> 101 return self._get_nodes_with_embeddings(query_bundle) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\core\\indices\\vector_store\\retrievers\\retriever.py:177, in VectorIndexRetriever._get_nodes_with_embeddings(self, query_bundle_with_embeddings) 173 def _get_nodes_with_embeddings( 174 self, query_bundle_with_embeddings: QueryBundle 175 ) -> List[NodeWithScore]: 176 query = self._build_vector_store_query(query_bundle_with_embeddings) --> 177 query_result = self._vector_store.query(query, **self._kwargs) 178 return self._build_node_list_from_query_result(query_result) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\vector_stores\\azurecosmosnosql\\base.py:393, in AzureCosmosDBNoSqlVectorSearch.query(self, query, **kwargs) 384 def query(self, query: VectorStoreQuery, **kwargs: Any) -> VectorStoreQueryResult: 385 \"\"\"Query index for top k most similar nodes. 386 387 Args: (...) 391 A VectorStoreQueryResult containing the results of the query. 392 \"\"\" --> 393 return self._query(query, **kwargs) File ____\\llm2024\\.venv\\Lib\\site-packages\\llama_index\\vector_stores\\azurecosmosnosql\\base.py:363, in AzureCosmosDBNoSqlVectorSearch._query(self, query, **kwargs) 360 print(f\"_query - query: {query}\") 361 print(f\"_query - parameters: {parameters}\") --> 363 for item in self._container.query_items( 364 query=query, 365 parameters=parameters, 366 enable_cross_partition_query=True, 367 ): 368 node = metadata_dict_to_node(item[self._metadata_key]) 369 node.set_content(item[self._text_key]) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\core\\paging.py:123, in ItemPaged.__next__(self) 121 if self._page_iterator is None: 122 self._page_iterator = itertools.chain.from_iterable(self.by_page()) --> 123 return next(self._page_iterator) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\core\\paging.py:75, in PageIterator.__next__(self) 73 raise StopIteration(\"End of paging\") 74 try: ---> 75 self._response = self._get_next(self.continuation_token) 76 except AzureError as error: 77 if not error.continuation_token: File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_query_iterable.py:99, in QueryIterable._fetch_next(self, *args) 89 def _fetch_next(self, *args): # pylint: disable=unused-argument 90 \"\"\"Return a block of results with respecting retry policy. 91 92 This method only exists for backward compatibility reasons. (Because (...) 97 :rtype: list 98 \"\"\" ---> 99 block = self._ex_context.fetch_next_block() 100 if not block: 101 raise StopIteration File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\execution_dispatcher.py:131, in _ProxyQueryExecutionContext.fetch_next_block(self) 129 self._execution_context = self._create_pipelined_execution_context(query_execution_info) 130 else: --> 131 raise e 133 return self._execution_context.fetch_next_block() File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\execution_dispatcher.py:123, in _ProxyQueryExecutionContext.fetch_next_block(self) 119 # TODO: NEED to change this - make every query retrieve a query plan 120 # also, we can't have this logic being returned to so often - there should be no need for this 121 # need to split up query plan logic and actual query iterating logic 122 try: --> 123 return self._execution_context.fetch_next_block() 124 except CosmosHttpResponseError as e: 125 if _is_partitioned_execution_info(e) or _is_hybrid_search_query(self._query, e): File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\base_execution_context.py:78, in _QueryExecutionContextBase.fetch_next_block(self) 69 def fetch_next_block(self): 70 \"\"\"Returns a block of results with respecting retry policy. 71 72 This method only exists for backward compatibility reasons. (Because (...) 76 :rtype: list 77 \"\"\" ---> 78 self._ensure() 79 res = list(self._buffer) 80 self._buffer.clear() File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\base_execution_context.py:63, in _QueryExecutionContextBase._ensure(self) 60 return 62 if not self._buffer: ---> 63 results = self._fetch_next_block() 64 self._buffer.extend(results) 66 if not self._buffer: File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\base_execution_context.py:163, in _DefaultQueryExecutionContext._fetch_next_block(self) 161 def _fetch_next_block(self): # pylint: disable=inconsistent-return-statements 162 while super(_DefaultQueryExecutionContext, self)._has_more_pages() and not self._buffer: --> 163 return self._fetch_items_helper_with_retries(self._fetch_function) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\base_execution_context.py:135, in _QueryExecutionContextBase._fetch_items_helper_with_retries(self, fetch_function) 132 def callback(): 133 return self._fetch_items_helper_no_retries(fetch_function) --> 135 return _retry_utility.Execute(self._client, self._client._global_endpoint_manager, callback) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_retry_utility.py:99, in Execute(client, global_endpoint_manager, function, *args, **kwargs) 97 result = ExecuteFunction(function, global_endpoint_manager, *args, **kwargs) 98 else: ---> 99 result = ExecuteFunction(function, *args, **kwargs) 100 if not client.last_response_headers: 101 client.last_response_headers = {} File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_retry_utility.py:199, in ExecuteFunction(function, *args, **kwargs) 192 def ExecuteFunction(function, *args, **kwargs): 193 \"\"\"Stub method so that it can be used for mocking purposes as well. 194 :param Callable function: the function to execute. 195 :param list args: the explicit arguments for the function. 196 :returns: the result of executing the function with the passed in arguments 197 :rtype: tuple(dict, dict) 198 \"\"\" --> 199 return function(*args, **kwargs) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\base_execution_context.py:133, in _QueryExecutionContextBase._fetch_items_helper_with_retries.<locals>.callback() 132 def callback(): --> 133 return self._fetch_items_helper_no_retries(fetch_function) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_execution_context\\base_execution_context.py:122, in _QueryExecutionContextBase._fetch_items_helper_no_retries(self, fetch_function) 119 new_options[\"continuation\"] = self._continuation 121 response_headers = {} --> 122 (fetched_items, response_headers) = fetch_function(new_options) 124 continuation_key = http_constants.HttpHeaders.Continuation 125 self._continuation = response_headers.get(continuation_key) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_cosmos_client_connection.py:1090, in CosmosClientConnection.QueryItems.<locals>.fetch_fn(options) 1089 def fetch_fn(options: Mapping[str, Any]) -> Tuple[List[Dict[str, Any]], CaseInsensitiveDict]: -> 1090 return self.__QueryFeed( 1091 path, 1092 \"docs\", 1093 collection_id, 1094 lambda r: r[\"Documents\"], 1095 lambda _, b: b, 1096 query, 1097 options, 1098 response_hook=response_hook, 1099 **kwargs) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_cosmos_client_connection.py:3124, in CosmosClientConnection.__QueryFeed(self, path, resource_type, resource_id, result_fn, create_fn, query, options, partition_key_range_id, response_hook, is_query_plan, **kwargs) 3121 if results: 3122 return __GetBodiesFromQueryResult(results), last_response_headers -> 3124 result, last_response_headers = self.__Post(path, request_params, query, req_headers, **kwargs) 3125 if last_response_headers.get(http_constants.HttpHeaders.IndexUtilization) is not None: 3126 INDEX_METRICS_HEADER = http_constants.HttpHeaders.IndexUtilization File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_cosmos_client_connection.py:2837, in CosmosClientConnection.__Post(self, path, request_params, body, req_headers, **kwargs) 2827 \"\"\"Azure Cosmos 'POST' http request. 2828 2829 :param str path: the url to be used for the request. (...) 2834 :rtype: tuple of (dict, dict) 2835 \"\"\" 2836 request = self.pipeline_client.post(url=path, headers=req_headers) -> 2837 return synchronized_request.SynchronizedRequest( 2838 client=self, 2839 request_params=request_params, 2840 global_endpoint_manager=self._global_endpoint_manager, 2841 connection_policy=self.connection_policy, 2842 pipeline_client=self.pipeline_client, 2843 request=request, 2844 request_data=body, 2845 **kwargs 2846 ) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_synchronized_request.py:214, in SynchronizedRequest(client, request_params, global_endpoint_manager, connection_policy, pipeline_client, request, request_data, **kwargs) 211 request.headers[http_constants.HttpHeaders.ContentLength] = 0 213 # Pass _Request function with its parameters to retry_utility's Execute method that wraps the call with retries --> 214 return _retry_utility.Execute( 215 client, 216 global_endpoint_manager, 217 _Request, 218 request_params, 219 connection_policy, 220 pipeline_client, 221 request, 222 **kwargs 223 ) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_retry_utility.py:97, in Execute(client, global_endpoint_manager, function, *args, **kwargs) 95 try: 96 if args: ---> 97 result = ExecuteFunction(function, global_endpoint_manager, *args, **kwargs) 98 else: 99 result = ExecuteFunction(function, *args, **kwargs) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_retry_utility.py:199, in ExecuteFunction(function, *args, **kwargs) 192 def ExecuteFunction(function, *args, **kwargs): 193 \"\"\"Stub method so that it can be used for mocking purposes as well. 194 :param Callable function: the function to execute. 195 :param list args: the explicit arguments for the function. 196 :returns: the result of executing the function with the passed in arguments 197 :rtype: tuple(dict, dict) 198 \"\"\" --> 199 return function(*args, **kwargs) File ____\\llm2024\\.venv\\Lib\\site-packages\\azure\\cosmos\\_synchronized_request.py:154, in _Request(global_endpoint_manager, request_params, connection_policy, pipeline_client, request, **kwargs) 152 raise exceptions.CosmosAccessConditionFailedError(message=data, response=response) 153 if response.status_code >= 400: --> 154 raise exceptions.CosmosHttpResponseError(message=data, response=response) 156 result = None 157 if data: CosmosHttpResponseError: (BadRequest) One of the input values is invalid.\r ActivityId: 3a2a7521-b72a-4723-bb4a-2fe8d49233c6, Windows/10.0.20348 cosmos-netstandard-sdk/3.18.0 Code: BadRequest Message: One of the input values is invalid.\r ActivityId: 3a2a7521-b72a-4723-bb4a-2fe8d49233c6, Windows/10.0.20348 cosmos-netstandard-sdk/3.18.0" }The text was updated successfully, but these errors were encountered: