search

langchain-ai / langchain-nvidia

MIT License
48 stars 15 forks source link

Logs not getting generated with langchain-nvidia #32

Closed tapansstardog closed 4 months ago

tapansstardog commented 4 months ago

Hi team,

I am running 0.7.1triton server with TensorRT model.

I0429 12:46:50.778379 1 grpc_server.cc:2519] Started GRPCInferenceService at 0.0.0.0:8001
I0429 12:46:50.778524 1 http_server.cc:4623] Started HTTPService at 0.0.0.0:8000
I0429 12:46:50.819315 1 http_server.cc:315] Started Metrics Service at 0.0.0.0:8002

I have configured logging:

{
    "log_file": "/logs/logs291.txt",
    "log_info": true,
    "log_warning": true,
    "log_error": true,
    "log_verbose_level": 1,
    "log_format": "default"
}

Logs are getting generated properly when I am making the HTTP call, however that is not the case for GRPC (Langchain-nvidia-trt). Anything I need to do differently? Please suggest.

Second concern:

In case of multiple threads, I am sometimes getting the error below. Is there any configuration specific for GRPC?

[StatusCode.UNAVAILABLE] failed to connect to all addresses; last error: UNKNOWN: ipv4:127.0.0.1:8081: Failed to connect to remote host: Connection refused

Regards Tapan

tapansstardog commented 4 months ago

Found the issue in my code.