After starting a container with containers.run, the stdout from a C++ App seems to be empty.

Hello everyone,

I am running a docker container based on an ubuntu:bionic based image where I start my C++ testApp with an entrypoint.

Please find below my different codes.

C++ testApp Code Sample:

#include <iostream>
#include <unistd.h>

int main(int argc, char **argv)
{
	printf("\nServer is starting... !!\n\n");
	while (1)
            usleep(100000);
	return 1;
}

DockerFile content:

FROM amd64/ubuntu:latest
ADD / test
WORKDIR /test
RUN chmod +x testApp
ENTRYPOINT ["/test/testApp"]

Python Code

   import docker, time

    imageName = 'test'
    imagePath = 'C:/test/image'
    dockerClient = docker.from_env()

    try:
        dockerClient.images.remove(image = imageName, force = True)
    except (docker.errors.ImageNotFound):
        pass
    
    dockerClient.images.build(path = imagePath, tag = imageName, rm=True)
    container = dockerClient.containers.run(imageName, detach=True, name=imageName)

    while True:
        print(container.logs())
        time.sleep(1)

When running my custom container from command line docker run -it test , I have got the logs but when running it with docker-py with dockerClient.containers.run , My logs stays empty.

Does anybody have an idea how to get my logs from my custom container using dockerClient.containers.run ?

It seems to be an issue.

Start with the Python entry points shown in the example: dockerClient.containers.run(..., detach=True) and container.logs(). Compare their behavior with the command-line docker run -it test case and the provided C++ testApp output. Done means the container's startup output can be retrieved consistently through the docker-py call.