Container Shipyard

Rarely has such young software raised so much attention in such a short time as the Docker [1] container manager. Docker is one of the fastest growing open source developments in the history of free software. Released in March 2013, it made it to version 1.0 in June 2014 and was declared ready for production. In the same month, Docker was already part of Red Hat's RHEL 7 release. Further cooperative agreements exist with Canonical, openSUSE, CoreOS, VMware, IBM and others. Moreover, Docker is a member of Google's Kubernetes project [2].

The image associated with the name Docker sums it up: a ship (computer) is loaded with containers that you can stack, transport, isolate, and protect (Figure 1). Docker containers in IT are isolated virtual environments for isolated execution of processes and applications. Docker is currently used primarily for developing, testing, and deploying applications and can isolate them along with their dependencies in containers. These applications have practically no overhead and can run nearly as fast as native applications.

Figure 1: The Docker logo says it all: The underlying computer is loaded with containers containing isolated images.

Unlike the well-known hypervisors [3] VirtualBox, VMware Workstation, HyperV, Parallels Desktop, or Xen, a Docker container doesn't need its own operating system. Each container uses the basic functions of its host system and is therefore self-sufficient and isolated from other containers. Only the host's kernel needs to match that of the guest.

The Basics

Docker uses the group (cgroup) [4] and kernel namespace [5] technologies. Namespaces serve to isolate and protect, whereas cgroups ensure that the container can use the resources selectively granted to it by the host. Until version 0.8 (1.3 is the current version) Docker also availed itself of the functions of Linux Containers (LXCs) [6] but eventually developed its own API named LibContainer in version 0.9 that is also available to other projects.

Currently, Docker is mainly used as a Platform as a Service (PaaS) [7], but it is becoming quite suitable as a resource-friendly alternative to desktop virtualization (Figure 2). Graphical tools are being developed to handle containers, and I will present one of them, Panamax, later in this article as one of the first interfaces its kind.

Figure 2: Comparing Docker with a virtual machine. Although VMs implement a whole OS for each instance, Docker shares an OS for all instances.

The Components

The Docker system consists of three main components: a daemon, a client, and images. The daemon is responsible for orchestrating the containers; the client is what you use to control Docker with commands. The client and daemon can run on the same machine or the client can contact the daemon on a data center server or over the cloud. Docker images are the equivalent of virtual machines (VMs) for VirtualBox or appliances in VMware. Compared with these, Docker images are extremely space-saving, can be stacked, and can be versioned in a CVS [8].

The Docker hub is laid out as Software as a Service (SaaS) [9] and includes the repositories that now have around 15,000 complete image recipes (Figure 3). These are different from base images, which represent a base operating system along with related images containing one or more applications. Creating your own images isn't rocket science after a bit of training, but you need to know the command language of the distribution. An example would be to create an image using WordPress, PHP, MariaDB, SSH, or Nginx.

Figure 3: The Docker hub provides thousands of ready-to-use images.

Overlaid Layers

Thanks to the overlay functions supplied by Aufs [10] or the Device mapper [11], Docker has an efficient back end to store the containers. Each container, to put it simply, is stored as a transparent layer. These functions are the same ones that allow us to seemingly write to a Live CD (Figure 4). Quite recently, Btrfs [12] has been playing a role as well with its snapshot function. Details on the different filesystem functions for storing and managing containers can be found in Red Hat's documentation for their Atomic [13] project, a container operating system [14] still being developed.

Figure 4: The Docker layer model is similar to the layers on a Live CD.

Before I start going into details, I'll clarify things a bit. A Docker image is just a read-only template, a kind of recipe. A base image [15] provides rudimentary operating system functions and can have a parent-child relationship with other images that interact with the base image on various layers. For example, this could be an ownCloud image, which I'll get to later.

As a rule, containers consist of several images that contain all the functionality and dependencies that the content of the container (an operating system, application, or service) needs at run time. The basic handling of Docker containers is through a command-line interface resembling Git [16] and is easy to learn. The "Containers, Images, and Docker Files" box provides more information on how things work.

Figure 5: Debian in execution.

A container thus consists of a base operating system with apps and metadata added by users. The image tells Docker what the container should include and, where appropriate, what command is used for startup. In the simplest case the container starts right away with a shell, although far more complicated processes can occur, such as automated test runs.

Because images are read-only, Docker opens a layer for the containers with help from a Union filesystem, such as Aufs or Device mapper. These layers are writeable (wr), and the application runs within them. If a change occurs, such as an update, another layer is applied. Instead of changing the entire image, changes at run time are integrated only when you convert the container into an image again using:

docker commit

Docker containers come two ways: interactively started or daemonized. If the container is to execute a predetermined sequence and then stop, start it interactively with:

docker run -t -i debian:sid </command_1;command_2>

A container started with

docker run -d debian:sid /bin/bash

waits in the background for input.