Search code examples
dockerkubernetesopenshiftopenshift-3hawkular

Why is Openshift reporting a far greater container memory usage than the same Docker container running locally?

I have a Docker container that reports this resource usage when running locally:

docker run -i --rm -p 8080:8080 my-application

As you can see the container uses 10.6MiB:

docker ps -q | xargs  docker stats --no-stream             
CONTAINER ID        NAME                 CPU %               MEM USAGE / LIMIT    MEM %               NET I/O             BLOCK I/O           PIDS
b73afe5ee771        mystifying_neumann   0.00%               10.6MiB / 7.777GiB   0.13%               11.7kB / 2.38kB     0B / 0B             21

Now I run that container in Openshift, setting the following memory limits:

        resources:
          limits:
            memory: 64Mi
          requests:
            memory: 64Mi

When the pod starts I would be expecting ~11MiB used of a total of 64MiB. However the container is using 53MiB!!! Why this difference?

enter image description here

Solution

  • I finally found the reason of this difference in these two references:

    https://github.com/openshift/origin-web-console/issues/1315

    https://access.redhat.com/solutions/3639191

    Summing up: Docker reports memory as the addition of several elements such as rss and cache:

    https://docs.docker.com/config/containers/runmetrics/#metrics-from-cgroups-memory-cpu-block-io

    cache The amount of memory used by the processes of this control group that can be associated precisely with a block on a block device. When you read from and write to files on disk, this amount increases. This is the case if you use “conventional” I/O (open, read, write syscalls) as well as mapped files (with mmap). It also accounts for the memory used by tmpfs mounts, though the reasons are unclear.

    rss The amount of memory that doesn’t correspond to anything on disk: stacks, heaps, and anonymous memory maps.

    Openshift 3.x reads that information using Heapster and can`t tell appart both types of memory.

    If you check the docker stats of the container that is running inside Openshift you will find the expected (lower) value.