Bedrock checkpoint saver failing with pop pending_sends #546
Description
Checked other resources
- This is a bug, not a usage question. For questions, please use the LangChain Forum (https://forum.langchain.com/).
- I added a clear and detailed title that summarizes the issue.
- I read what a minimal reproducible example is (https://stackoverflow.com/help/minimal-reproducible-example).
- I included a self-contained, minimal example that demonstrates the issue INCLUDING all the relevant imports. The code run AS IS to reproduce the issue.
Example Code
chat_config = {"configurable": {"thread_id": get_or_create_bedrock_session(connection_id)}}
for msg, metadata in chat_graph.stream(messages_json, config=chat_config, stream_mode="messages", ):
# for msg, metadata in chat_graph.stream(messages_json, stream_mode="messages", ):
logger.debug("New message received", extra={"marker": "start"})
logger.debug("Message content", extra={"UGHH": str(msg)})
logger.debug("Message metadata", extra={"metadata": msg.response_metadata})
logger.debug("End of message processing", extra={"marker": "end"})
msg: AIMessageChunk = msg
if not msg.response_metadata.get("finish_reason", "NOT FINISHED") == "stop":
yield msg.content
else:
logger.debug("Message finished", extra={"finish_reason": msg.response_metadata.get("finish_reason")})
yield "------END OF RESPONSE------"
def get_or_create_bedrock_session(connection_id: str):
"""
Look for a bedrock session ID in the connections table for the given connection.
If none is found, create a new bedrock checkpointer session and update the item.
Returns the session ID.
"""
try:
response = connection_table.get_item(Key={"connectionId": connection_id})
if "Item" in response:
# Connection exists, check if it has a bedrock session ID
item = response["Item"]
if "bedrockSessionId" in item and item["bedrockSessionId"]:
logger.debug("Found existing bedrock session ID", extra={"session_id": item["bedrockSessionId"]})
return item["bedrockSessionId"]
logger.debug("No bedrock session ID found, creating a new one")
bedrock_session = checkpointer.session_client.create_session()
bedrock_session_id = bedrock_session.session_id
logger.debug("Created new bedrock session ID", extra={"session_id": bedrock_session_id})
response = connection_table.update_item(
Key={"connectionId": connection_id},
UpdateExpression="SET bedrockSessionId = :sessionId",
ExpressionAttributeValues={":sessionId": bedrock_session_id},
# Create the item if it doesn't exist
ReturnValues="UPDATED_NEW"
)
logger.debug("Updated connection table with bedrock session ID", extra=response)
return bedrock_session_id
except ClientError as e:
logger.exception(f"Error getting or creating bedrock session: {e}")
logger.error(f"Error details: {traceback.format_exc()}")
raise e
except Exception as e:
logger.exception(f"Unexpected error in get_or_create_bedrock_session: {e}")
raise eError Message and Stack Trace (if applicable)
..\.venv\Lib\site-packages\langgraph\pregel\__init__.py:2467: in stream
with SyncPregelLoop(
..\.venv\Lib\site-packages\langgraph\pregel\loop.py:1051: in __exit__
return self.stack.__exit__(exc_type, exc_value, traceback)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\contextlib.py:619: in __exit__
raise exc
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\contextlib.py:604: in __exit__
if cb(*exc_details):
^^^^^^^^^^^^^^^^
..\.venv\Lib\site-packages\langgraph\pregel\executor.py:118: in __exit__
task.result()
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\concurrent\futures\_base.py:449: in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\concurrent\futures\_base.py:401: in __get_result
raise self._exception
..\.venv\Lib\site-packages\langgraph\pregel\loop.py:952: in _checkpointer_put_after_previous
prev.result()
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\concurrent\futures\_base.py:449: in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\concurrent\futures\_base.py:401: in __get_result
raise self._exception
..\.venv\Lib\site-packages\langgraph\pregel\executor.py:81: in done
task.result()
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\concurrent\futures\_base.py:449: in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\concurrent\futures\_base.py:401: in __get_result
raise self._exception
D:\Tools\uv\pythons\cpython-3.13.5-windows-x86_64-none\Lib\concurrent\futures\thread.py:59: in run
result = self.fn(*self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
..\.venv\Lib\site-packages\langgraph\pregel\loop.py:954: in _checkpointer_put_after_previous
cast(BaseCheckpointSaver, self.checkpointer).put(
..\.venv\Lib\site-packages\langgraph_checkpoint_aws\saver.py:364: in put
session_checkpoint = create_session_checkpoint(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
checkpoint = {'channel_values': {'__start__': {'messages': [{'content': 'hello', 'role': 'user'}]}}, 'channel_versions': {'__start__': 1}, 'id': '1f06690f-388f-6732-bfff-8fdf16046c32', 'ts': '2025-07-22T00:15:19.564874+00:00', ...}
config = {'callbacks': None, 'configurable': {'checkpoint_ns': '', 'thread_id': 'ad9415f5-6b95-471b-a403-26ffe25f7e6e'}, 'metadata': ChainMap({'thread_id': 'ad9415f5-6b95-471b-a403-26ffe25f7e6e'}), 'recursion_limit': 25, ...}
metadata = {'parents': {}, 'source': 'input', 'step': -1}
serializer = <langgraph.checkpoint.serde.jsonplus.JsonPlusSerializer object at 0x0000025DC6A1EA50>
new_versions = {'__start__': 1}
def create_session_checkpoint(
checkpoint: Checkpoint,
config: RunnableConfig,
metadata: CheckpointMetadata,
serializer: SerializerProtocol,
new_versions: ChannelVersions,
) -> SessionCheckpoint:
"""
Create a SessionCheckpoint object from the given checkpoint and related data.
This function processes the checkpoint, extracts necessary information from the config,
and serializes various components to create a SessionCheckpoint object.
Args:
checkpoint (Checkpoint): The checkpoint to process.
config (RunnableConfig): Configuration containing thread and checkpoint information.
metadata (CheckpointMetadata): Metadata associated with the checkpoint.
serializer (SerializerProtocol): Serializer for data conversion.
new_versions (ChannelVersions): New versions of channel data.
Returns:
SessionCheckpoint: A SessionCheckpoint object containing the processed and serialized data.
"""
# Create copy to avoid modifying original checkpoint
checkpoint_copy = checkpoint.copy()
# Remove pending sends as they are handled separately
> checkpoint_copy.pop("pending_sends")
E KeyError: 'pending_sends'
..\.venv\Lib\site-packages\langgraph_checkpoint_aws\utils.py:265: KeyErrorDescription
This is happening in a server environment and I am able to reproduce it in my workstation. I will try to post complete code. I wanted to ask if there is something that everyone knows and I don't.
This is for Bedrock Session Saver.
System Info
System Information
OS: Windows
OS Version: 10.0.26100
Python Version: 3.13.5 (main, Jun 12 2025, 12:42:35) [MSC v.1943 64 bit (AMD64)]
Package Information
langchain_core: 0.3.69
langsmith: 0.4.6
langchain_aws: 0.2.28
langchain_openai: 0.3.28
langgraph_checkpoint_aws: 0.1.0
langgraph_sdk: 0.1.73
Optional packages not installed
langserve
Other Dependencies
boto3: 1.39.5
httpx: 0.28.1
httpx>=0.25.2: Installed. No version info available.
jsonpatch<2.0,>=1.33: Installed. No version info available.
langchain-core<1.0.0,>=0.3.68: Installed. No version info available.
langgraph: 0.5.3
langgraph-checkpoint: 2.1.0
langsmith-pyo3: Installed. No version info available.
langsmith>=0.3.45: Installed. No version info available.
numpy: 2.3.1
openai-agents: Installed. No version info available.
openai<2.0.0,>=1.86.0: Installed. No version info available.
opentelemetry-api: Installed. No version info available.
opentelemetry-exporter-otlp-proto-http: Installed. No version info available.
opentelemetry-sdk: Installed. No version info available.
orjson: 3.11.0
orjson>=3.10.1: Installed. No version info available.
packaging: 25.0
packaging>=23.2: Installed. No version info available.
pydantic: 2.11.7
pydantic>=2.7.4: Installed. No version info available.
pytest: 8.4.1
PyYAML>=5.3: Installed. No version info available.
requests: 2.32.4
requests-toolbelt: 1.0.0
rich: Installed. No version info available.
tenacity!=8.4.0,<10.0.0,>=8.1.0: Installed. No version info available.
tiktoken<1,>=0.7: Installed. No version info available.
typing-extensions>=4.7: Installed. No version info available.
zstandard: 0.23.0