Issue with the daemon grpc server

[Igosuki]

Hi, we deploy dagster on ECS and use PostgresRunStorage. We are trying to get tracing up for the daemon with dd-trace, but it logs a lot of error traces like :
/api.DagsterApi/Ping -> StatusCode.UNAVAILABLE: failed to connect to all addresses; last error: UNKNOWN: unix:/tmp/tmpa9xdjdq7: No such file or directory
Which end up saturating the process.

I think this originates in this part of the code :

def wait_for_grpc_server(server_process, client, subprocess_args, timeout=60):
    start_time = time.time()

    last_error = None

    while True:
        try:
            client.ping("")
            return
        except DagsterUserCodeUnreachableError:
            last_error = serializable_error_info_from_exc_info(sys.exc_info())

        if timeout > 0 and (time.time() - start_time > timeout):
            raise Exception(
                f"Timed out waiting for gRPC server to start after {timeout}s with arguments:"
                f" \"{' '.join(subprocess_args)}\". Most recent connection error: {last_error}"
            )

        if server_process.poll() is not None:
            raise Exception(
                f"gRPC server exited with return code {server_process.returncode} while starting up"
                f" with the command: \"{' '.join(subprocess_args)}\""
            )

        sleep(0.1)

Is this a dd-trace problem, or a dagster problem ? We have not done any particular configuration for the daemon's grpc server.