In the first two articles in this series, I explained why Kubernetes like a dump truck , and you want to understand as Kubernetes (and dump trucks, cranes, etc.) so elegant, professional tools is always a learning curve of. This article is the next step: to learn how to drive.
Recently, I saw on Reddit about important Kubernetes project posts. People seem to want to know that they should learn how to get started Kubernetes. "Driving a dump truck analogy" helps to ensure that this problem go back to the track. Someone mentioned in this post, unless absolutely necessary, you should not run your own mirror warehouse, so people began to accept control Kubernetes instead of building its ideas.
API is Kubernetes engine and transmission. Like dumpers steering wheel, clutch, brake and gas pedals, for building applications YAML or JSON file is the main interface of the machine. When you first learn Kubernetes, this should be your main concern. Know your control components. Do not be distracted at the latest and largest of those projects. When you learned to drive, do not try to drive experimental dumpers. Instead, focus on the basics.
Defined state and the actual state
First, Kubernetes follow the principles defined state and the actual state.
Defined state and actual state
Human (developer / system administrator / operation and maintenance personnel) to use their submission to Kubernetes API of YAML / JSON file specifies the state definition. Then, Kubernetes to analyze the difference between the new state and the cluster YAML / JSON defined in the actual state of use of the controller.
在上面的例子中，Replication Controller 可以看到用户指定的三个 pod 之间的差异，其中一个 pod 正在运行，并调度另外两个 Pod。如果你登录 Kubernetes 并手动杀死其中一个 Pod，它会不断启动另一个来替换它。在实际状态与定义的状态匹配之前，Kubernetes 不会停止。这是非常强大的。
原语
接下来，你需要了解可以在 Kubernetes 中指定的原语。
Kubernetes primitives
这些原语不仅仅有 Pod，还有部署Deployment、持久化卷声明Persistent Volume Claim、服务Service，路由route等。使用支持 Kubernetes 的平台 OpenShift，你可以添加构建build和 BuildConfig。你大概需要一天左右的时间来了解这些原语。你可以在你的用例变得更加复杂时再深入了解。
将原生开发者映射到传统 IT 环境
最后，考虑这该如何映射到你在传统 IT 环境中的操作。
Mapping developer-native to traditional IT environments
尽管是一个技术问题，但用户一直在尝试解决业务问题。从历史上看，我们使用诸如剧本playbook之类的东西将业务逻辑与单一语言的 IT 系统绑定起来。对于运维人员来说，这很不错，但是当你尝试将其扩展到开发人员时，它会变得更加繁琐。
直到 Kubernete 出现之前，我们从未能够以原生开发者的方式真正同时指定一组 IT 系统应如何表现和交互。如果你考虑一下，我们正在使用在 Kubernetes 中编写的 YAML/JSON 文件以非常便携和声明的方式扩展了管理存储、网络和计算资源的能力，但它们总会映射到某处的“真实”资源。我们不必以开发者身份担心它。
Therefore, attention quickly give up Kubernetes ecosystems new project, focusing instead start using it. In the next article, I'll share some can help you use Kubernetes tools and workflows.
via: https://opensource.com/article/19/6/kubernetes-basics
Author: Scott McCarty topics: lujun9972 Translator: geekpi proofread: wxy
This article from the LCTT original compiler, Linux China is proud