The simple library to gather CPU and memory resource limits from within OS Container bases on Linux.

pip install oscontainer

The library can be used to acquire current limits exposed to the container or process as following:

from oscontainer import OSContainer

# Get the container instance
container = OSContainer()

# Check if current process runs inside container
if container.is_containerized():
    # Get active processors
    cpus_count = container.active_processor_count()
    
    # Get the limit of available memory for this process.
    mem_limit = container.memory_limit_in_bytes()
else:
    print("No Container support detected")

For extra metrics available see OSContainer.

To run library in the test container you need to build the docker image and run it as container.

Checkout the repository.

git clone [email protected]:yaricom/oscontainer.git

Build local image with oscontainer name.

cd oscontainer

docker build -f Dockerfile -t oscontainer . 

Run the image we built within container with specific CPU limits set. For details about setting container resources limits with Docker, see: Runtime options with Memory, CPUs, and GPUs.

The following command will start container and open interactive shell:

docker run -it --cpus=".5" oscontainer /bin/bash

Now, execute main.py script within shell to see resource constraints that was detected:

python main.py 

This will produce the output like following:

OSContainer:
==========================
   active processors: 1
      container type: cgroup2
--------------------------
 > MEMORY:
--------------------------
memory limit (bytes): -1
memory usage (bytes): 8413184
--------------------------
 > CPU:
--------------------------
               quota: 50000
              period: 100000
              shares: -1
         cpuset cpus: 0-2
==========================
System information:
==========================
multiprocessing.cpu_count: 3
     process cpu affinity: {0, 1, 2}
==========================