Md. Azmal
Md. Azmal
using tacer.start_as_current_span does prevent creating multiple traces, but the span is empty
@dosu the solutions didnt work, i tried what you mentioned, any specific workaround or guides you could suggest? or anybody in the team who can help?
Upon debugging i realised when invoking the agent in sync (with just invoke()), the TraceID is same for all the invokations happening internally. In my case each of the graph...
@RogerHYang Thanks for responding, I haven't verified if a simple graph is having this issue, I'll let you know once it's verified. I tried enabling logs for langgraph and it...
@RogerHYang I checked with a very basic agent written by gpt and the issue still persists ``` from langchain_core.messages import AIMessage, HumanMessage from langchain_openai import ChatOpenAI from langgraph.checkpoint.postgres.aio import AsyncPostgresSaver...
Hi @RogerHYang, although passing the config to the nodes themselves is trivial as they only expect the state variable as argument (i am not sure if you can pass any...
@RogerHYang Oh i didn't knew that! thanks for mentioning.
Just as a note, is there a way to provide a traceId manually? it seems the sessionid and uid context is penetrating pretty well, if there's a way to have...
got it, let me know if there's any fix for the current issue
Hi, i tried this it seems to solve some parts of problem, basically when i am adding config any of the ainvoke it works fine for llm's but in my...