Shulou(Shulou.com)05/31 Report--

This article introduces the relevant knowledge of "what are the functions of Kubernetes v1.17". In the operation of actual cases, many people will encounter such a dilemma, so let the editor lead you to learn how to deal with these situations. I hope you can read it carefully and be able to achieve something!

Kubernetes v1.17 contains 22 enhancements, of which 14 have gradually stabilized, 4 have entered the beta version, and 4 have entered the alpha version. This version update mainly focuses on several major topics, such as Cloud Provider Labels reaching GA, Volume Snapshot entering Beta version, CSI Migration reaching Beta, and so on.

Cloud Provider Labels achieves GA

When creating nodes and volumes, Kubernetes applies a set of standard tags based on the underlying cloud provider of the cluster. The node gets the label of the instance type, and both the node and the volume get two labels, usually describing the location of the resource in the cloud provider topology in terms of region and zone.

Kubernetes components use standard tags to support certain features. For example, the scheduler will ensure that the Pod is placed in the same zone as the volume they declare; when scheduling the Pod that belongs to a deployment, the scheduler will first distribute them in each zone. You can also use tags in Pod specs to configure things such as node affinity. Standard tags allow developers to migrate Pod specs between different cloud providers.

In Kubernetes v1.17, these tags are already widely used.

Volume Snapshot enters the Beta version

In Kubernetes v1.17, Volume Snapshot enters the Beta phase. Now let's get to know what Volume Snapshot is.

Many storage systems, such as Google Cloud Persistent Disks,Amazon Elastic Block Storage, can create "snapshots" of persistent volumes. A snapshot represents a point-in-time copy of a volume that can be used to set up a new volume or restore an existing volume to a previous state.

Why add Volume Snapshot to Kubernetes?

Kubernetes Volume's plug-in system provides powerful abstraction capabilities that can automatically generate, load, and mount blocks and files.

The goal of supporting these features is the workload portability of Kubernetes: the goal of Kubernetes is to create an abstraction layer between distributed system applications and the underlying cluster so that applications can be independent of the specific circumstances of the cluster they are running on and do not need to be "cluster-specific" at deployment time.

Kubernetes Storage SIG sees snapshot operations as a key feature of many stateful workloads. For example, a database administrator might want to take a snapshot of a database volume before database operation and maintenance. By providing a standard way to trigger snapshot operations in Kubernetes API, Kubernetes users can easily deal with these scenarios without using Kubernetes API (performing storage system-specific operations manually).

Kubernetes users are now authorized to incorporate snapshot operations into their tools and policies in a cluster-independent manner and know that it works for any Kubernetes cluster, regardless of the underlying storage.

In addition, these Kubernetes snapshot primitives serve as basic features that can be used to develop advanced enterprise-level storage management capabilities for Kubernetes, including applications or cluster-level backup solutions.

CSI Migration achieves Beta

In Kubernetes v1.17, the container storage interface CSI Migration has entered the Beta phase.

Why should we migrate the in-tree plug-in to CSI?

Before CSI, Kubernetes provided a powerful volume plug-in system. These plug-ins are in "in-tree" mode, which means that their code is part of the core Kubernetes code and is released with the core Kubernetes binaries. However, it is difficult to add a new volume plug-in to Kubernetes. First of all, storage vendors must be consistent with Kubernetes's release process if they want to add support for other storage systems to Kubernetes. Second, third-party stored code can cause reliability and security problems in the core Kubernetes binaries, and it is difficult for operators to test and maintain. Using CSI in Kubernetes can solve these problems.

In the future, more and more CSI drivers will be developed and applied in practice, and more and more Kubernetes users will benefit from the CSI model.

With CSI Migration, you can replace an existing in-tree storage plug-in, such as kubernetes.io/gce-pd or kubernetes.io/aws-ebs, with the appropriate CSI driver. After migration, the end users of Kubernetes will not realize the difference between the two and can continue to rely on all the functions of the in-tree storage plug-in using the existing interface.

When the Kubernetes cluster administrator updates the cluster to enable CSI migration, the existing stateful deployment and workload will continue to function as usual; but in fact, Kubernetes has handed over all storage management operations (previously for the in-tree driver) to the CSI driver.

Building a full-stack cloud native service platform based on Kubernetes

Hourly Cloud is the leading full-stack cloud native technology service provider in China. The company has built a full-stack cloud native service platform with container + Kubernetes as the core of the architecture. The hourly speed cloud full stack cloud native service platform includes two core product systems of infrastructure and cloud native application architecture, among which cloud native infrastructure products include container cloud PaaS, cloud native DevOps, middleware services and edge cloud computing, which provide agile and highly resilient running support environment for cloud native applications and improve the efficiency of application delivery, operation and maintenance. Cloud native application architecture products include micro-service governance, service grid, API gateway and Serverless services, which help enterprises easily cope with high-concurrency and high-performance business scenarios under large-scale distributed systems, and improve the agility and scalability of business application architecture.

The combination of hourly speed cloud full stack cloud native service platform and Kubernetes is embodied in the following points:

With the improvement of Kubernetes deployment mode, a secure Kubernetes cluster can be deployed with one click, and the high availability of the environment can be maximized. Compared with the native solution, the deployment is simpler and the availability is higher.

Tuning the operating system and Kubernetes, controlling the resource allocation of system components and node resource reservation, ensuring the stable operation of nodes and services for a long time.

Through the friendly extension of Kubernetes model + controller, the complete lifecycle management of database, cache, message queue and third-party products is realized.

Based on the DevOps product system developed by Kubernetes, DevOps and PaaS have the same architectural background and technical system, and the experience of any PaaS layer can be used to apply and innovate on the DevOps platform.

At present, the hourly speed cloud full stack cloud native service platform has served many large enterprises and Fortune 500 customers in the fields of finance, energy, manufacturing, radio and television, operators and so on. Benchmark customers include State Grid, Taiping Group, Hisense Group, China Southern Airlines, China Telecom, China Nuclear Corporation, Mercedes-Benz, China FAW and so on.

It is the core mission of cloud-speed cloud to assist enterprises in digital transformation through cloud native technology. The company will closely follow the development trend of cutting-edge technology and actively promote the development and landing of cloud-based technology. At the same time, it will also integrate more excellent and new features of Kubernetes cloud native platform into its own products to help more enterprises successfully complete their digital transformation.

Updates to other Kubernetesv1.17 features:

Node stain based on condition

Configurable Pod process namespace sharing

Scheduling DaemonSet Pod through kube-scheduler

Dynamic maximum Volume count

Kubernetes CSI topology support

Provide environment variable extensions in SubPath Mount

Default settings for custom resources

Move frequent Kubelet heartbeats to Lease Api

Split Kubernetes test package

Support for monitoring tags

This is the end of the content of "what are the functions of Kubernetes v1.17". Thank you for reading. If you want to know more about the industry, you can follow the website, the editor will output more high-quality practical articles for you!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.