langchain-ai/langgraph
1
0
Fork 0
mirror of https://github.com/langchain-ai/langgraph.git synced 2026-07-15 12:35:43 -04:00
Code Issues 314 Packages Projects Releases 17 Wiki Activity

When a graph containing async checkpointer is incorrectly used with invoke or get_state, the program will hang #241

New Issue
Closed
opened 2026-02-20 17:33:34 -05:00 by yindo · 3 comments
yindo commented 2026-02-20 17:33:34 -05:00
Owner
Copy Link
Copy Source

Originally created by @gbaian10 on GitHub (Sep 23, 2024).

Checked other resources

  • I added a very descriptive title to this issue.
  • I searched the LangGraph/LangChain documentation with the integrated search.
  • I used the GitHub search to find a similar question and didn't find it.
  • I am sure that this is a bug in LangGraph/LangChain rather than my code.
  • I am sure this is better as an issue rather than a GitHub discussion, since this is a LangGraph bug and not a design question.

Example Code

import asyncio

from langgraph.checkpoint.sqlite.aio import AsyncSqliteSaver
from langgraph.graph import MessagesState, StateGraph

thread_config = {"configurable": {"thread_id": "1"}}


async def main() -> None:
    async with AsyncSqliteSaver.from_conn_string(":memory:") as checkpointer:
        workflow = StateGraph(MessagesState)
        app = workflow.compile(checkpointer=checkpointer)

        app.invoke({"messages": []}, thread_config)  # it will hang here

        # await app.ainvoke({"messages": []}, thread_config)
        # app.get_state(thread_config)  # it also hangs here


if __name__ == "__main__":
    asyncio.run(main())

Error Message and Stack Trace (if applicable)

In langgraph-checkpoint-sqlite == 1.0.1, the program will throw an exception, which is what I expected.

  File ".../python3.12/site-packages/langgraph/pregel/__init__.py", line 601, in get_state
    saved = checkpointer.get_tuple(config)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File ".../python3.12/site-packages/langgraph/checkpoint/sqlite/aio.py", line 37, in wrapper
    raise NotImplementedError(
NotImplementedError: The AsyncSqliteSaver does not support synchronous methods. Consider using the SqliteSaver instead.
from langgraph.checkpoint.sqlite import SqliteSaver
See https://langchain-ai.github.io/langgraph/reference/checkpoints/langgraph.checkpoint.sqlite.SqliteSaver for more information.

But in versions of langgraph-checkpoint-sqlite >= 1.0.2, the program will hang without reporting any errors.

The same issue will also occur with langgraph-checkpoint-postgres.

Description

The program can't continue running, but it doesn't report an error. If you don't realize it's because you're not using the async method, it might take you a lot of time to debug.
In my case, I was originally using MemorySaver(), and it worked fine, but when I switched to the asynchronous checkpointer, it suddenly stopped responding.

The issue is likely because the async checkpointer implements get_tuple.
If this situation is unavoidable, I hope at least a warning can be printed to alert users to switch methods.

System Info

langchain-core==0.3.5
langgraph==0.2.23
langgraph-checkpoint-sqlite==1.0.3
langgraph-checkpoint-postgres==1.0.7

platform==linux
python-version==3.12.6

Originally created by @gbaian10 on GitHub (Sep 23, 2024). ## Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the [LangGraph](https://langchain-ai.github.io/langgraph/)/LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangGraph/LangChain rather than my code. - [X] I am sure this is better as an issue [rather than a GitHub discussion](https://github.com/langchain-ai/langgraph/discussions/new/choose), since this is a LangGraph bug and not a design question. ## Example Code ```python import asyncio from langgraph.checkpoint.sqlite.aio import AsyncSqliteSaver from langgraph.graph import MessagesState, StateGraph thread_config = {"configurable": {"thread_id": "1"}} async def main() -> None: async with AsyncSqliteSaver.from_conn_string(":memory:") as checkpointer: workflow = StateGraph(MessagesState) app = workflow.compile(checkpointer=checkpointer) app.invoke({"messages": []}, thread_config) # it will hang here # await app.ainvoke({"messages": []}, thread_config) # app.get_state(thread_config) # it also hangs here if __name__ == "__main__": asyncio.run(main()) ``` ## Error Message and Stack Trace (if applicable) In langgraph-checkpoint-sqlite == 1.0.1, the program will throw an exception, which is what I expected. ```shell File ".../python3.12/site-packages/langgraph/pregel/__init__.py", line 601, in get_state saved = checkpointer.get_tuple(config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File ".../python3.12/site-packages/langgraph/checkpoint/sqlite/aio.py", line 37, in wrapper raise NotImplementedError( NotImplementedError: The AsyncSqliteSaver does not support synchronous methods. Consider using the SqliteSaver instead. from langgraph.checkpoint.sqlite import SqliteSaver See https://langchain-ai.github.io/langgraph/reference/checkpoints/langgraph.checkpoint.sqlite.SqliteSaver for more information. ``` But in versions of langgraph-checkpoint-sqlite >= 1.0.2, the program will hang without reporting any errors. The same issue will also occur with langgraph-checkpoint-postgres. ## Description The program can't continue running, but it doesn't report an error. If you don't realize it's because you're not using the async method, it might take you a lot of time to debug. In my case, I was originally using `MemorySaver()`, and it worked fine, but when I switched to the asynchronous checkpointer, it suddenly stopped responding. The issue is likely because the async checkpointer implements get_tuple. If this situation is unavoidable, I hope at least a warning can be printed to alert users to switch methods. ## System Info langchain-core==0.3.5 langgraph==0.2.23 langgraph-checkpoint-sqlite==1.0.3 langgraph-checkpoint-postgres==1.0.7 platform==linux python-version==3.12.6
yindo closed this issue 2026-02-20 17:33:35 -05:00
yindo commented 2026-02-20 17:33:35 -05:00
Author
Owner
Copy Link
Copy Source

@nfcampos commented on GitHub (Sep 23, 2024):

Thanks, opened PR to raise exception for this case

@nfcampos commented on GitHub (Sep 23, 2024): Thanks, opened PR to raise exception for this case
yindo commented 2026-02-20 17:33:35 -05:00
Author
Owner
Copy Link
Copy Source

@puyuanOT commented on GitHub (Sep 23, 2024):

I got the same error. What would be the correct way to get the state in this case?

@puyuanOT commented on GitHub (Sep 23, 2024): I got the same error. What would be the correct way to get the state in this case?
yindo commented 2026-02-20 17:33:35 -05:00
Author
Owner
Copy Link
Copy Source

@gbaian10 commented on GitHub (Sep 24, 2024):

I got the same error. What would be the correct way to get the state in this case?

@puyuanOT Use any async function
ex: ainvoke(), aget_state(), aget_tuple()....

@gbaian10 commented on GitHub (Sep 24, 2024): > I got the same error. What would be the correct way to get the state in this case? @puyuanOT Use any async function ex: ainvoke(), aget_state(), aget_tuple()....
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
agents
awaiting author revision
bug
change
chat models
checkpointer
ci
code quality
code quality
dependencies
documentation
drawing
duplicate
enhancement
good first issue
help wanted
invalid
maintainer
meta
missing information
not maintained by langchain
open-swe-auto
open-swe-max
open-swe-max-auto
pending
performance
pull-request
Mirrored from GitHub Pull Request
 question
requires design
sdk
server
stale
studio
tests
type checking
usage question
v0.5
v1.1
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/langgraph#241
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?