Shulou(Shulou.com)06/03 Report--

This article shows you what the Operators of Kubernetes in DevOps has, which is concise and easy to understand, which will definitely brighten your eyes. I hope you can get something through the detailed introduction of this article.

Kubernetes Operators is built for specific applications to simplify the creation, configuration, management, and operation of these applications on Kubernetes. In this article, we will delve into the mechanism of Kubernetes Operators and outline the 10 Operators that each DevOps needs to know.

The emergence of Kubernetes and container technology not only makes the application more portable and extensible, but also improves the utilization of resources. For DevOps, Kubernetes has a broader appeal: the ability to configure, manage and operate containerized microservices on a large scale.

At the same time, Kubernetes also allows them to achieve a certain degree of automation in the creation, deployment, extension and configuration of applications, thus greatly reducing the overhead and errors of program management.

However, there is one thing to note: for stateful applications, it may not be as perfect as we would like. Deploying, scaling, operating, and configuring stateful applications, as well as building in automation, requires a lot of interaction with DevOps.

What is Kubernetes Operators?

Kubernetes Operators is a high-level native way to build and drive each application on a Kubernetes cluster. Working closely with Kubernetes API, it provides a consistent way to automate all application operations without any human response.

In other words, Operator is a way to package, run, and manage Kubernetes applications.

Kubernetes Operators makes it easier to create, configure, and manage these applications on Kubernetes.

Many Operators also extend the entire application lifecycle, making it easier to perform operational tasks such as extending, upgrading, backing up, and restoring complex stateful applications.

Because they use and extend Kubernetes API, they are tightly integrated into the Kubernetes framework.

The following is a list of some of the most common features of Kubernetes Operators:

Install the application using configuration items and the number of copies.

Auto scaling application

Start upgrade, automatic backup and fault recovery

Perform administrative tasks in a code manner

How does Kubernetes Operators work?

Operators takes advantage of the extensibility and modularity of Kubernetes to automate the creation, configuration, and management of Kubernetes applications.

Operators is based on the concepts of custom Kubernetes controllers (CRD) and custom resources, and is used by DevOps to manage applications on Kubernetes. They work with CRD to ensure that the actual state of the cluster matches the state defined in CRD.

Prometheus Operator from CoreOs is a good example. It is deployed as a custom Kubernetes controller that monitors Kubernetes API's four custom resource definitions: Prometheus,ServiceMonitor,PrometheusRule and AlertManager.

After deployment, Prometheus Operator will install and configure the complete Prometheus Stack, including the Prometheus server, Alertmanager,Grafana, host node_exporter, and kube-state-metrics. DevOps can then easily expand the number of individual copies of each component, make configuration changes, update alert rules, or automatically monitor new services.

Now that we have introduced the concept of Kubernetes Operators, let's introduce some of the most commonly used Operators in DevOps.

Which Kubernetes Operators should you know?

RBAC Manager Operator

RBAC Manager (https://github.com/FairwindsOps/rbac-manager) is a kubernetns Operator from Fairwentz that is designed to make Kubernetes on RBAC easy to install, configure and manage.

Kubernetes authorization is usually complex, repetitive, requires a lot of manual configuration, and is difficult to extend. RBAC Manager greatly reduces the configuration involved in managing RBAC and creating, deleting, or updating role bindings, cluster role bindings, and service accounts.

By summarizing role bindings between multiple namespaces in a single RBAC definition file, it can be used as the only factual source of understanding of RBAC status.

MongoDB Enterprise Kubernetes Operator

MongoDB Operator (https://docs.mongodb.com/manual/reference/Operator/) helps DevOps standardize the process of creating large-scale MongoDB clusters. The Operator can take over typical administrative tasks involved in splitting and managing MongoDB clusters, including configuring storage and computing, configuring network connections, and setting up users.

Kubernetes Operator also integrates other MongoDB management tools, such as MongoDB Ops Manager and MongoDB Cloud Manager, to provide backup, monitoring, and performance optimization.

HPA Kubernetes Operator

Banzai's HPA Operator (https://github.com/banzaicloud/hpa-Operator) is another useful Operator that adds Pod automatic scaling to Helm charts.

It monitors Kubernetes deployment or StatefulSet and automatically creates, deletes, or updates "Horizontal Pod Autoscalers" (HPA) based on the comments defined in the configuration. The Github page of HPA takes Kafka as an example.

Kafka's Helm charts does not define any HPA rules for the cluster, which means that deploying Kafka using this helm chart does not implement HPA in Kafka.

To ensure that HPA is part of Helm chart, DevOps can add comments to min and maxReplicas. Once added, HPA Operator will increase the required number of HPA copies based on the comments. HPA Operator also uses Prometheus-based custom metrics exposed by Kube Metrics Adapter.

Cert-manager Kubernetes Operator

Cert-manager Operator (https://github.com/jetstack/cert-manager) is from Kubernetes Operator with Jetstack. It is designed to automate the management and issuance of TLS certificates.

DevOps can use this Operator to automate repetitive tasks, such as ensuring that certificates are valid, up-to-date, and updated correctly. After deployment, Cert-manager will run as Kubernetes deployment.

DevOps can configure certificates and certificate issuer lists as Kubernetes CRD. Once the certificate is configured, the application can make an immediate request using a trusted certificate.

ArgoCD Operator

ArgoCD Operator (https://github.com/jmckind/argocd-Operator) manages the entire lifecycle of ArgoCD and its associated components. ArgoCD is one of the most highly rated continuous delivery tools for Kubernetes in the field of CNCF.

The Operator allows easy configuration and installation of ArgoCD, as well as easy upgrade, backup, restore and extension of ArgoCD components. The Operator does this by monitoring three Kubernetes CRD, including ArgoCD (defining the required state of the ArgoCD cluster) and ArgoCDExport (defining the required state for ArgoCD component export and recovery).

Istio Operator

Istio (https://istio.io/) is a service grid tool that manages, coordinates, protects, and monitors communications between microservices deployed on Kubernetes. Istio Operator (https://istio.io/blog/2019/introducing-istio-Operator/) makes it easier to install, upgrade, and troubleshoot Istio.

Installation requires only istioctl as a prerequisite because they do not need to change API and version-specific upgrade hooks can be easily implemented. Installing Istio with this Operator also ensures that all API fields are validated.

The OperatorAPI supports all 6 built-in installation profiles, including default, demo, minimum and remote. DevOps and SRE can start with either of these and then make further configuration changes to customize the service grid to their specific needs.

Etcd Operator

Etcd acts as the storage database for all cluster data on the Kubernetes, so it is one of the most critical components of each cluster. Managing and configuring etcd clusters on Kubernetes is a relatively time-consuming task, which requires some manual configuration work, which increases the complexity to ensure the high availability, monitoring and disaster recovery of etcd clusters.

ETCD Operator (https://coreos.com/Operators/etcd/docs/latest/) makes it easier for DevOps and SRE to create, configure, and manage etcd instances in Kubernetes clusters.

Teams can easily start multiple highly available etcd instances without specifying detailed configuration settings, modifying cluster specifications to resize the cluster, configuring an automatic backup policy for disaster recovery, and starting a normal upgrade without downtime.

Elastic Cloud on Kubernetes (Elastic Kubernetes Operator)

Elastic cloud (ECK) on Kubernetes (https://github.com/elastic/cloud-on-k8s) is the official Operator of elastic.co 's Kubernetes, designed to provide a seamless experience for deploying, managing, and operating the entire Elastic Stack on Kubernetes. In addition to making it easier to deploy elastic search and Kibana on Kubernetes, it simplifies key operational tasks, including managing and monitoring multiple clusters, starting a normal upgrade, expanding cluster capacity and local storage, making configuration changes, and backing up.

The default ECK distribution is free and open source, with built-in features, including frozen indices for dense storage, Kibana space, Canvas and Elastic Maps, and support for monitoring Kubernetes logs and infrastructure.

Grafana Operator

Grafana Operator provided by RedHat (https://github.com/integr8ly/grafana-Operator) simplifies the process of creating, configuring, and managing Grafana instances on Kubernetes.

In addition to helping deploy Grafana, it also supports making Grafana highly available through ingress, automated dashboard and data source discovery, and dashboard dependency installation. You can install Operator manually and create custom resources using Ansible or through the kubectl command.

After installation, Operator monitors the dashboard definition in its own namespace or in all namespaces based on the flags passed during deployment, discovers the dashboard, adds error messages to the dashboard status field in the event of any invalid json, and automatically installs the specified plug-in.

DevOps and SRE can then add data sources and other configuration files to GrafanaDataSource.

Jaeger Kubernetes Operator

Jaeger Kubernetes Operator (https://github.com/jaegertracing/jaeger-Operator) helps deploy, manage, and configure Jaeger instances. When installing DevOps and SRE, you can specify configuration options for jaeger, including storage options, derived dependencies, injection Jaeger agent sidecar and UI configuration, and so on.

After the installation is complete, you can use Operator to create an Jaeger instance and associate it with the deployment policy. DevOps and SRE can choose from the following three supported deployment strategies: allInOne,production and streaming. When using the production policy, Operator starts a more scalable and highly available environment and deploys each back-end component separately.

The allInOne policy is intended for testing and development purposes, while the streaming policy enhances the production policy and provides a stream between the collector and the back-end storage.

What are the above contents of the Operators of Kubernetes in DevOps? have you learned any knowledge or skills? If you want to learn more skills or enrich your knowledge reserve, you are welcome to follow the industry information channel.

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.