langgraph/tutorials/langgraph-platform/local-server/

[giscus[bot]]

langgraph/tutorials/langgraph-platform/local-server/

Build language agents as graphs

https://langchain-ai.github.io/langgraph/tutorials/langgraph-platform/local-server/

[wm987]

I tried setting it up using the agent template by following the instructions here, but when I tried running langgraph dev, I got this error:

$ langgraph dev          
INFO:langgraph_api.cli:

        Welcome to

╦  ┌─┐┌┐┌┌─┐╔═╗┬─┐┌─┐┌─┐┬ ┬
║  ├─┤││││ ┬║ ╦├┬┘├─┤├─┘├─┤
╩═╝┴ ┴┘└┘└─┘╚═╝┴└─┴ ┴┴  ┴ ┴

- 🚀 API: http://127.0.0.1:2024
- 🎨 Studio UI: https://smith.langchain.com/studio/?baseUrl=http://127.0.0.1:2024
- 📚 API Docs: http://127.0.0.1:2024/docs

This in-memory server is designed for development and testing.
For production use, please use LangGraph Cloud.


2024-11-25T06:33:11.602737Z [info     ] Will watch for changes in these directories: ['C:\\Users\\William\\Documents\\GitHub\\ArchiLabs-AI-MonoRepo\\langgraph'] [uvicorn.error] api_variant=test
2024-11-25T06:33:11.603738Z [info     ] Uvicorn running on http://127.0.0.1:2024 (Press CTRL+C to quit) [uvicorn.error] api_variant=test color_message=Uvicorn running on %s://%s:%d (Press CTRL+C to quit)
2024-11-25T06:33:11.603738Z [info     ] Started reloader process [120400] using WatchFiles [uvicorn.error] api_variant=test color_message=Started reloader process [120400] using WatchFiles
2024-11-25T06:33:12.579357Z [info     ] Started server process [110076] [uvicorn.error] api_variant=test color_message=Started server process [%d]
2024-11-25T06:33:12.579357Z [info     ] Waiting for application startup. [uvicorn.error] api_variant=test
2024-11-25T06:33:12.663118Z [error    ] Traceback (most recent call last):
  File "C:\Python311\Lib\site-packages\starlette\routing.py", line 693, in lifespan
    async with self.lifespan_context(app) as maybe_state:
  File "C:\Python311\Lib\contextlib.py", line 204, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Python311\Lib\site-packages\langgraph_api\lifespan.py", line 30, in lifespan
    await collect_graphs_from_env(True)
  File "C:\Python311\Lib\site-packages\langgraph_api\graph.py", line 230, in collect_graphs_from_env
    graph = await run_in_executor(None, _graph_from_spec, spec)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Python311\Lib\site-packages\langchain_core\runnables\config.py", line 588, in run_in_executor
    return await asyncio.get_running_loop().run_in_executor(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Python311\Lib\concurrent\futures\thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Python311\Lib\site-packages\langchain_core\runnables\config.py", line 579, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "C:\Python311\Lib\site-packages\langgraph_api\graph.py", line 271, in _graph_from_spec
    raise e
  File "C:\Python311\Lib\site-packages\langgraph_api\graph.py", line 268, in _graph_from_spec
    modspec.loader.exec_module(module)
  File "<frozen importlib._bootstrap_external>", line 940, in exec_module
  File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
  File "C:\Users\William\Documents\GitHub\ArchiLabs-AI-MonoRepo\langgraph\./src/agent/graph.py", line 11, in <module>
    from agent.configuration import Configuration
ModuleNotFoundError: No module named 'agent'
Could not import python module for graph: GraphSpec(id='agent', path='./src/agent/graph.py', module=None, variable='graph', config=None)
 [uvicorn.error] api_variant=test
2024-11-25T06:33:12.665766Z [error    ] Application startup failed. Exiting. [uvicorn.error] api_variant=test

Any idea why it's failing to find the module?

[hinthornw]

Did you install the current module?

pip install .

[ujitkumar1]

This issue might be caused by an incorrect Python path or because the required module has not been installed.

To set the Python path:
Run: export PYTHONPATH=. in the project's root directory.

To install the module:
Run: pip install ."

Also check if you are in the right path in the terminal

[viren-vii]

I am getting same error even after installing the dependencies. It says
ModuleNotFoundError: No module named 'react_agent'

[jakehemmerle]

I'm having a situation where langgraph dev just hangs. It was working earlier today. I added a .env tried wiping the venv, reinstalling CLI, trying new templates, no luck. Using uv with project-independant venvs.

~/codebases/infra/asdf main* ❯ langgraph dev                                                         asdf
INFO:langgraph_api.cli:

        Welcome to

╦  ┌─┐┌┐┌┌─┐╔═╗┬─┐┌─┐┌─┐┬ ┬
║  ├─┤││││ ┬║ ╦├┬┘├─┤├─┘├─┤
╩═╝┴ ┴┘└┘└─┘╚═╝┴└─┴ ┴┴  ┴ ┴

- 🚀 API: http://127.0.0.1:2024
- 🎨 Studio UI: https://smith.langchain.com/studio/?baseUrl=http://127.0.0.1:2024
- 📚 API Docs: http://127.0.0.1:2024/docs

This in-memory server is designed for development and testing.
For production use, please use LangGraph Cloud.

[prasanjeevi]

Faced the same issue downgrading from langgraph-checkpoint==2.0.7 to langgraph-checkpoint==2.0.6 works for me

[hinthornw]

Do you still see the locking if you use --no-browser?

[jwj1342]

When using langgraph dev, I encountered the following error:

Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "D:\ANACONDA\envs\LangSmith\Scripts\langgraph.exe\__main__.py", line 7, in <module>
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\click\core.py", line 1157, in __call__
    return self.main(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\click\core.py", line 1078, in main
    rv = self.invoke(ctx)
         ^^^^^^^^^^^^^^^^
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\click\core.py", line 1688, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\click\core.py", line 1434, in invoke
    return ctx.invoke(self.callback, **ctx.params)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\click\core.py", line 783, in invoke
    return __callback(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\langgraph_cli\analytics.py", line 96, in decorator
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\langgraph_cli\cli.py", line 608, in dev
    run_server(
  File "D:\ANACONDA\envs\LangSmith\Lib\site-packages\langgraph_api\cli.py", line 221, in run_server
    uvicorn.run(
TypeError: run() got an unexpected keyword argument 'auth'

This error occurs because the uvicorn.run() function does not accept an auth argument.

Solution:

To resolve this issue, modify the run_server function in the cli.py file by removing the **kwargs argument from the uvicorn.run() call. After making this change, the program should function correctly.

[hinthornw]

Hi @jwj1342 could you upgrade to 0.1.63? Apologies for the broken release!

[ATAKing1023]

Requirements:
LangGraph CLI: Requires langchain-cli[inmem] >= 0.1.58

but the latest version of langchain-cli in PyPI is 0.0.35.

[Kashyab19]

Hi, I followed all the steps mentioned in the documentation

Ran this step after creating an env file with contents:
langgraph dev

My Langgraph server is running but it does not show any of the threads in the file. Error:

Failed to load assistants
Please verify if the API server is running or accessible from the browser.
TypeError: Failed to fetch

Am I missing something?

[aileen5150]

I encountered a similar issue. Using the F12 developer tools, I observed that an HTTP request was made to my local service at http://{ip}:{port}/assistants/search, but a cross-origin resource sharing (CORS) error occurred. However, when testing the /assistants/search endpoint via http://{ip}:{port}/docs, the call succeeded. The only deviation from the official tutorial is that, instead of using localhost, I am using the server's IP address.
I resolved the issue by implementing a local proxy.

[martinobettucci]

Mixed Content: The page at 'https://smith.langchain.com/studio/thread?baseUrl=http%3A%2F%2F0.0.0.0%3A2024' was loaded over HTTPS, but requested an insecure resource 'http://0.0.0.0:2024/assistants/search'. This request has been blocked; the content must be served over HTTPS.

It is impossible to load a local resource from a an https resource which is also on another domain. This can't work

[eyurtsev]

Which web-browser are you using?

[zaferfatihk]

Guys come on. This doesn't work. You put a quickstart tutorial on one of the most hyped tools at the moment and you don't have time to fix it. Come on guys!!!

[eyurtsev]

Could you file an issue with an explanation of what doesn't work?

[xtbwqtq]

What hook functions can be used to perform cleanup after langgraph-cli is shut down?

[eyurtsev]

What are you trying to achieve?