Issue when inserting in to vector db

When inserting PDF in to qdrant vector db using the llama index framework am getting the below error

**_InactiveRpcError: <_InactiveRpcError of RPC that terminated with:
status = StatusCode.INVALID_ARGUMENT
details = "cannot convert to json number"
debug_error_string = "UNKNOWN:Error received from peer , grpc_status:3, grpc_message:"cannot convert to json number"}"

**

code used

from llama_index import (
    VectorStoreIndex,
    SimpleDirectoryReader,
    StorageContext,ServiceContext
)
vector_store = QdrantVectorStore(client=client, collection_name=COLLECTION)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
# to change the default values
service_context = ServiceContext.from_defaults(embed_model=embed_model,\
                                               llm=llm,\
                                               chunk_size = 256,
                                               chunk_overlap = 50)

documents = SimpleDirectoryReader(
                                DIRECTORY_PATH, 
                                file_extractor={".pdf": UnstructuredReader()},
                                recursive=True, 
                                file_metadata=file_metadata,
                                filename_as_id=True,
                                required_exts=FILE_EXTENSION).load_data()

index = VectorStoreIndex.from_documents(
    documents=documents,
    storage_context=storage_context, 
    service_context=service_context,
    store_nodes_override=True

```)

Start by reproducing the provided Python insertion using LlamaIndex, QdrantVectorStore, a PDF, and the shown configuration. Check the Qdrant and client versions plus the inserted document metadata and payload values to isolate what produces the JSON-number conversion error. Done means identifying a reproducible cause and documenting or validating a focused fix.