Enabling Tracing in Phoenix for Asynchronous Langgraph Execution
Hello, everyone. I have a question about tracing in Phoenix. I鈥檓 using langgraph and currently executing my graph asynchronously using graph.astream(). I was wondering how I can enable or capture tracing in Phoenix while the graph is running in streaming mode. Any guidance would be appreciated.
async for chunk in agent_graph.astream(base_state, stream_mode="messages"):
if chunk[1]["langgraph_node"] == "node_response":
yield chunk[0].contentI get an error like below.
packages/langchain_core/runnables/config.py", line 616, in run_in_executor return await asyncio.get_running_loop().run_in_executor( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ...<2 lines>... ) ^ asyncio.exceptions.CancelledError: <object object at 0x1141d1860>
