How does Docker for Windows run Linux containers?

In the old versions of Docker for Windows, I remember it explicitly said it used a linux VM for the kernel.

But since the new stable version (released in July 2016 I think), it says

  • Pull image from private Docker registry in SaltStack
  • Docker Compose Shared Folder
  • Port forwarding with Elastic docker image
  • Cassandra read/write bandwidth
  • Clear logs in native Docker on Mac
  • how to redirect no-www to www under jwilder/nginx-proxy?
  • Docker for Windows is a native Windows application with a native user interface and auto-update capability, deeply integrated with Windows native virtualization, Hyper-V, networking and file system

    If I understand correctly, the specified base image is for the user space and the host’s kernel is used.
    So, if I specify that I’m using an ubuntu base image to run the echo command, how does the Windows kernel come into play?

    Or am I completely misunderstanding something?

  • Docker container not starting (docker start)
  • Error: unknown shorthand flag: 'r' in -r
  • Docker container cannot clone public Github repo
  • “Memory used” metric: Go tool pprof vs docker stats
  • Docker compose version 2.1+ on mac
  • Issues connecting a Symfony2 app to a database container in Docker
  • One Solution collect form web for “How does Docker for Windows run Linux containers?”

    Docker for Windows still uses a Linux VM to run Linux containers. But instead of using Virtual Box (which is what is used with Docker Toolbox), the Linux VM is run using Hyper-V – a Windows-native hypervisor. This means that Docker for Windows ships fewer components and has less moving parts.

    If you install Docker for Windows and run docker version you’ll see that the Docker Linux daemon is running on “Moby Linux”.

    Docker will be the best open platform for developers and sysadmins to build, ship, and run distributed applications.