cancel
Showing results for 
Search instead for 
Did you mean: 
Data Engineering
Join discussions on data engineering best practices, architectures, and optimization strategies within the Databricks Community. Exchange insights and solutions with fellow data engineers.
cancel
Showing results for 
Search instead for 
Did you mean: 

UC_COMMAND_NOT_SUPPORTED.WITHOUT_RECOMMENDATION in shared access mode

Awoke101
New Contributor II

I'm using a shared access cluster and am getting this error while trying to upload to Qdrant. This is the error. Anyway I can make it worked on shared access mode? It works on the personal cluster.

[UC_COMMAND_NOT_SUPPORTED.WITHOUT_RECOMMENDATION] The command(s): AppendData are not supported in Unity Catalog.  SQLSTATE: 0AKUC
File <command-2604385794497207>, line 65
     48 #embeddings_df = embeddings_df.limit(5)
     50 options = {
     51     "qdrant_url": QDRANT_GRPC_URL,
     52     "api_key": QDRANT_API_KEY,
   (...)
     60     "batch_size":"128",
     61 }
     63 embeddings_df.write.format("io.qdrant.spark.Qdrant").options(**options).mode(
     64     "append"
---> 65 ).save()

File /databricks/spark/python/pyspark/sql/connect/readwriter.py:670, in DataFrameWriter.save(self, path, format, mode, partitionBy, **options)
    668     self.format(format)
    669 self._write.path = path
--> 670 self._spark.client.execute_command(
    671     self._write.command(self._spark.client), self._write.observations
    672 )
File /databricks/spark/python/pyspark/sql/connect/client/core.py:1203, in SparkConnectClient.execute_command(self, command, observations, extra_request_metadata)
   1201     req.user_context.user_id = self._user_id
   1202 req.plan.command.CopyFrom(command)
-> 1203 data, _, _, _, properties = self._execute_and_fetch(
   1204     req, observations or {}, extra_request_metadata
   1205 )
   1206 if data is not None:
   1207     return (data.to_pandas(), properties)
File /databricks/spark/python/pyspark/sql/connect/client/core.py:1624, in SparkConnectClient._execute_and_fetch(self, req, observations, extra_request_metadata, self_destruct)
   1621 schema: Optional[StructType] = None
   1622 properties: Dict[str, Any] = {}
-> 1624 for response in self._execute_and_fetch_as_iterator(
   1625     req, observations, extra_request_metadata or []
   1626 ):
   1627     if isinstance(response, StructType):
   1628         schema = response
File /databricks/spark/python/pyspark/sql/connect/client/core.py:1601, in SparkConnectClient._execute_and_fetch_as_iterator(self, req, observations, extra_request_metadata)
   1599                     yield from handle_response(b)
   1600 except Exception as error:
-> 1601     self._handle_error(error)
File /databricks/spark/python/pyspark/sql/connect/client/core.py:1910, in SparkConnectClient._handle_error(self, error)
   1908 self.thread_local.inside_error_handling = True
   1909 if isinstance(error, grpc.RpcError):
-> 1910     self._handle_rpc_error(error)
   1911 elif isinstance(error, ValueError):
   1912     if "Cannot invoke RPC" in str(error) and "closed" in str(error):
File /databricks/spark/python/pyspark/sql/connect/client/core.py:1985, in SparkConnectClient._handle_rpc_error(self, rpc_error)
   1982             info = error_details_pb2.ErrorInfo()
   1983             d.Unpack(info)
-> 1985             raise convert_exception(
   1986                 info,
   1987                 status.message,
   1988                 self._fetch_enriched_error(info),
   1989                 self._display_server_stack_trace(),
   1990             ) from None
   1992     raise SparkConnectGrpcException(status.message) from None

 This is the code.

#embeddings_df = embeddings_df.limit(5)

options = {
    "qdrant_url": QDRANT_GRPC_URL,
    "api_key": QDRANT_API_KEY,
    "collection_name": QDRANT_COLLECTION_NAME,
    "vector_fields": "dense_vector",
    "vector_names": "dense",
    "schema": embeddings_df.schema.json(),
    "batch_size":"128",
}

embeddings_df.write.format("io.qdrant.spark.Qdrant").options(**options).mode(
    "append"
).save()

 

1 REPLY 1

Kaniz_Fatma
Community Manager
Community Manager

Hi @Awoke101

  • Shared access clusters in Databricks have certain restrictions due to Unity Catalog limitations.
  • I recommend trying the single-user Unity Catalog cluster or configuring data persistence outside the container. Let me know if you need further assistance! 
Join 100K+ Data Experts: Register Now & Grow with Us!

Excited to expand your horizons with us? Click here to Register and begin your journey to success!

Already a member? Login and join your local regional user group! If there isn’t one near you, fill out this form and we’ll create one for you to join!