Today, apps are no longer developed and deployed the way they were used to before DevOps. Ever since DevOps became a mainstream thing in the IT and software industry, it has radically changed the operational infrastructure. Now, Developers and Operators work together to build, test, and deploy apps quickly and efficiently. That’s precisely where Docker fits in the picture.
Docker is a containerization software, or more precisely, a PaaS tool that allows you to fragment apps and their environments into smaller chunks that can be stored and deployed in any environment and machine in a hassle-free manner. With containerization, multiple containers (including apps, their system files, and environment) to work smoothly in one virtual machine without even sharing all the resources.
As a result, you can run multiple applications in the same virtual machine, where each application will perform as if it possesses its operating system, CPU, memory, storage, and other crucial resources.
The Docker ecosystem comprises four major projects, namely Moby, Kubernetes, and Containerd. However, that’s not all – there’s a lot more to Docker than these three projects. Each day the Docker community actively works on developing best-in-class projects and tools both with and for Docker so that Developers around the world can build and deploy apps that can run anywhere. If you want to learn DevOps, Docker is an important section to learn.
In this post, we’ll highlight some of the best Docker projects that every aspiring Developer should know of!
Top Docker Project Ideas
Anyone who uses Docker probably knows what Rancher is. It is a containers-as-service vendor that includes a host of some of the most widely used container orchestration and scheduling frameworks – Docker Swarm, Kubernetes, and Mesos. RancherVM is one of Rancher’s lesser-known projects that allows you to run a VM inside a Docker container. It is equipped with a Web UI for managing public keys, compute nodes, VMs, and to access VNC consoles from a web browser.
With RancherVM, you can build VMs that run inside Kubernetes pods, also known as VM Pods. Each VM pod includes a container running a virtual machine instance. At present, RancherVM only supports KVM images. So, you can package any KVM image as a Docker image, distribute it using any Docker registry (example, DockerHub), and run it on RancherVM.
Furthermore, RancherVM extends the Kubernetes API with Custom Resource Definitions (CRDs). You can define a VirtualMachine CRD specification specifying what base image, how much compute resources, and what keypairs are authorized to initiate an SSH session. Then, a Kubernetes controller will create VM pods according to the specifications provided by you and display the same in the VirtualMachine CRD status.
In November 2015, Microsoft released Dockercraft before the public eye. It was launched as an open-source on GitHub, explicitly designed to allow developers to manage their Docker containers in Minecraft, a popular game that Microsoft bought for US$ 2.5 billion!
According to GitHub, Dockercraft is a “simple Minecraft Docker client, to visualize and manage Docker containers” in Minecraft’s 3D environment. It isn’t exactly a complicated Docker manager implemented in Minecraft, but rather a simplistic way of visualizing and flipping switches in Docker containers.
It is important to remember that you must have the full version of Minecraft on your PC to use Dockercraft. If you have it already, you can download Dockercraft from Github. After you’ve completed the setup, you have to run a Dockercraft container and join a custom Minecraft server – Cuberite.
Docker’s biggest drawback is that Dockerized apps are not cross-platform, meaning you cannot natively run a Docker Windows app on Linux. To solve this issue, Wine was created. Wine on Docker was developed to enable Windows apps to run on Linux OS just as a local application would run from anywhere in the world. On Docker, it is known as Wine, whereas in GitHub, we call it Docker-Wine.
Wine uses the Wine compatibility layer to support Windows apps on Linux. Since Wine comes with Dynamic Video drivers and VirtualGL with both local and remote support, it should perform well with out-of-the-box Nvidia cards and Nvidia drivers and even other cards that use Mesa drivers. You can configure Wine to auto adapt to whatever drivers you may have installed, provided they are the most recent ones for your branch.
At present, there are only a handful of monitoring services for Docker images, for instance, Shipyard. And it is a well-established fact that developers/programmers love tools that let them access all the essential requirements inside the shell. Docker-Mon is one such console Docker monitoring tool. Docker-mon is based on blessed-contrib, a shell-centered dashboard framework.
This project is still a work in progress, and it relies on the docker stats feature that is yet to be launched. So, to use Docker-Mon, you will need either a master build or the upcoming v1.5.0 (you can check the launch for v1.5.0-rc1 here).
We all know that Phusion Passenger is a favorite of Rails developers. The primary goal of this app server is to create a suitable environment where developers can run their applications smoothly. This idea is what sparked the creation of the Passenger-Docker project. The aim was to develop a sensible default base image for developers who wish to implement Docker in their production environment.
Passenger-Docker is a suite of Docker images designed to serve as bases for Ruby, Python, Node.js, and Meteor web app images. Maintaining the core mantra of Phusion Passenger, the Passenger-Docker project seeks to simplify and quicken the process of making Docker images for web apps.
KubeVirt is a VM management add-on for Kubernetes. It was developed to offer a common ground for building virtualization solutions on top of Kubernetes. Essentially, Kubernetes was created to manage containers and not other types of infrastructure. However, you can use Kubernetes to orchestrate VM clusters. This is what KubeVirt does – it adds extra functionalities to Kubernetes so that it can seamlessly support VMs.
KubeVirt aims to extend the functionalities of Kubernetes by incorporating additional virtualization resource types through Kubernetes’s Custom Resource Definitions API. You can use KubeVirt to build a predefined VM, launch a VM, stop a VM, delete a VM, and also schedule a VM on a Kubernetes cluster. However, KubeVirt is still a work in progress.
Docket is a custom Docker registry that facilitates lightning-fast deploys via BitTorrent. It lets you deploy Docker images across a large cluster of machines at an impressive speed. Docket was designed and developed in 48 hours in the Gopher Gala Golang 48 hour hackathon! Written in Golang, Docket is user-friendly, and it can work well with your private Docker registry.
Docket is made up of three main components –
- Docket Registry: It is a REST service that functions as a registry. It can receive Docker image tarballs from the client, store metadata into a database, and create torrents out of them as well as seed them.
- Docket Client: It is the interface with which the end-user interacts. It allows you to view available images in the registry, push an image to the registry, and pull an image from the registry (this triggers a BitTorrent deployment).
- BitTorrent Tracker: Although Docket has the BYOT (bring your tracker) feature, we recommend that you install opentracker for tracking BitTorrent.
8. Container Migration Tool (CMT)
A good thing about Docker containers is that they are portable, as in, they can run on any type of host that has Docker installed. However, Docker lacks the built-in feature that lets you move running containers from one cluster to another over a network. To add to the challenge, Checkpoint & Restore (CR) is still a relatively new feature that is not generically available to container users. CR demands a level of understanding about its functionality, and usually, most users who aren’t well-versed with CR functionality get errors while performing CR owing to the differences/limitations between the source and the target host.
Container Migration Tool or CMT was specifically developed to combat this issue of migration. The idea behind the creation of CMT was to create an external command-line tool that can be used with Docker or runC that facilitates the live migration of containers between different hosts by performing pre-migration validations and auto-discovering suitable target hosts.
You can use the dockerfile_lint rules to check file syntax, arbitrary semantics, and best practice attributes that are determined by the rule file writer. Also, you can use the linter to check LABEL rules against Docker images. The rule files are written in YAML. All the rules are implemented using regular expressions, and they can run on one instruction of the Dockerfile at a time. The rule file consists of four sections –
- Profile section: It provides necessary information about the rule file that helps the user to select an appropriate rule file for a specific Dockerfile.
- General section: It contains general syntax rules.
- Line rule section: It contains rules that must be run on a given instruction in the Dockerfile. There are rules for checking the syntax of each instruction.
- Required instruction section: It contains a list of instructions that must exist in the Dockerfile so that it can be considered valid.
Dockermix is a command-line tool used for developing and managing multi-container Docker environments. In this project, the container sets are defined in the YAML format that emulates the options available in the Docker API. The aim behind creating Dockermix was to allow developers to effortlessly launch, orchestrate, and deploy complex multi-node environments for testing and development.
11. Docker Cookbook
Docker Cookbook includes all the necessary resources for installing Docker and building, managing, and running Docker containers. This Docker project is only associated with the Docker container engine that is distributed by Docker, Inc., and it does not address Docker ecosystem tooling or other technologies like Cgroups or AUFS.
To use the Docker Cookbook, you must have Chef v12.15 or later installed in your machine. The Cookbook offers support for multiple platforms, including Amazon Linux, Fedora, CentOS 7, Debian 8/9, and Ubuntu 14.04/16.04. Also, it can automatically set up the upstream Docker package repositories.
Dokku is a Docker-powered mini-Heroku, the smallest PaaS implementation to have ever been developed. This PaaS solution allows users to configure and deploys applications to a production environment on a separate server. It uses Docker to help developers build and manage the complete lifecycle of applications.
You can install Dokku on any machine and even use it on inexpensive cloud providers. Additionally, you can write Dokku plugins in any language, share them online, and extend the already available plugins.
These are only some of the many, many projects that have been developed with Docker as their foundation. Although you don’t need all of these Docker tools for your next project, it is good to know that you have a multitude of options!
If you’re interested to learn more about big data, check out upGrad & IIIT-B’s PG Diploma in Full-stack Software Development which is designed for working professionals and offers 500+ hours of rigorous training, 9+ projects and assignments, IIIT-B Alumni status, practical hands-on capstone projects & job assistance with top firms.