Platform Center
Home / Platform management / Clusters / Clusters / Plugin management / Plugin Overview

Plugin Overview

You can extend the capabilities of the cluster by deploying feature plugins for the cluster, and you can manage the cluster through the Platform Management plugin (upgrade, uninstall, update configuration parameters).

Tip: Plugins can only be upgraded by upgrading the product. When the product is upgraded, the plugins in the cluster will be upgraded to the latest version along with the product.

Plugin Function Introduction and Deployment Instructions are shown in the table below.

Monitoring

Plugin Name Function Applicable Cluster Types Deployment Instructions Common Issues
Prometheus Collects, stores, queries, pushes, and visualizes monitoring data for nodes in the cluster. - Self-built cluster
- Standard Kubernetes cluster connected to the platform
- AWS EKS cluster connected to the platform		 Cannot be deployed on the same cluster as OCP Monitoring Registration Component or VictoriaMetrics.
Deployment Guide 		- How to choose a monitoring component?

- How to update alert rules after replacing Prometheus with VictoriaMetrics?

- How to back up monitoring data for Prometheus?
VictoriaMetrics Collects, stores, queries, pushes, and visualizes monitoring data for nodes in the cluster. - Self-built cluster
- Standard Kubernetes cluster connected to the platform		 - Cannot be deployed on the same cluster as OCP Monitoring Registration Component or Prometheus.
Deployment Guide 		- How to choose a monitoring component?

- How to update alert rules after replacing VictoriaMetrics with Prometheus?

- How to back up monitoring data for VictoriaMetrics?
OCP Monitoring Registration Component Registers and connects the deployed monitoring components on the OpenShift cluster to the platform. After successful registration, the monitoring data of the OpenShift cluster can be viewed and displayed on the platform. OpenShift cluster connected to the platform Cannot be deployed on the same cluster as Prometheus or VictoriaMetrics.
Deployment Guide
Note: When uninstalling the component, only the registration and connection information of the monitoring component (on the OpenShift cluster) on the platform will be deleted, and the monitoring component will not be uninstalled from the cluster.		 -
etcd Extended Monitoring Panel etcd is a core component of the Kubernetes cluster and stores all cluster information. To better monitor the overall operation status of the etcd cluster and help troubleshoot and handle etcd-related issues, you can install the etcd extended monitoring panel. Note: This plugin has a certain performance impact and should be uninstalled promptly after troubleshooting. Self-built cluster with X86 hardware architecture, and the operating system type and corresponding kernel version of the cluster must meet the following corresponding relationships:
- CentOS 7.6/7.7/7.8/7.9: Kernel version can be 3.10.0-1160 or 3.10.0-1127
- Redhat 7.8: Kernel version is 3.10.0-1127.el7.x86_64
- openEluer 22.03 SP3: Kernel version is 5.10.0-60.18.0.50.oe2203.x86_64
- Ubuntu 22.04: Kernel version is 5.15.0-56-generic		 Before deploying this plugin, please deploy Prometheus or VictoriaMetrics first.
Deployment Guide 		-

Logging

Plugin Name Function Applicable Cluster Types Deployment Instructions Common Issues
Log Storage Component A plugin based on Elasticsearch and Kafka, used to store logs, events, and audit data in the cluster, and provide message distribution capabilities. All types of clusters managed by the platform Deployment Guide
Note: Before uninstalling the component, you need to uninstall the Log Agent deployed as the storage cluster for the current cluster.		 -
Log Agent A plugin based on Filebeat, used to collect and forward log data from Linux nodes in the cluster. All types of clusters managed by the platform This component can only be deployed when at least one cluster on the platform has deployed the log storage component.
Deployment Guide 		-

Storage

Plugin Name Function Applicable Cluster Types Deployment Instructions Common Issues
Container Object Interface Container object storage interface, aims to be a common abstraction layer across multiple object storage providers, dynamically providing object storage services for workloads. All types of managed clusters Refer to the section Deploy the container object storage interface plugin in Creating StorageClasses: Ceph Object Storage (Alpha) .
NFS Storage Plugin Based on the community NFS CSI (Container Storage Interface) storage access plugin, provides the ability to access multiple NFS storage systems or accounts. All types of managed clusters Refer to the section Deploy NFS storage plugin in Creating StorageClasses: NFS Shared Storage .
Snapshot Controller PV volume snapshot component, can be used with snapshots in StorageClasses that support them after deployment. All types of managed clusters Refer to Deploying the volume snapshot component .

Networking

Plugin Name Function Applicable Cluster Types Deployment Instructions Common Issues
MetalLB Provides LoadBalance type Service support in a native way in Kubernetes, out of the box. All types of managed clusters Refer to the Deploy metallb plug-ins section in External Address Pool .
Network Health Check Plugin This component is responsible for the network status of the cluster. It detects the destination address in the probe items using both host network and container network agents on all nodes in the cluster, using a specific protocol, to help you understand the network health status of the cluster. All types of managed clusters Deployment Guide
CoreDNS CoreDNS is a flexible and scalable DNS server that provides service discovery capabilities in a Kubernetes cluster. Self-built clusters Deployment Guide
Multus CNI Serves as an intermediary between other CNI plugins and Kubernetes, enabling Kubernetes to support multiple network cards for Pods. All types of managed clusters Deployment Guide -

Cost Analysis

Plugin Name Function Applicable Cluster Types Deployment Instructions Common Issues
Kubecost A cost analysis tool based on Kubernetes. It provides dynamic and multidimensional cost analysis of Kubernetes applications by analyzing cluster resources dynamically. This helps users analyze and control the cost of Kubernetes applications. All types of managed clusters Deployment Guide -
CostManager A self-developed cost management component of the platform. It offers multidimensional cost aggregation and can efficiently process large amounts of data, providing accurate cost analysis capabilities. This assists enterprises in managing resource allocation and optimization decisions effectively. global Deployment Guide -
BillingServer The platform’s self-developed components, together with the billing client, support billing and analysis of multiple resources within the platform, assisting enterprises to better operate the platform’s overall resources and improve the efficiency of use. global Deployment Guide -
BillingAgent The platform’s self-developed component, used in conjunction with the billing server, supports the collection of metering data from within the cluster for billing and analysis, and relies on the cluster having the Prometheus component installed. All types of managed clusters Deployment Guide -

Safety and security

Plugin Name Function Applicable Cluster Types Deployment Instructions Common Issues
Security Component Provides container runtime security detection capabilities. Self-built clusters with X86 hardware architecture, and the cluster’s operating system type and corresponding kernel version must meet the following corresponding relationships:
- Centos: Kernel version can be 3.10.0-1160.11.1.el7.x86_64, 3.10.0-1160.el7.x86_64, 3.10.0-1127.el7.x86_64;
- Redhat: Kernel version is 3.10.0-1127.el7.x86_64;
- Euler: Kernel version is 3.10.0-862.14.1.5.h442.eulerosv2r7.x86_64.		 This component cannot be deployed when the platform does not enable the Safety panel and Security rules features.
Deployment Guide 		-
Gatekeeper Gatekeeper is a policy controller for Kubernetes, which can be seen as a customizable Kubernetes Admission Webhook. Its main goal is to help organizations apply policies and security best practices to Kubernetes environments quickly and directly. All types of clusters managed by the platform Deployment Guide -
NeuVector End-to-end container security system plugin that provides enterprise-level zero-trust security solutions for containerized workloads in the cluster. Linux clusters with X86 hardware architecture Deployment Guide -

DevOps Tools

Plugin Name Function Applicable Cluster Type Deployment Instructions
Bitbucket Integration This plugin is a prerequisite for integrating with the Bitbucket code repository. As a Git-based code hosting and collaboration tool, Bitbucket provides functions such as code review, PR merging, issue tracking, builds, and pipeline. After deploying the plugin and integrating with Bitbucket, you can use Bitbucket in GitOps applications, CI/CD, code management, and other functions in ****. Self-built cluster
  1. Download the plugin from the Customer Portal App Market.
  2. Refer to the “DevOps Plugin Center Registration Plan” in the Customer Portal knowledge base to register the plugin.
  3. Deploy the plugin in the global cluster and the cluster where the Bitbucket resources to be used are located.
What’s next: Integrate Bitbucket in the DevOps Toolchain.
HCS OBS Integration This plugin is a prerequisite for integrating with HCS OBS (Object Storage Service). OBS is an object-based storage service that provides massive, secure, highly reliable, and low-cost data storage capabilities, as well as multiple selectable StorageClasses types. After deploying the plugin and integrating with HCS OBS, you can manage, deploy, or update binary artifacts in **** on the ServiceStage platform. Self-built cluster
  1. Download the plugin from the Customer Portal App Market.
  2. Refer to the “DevOps Plugin Center Registration Plan” in the Customer Portal knowledge base to register the plugin.
  3. Deploy the plugin in the global cluster and the cluster where the OBS resources to be used are located.
What’s next: Integrate HCS OBS in the DevOps Toolchain.
HCS SWR Integration This plugin is a prerequisite for integrating with HCS SWR (SoftWare Repository for Container). SWR is a repository that supports the full lifecycle management of images, providing simple, easy-to-use, secure, and reliable image management functions. After deploying the plugin and integrating with HCS SWR, you can manage, deploy, or update image artifacts in **** on the ServiceStage platform for containerized applications. Self-built cluster
  1. Download the plugin from the Customer Portal App Market.
  2. Refer to the “DevOps Plugin Center Registration Plan” in the Customer Portal knowledge base to register the plugin.
  3. Deploy the plugin in the global cluster and the cluster where the SWR resources to be used are located.
What’s next: Integrate HCS SWR in the DevOps Toolchain.

Other

Plugin Name Function Applicable Cluster Types Deployment Instructions Common Issues
AWS EKS Cluster Management Component Platform extension component that supports AWS integration for creating and managing EKS clusters. global Deployment Guide This plugin cannot be uninstalled if an AWS EKS cluster has been created through the platform.
Huawei Cloud CCE Cluster Management Component Platform extension component that supports Huawei Cloud integration for creating and managing CCE clusters. global Deployment Guide This plugin cannot be uninstalled if a Huawei Cloud CCE cluster has been created through the platform.
Azure AKS Cluster Management Component Platform extension component that supports Azure integration for creating and managing AKS clusters. global Deployment Guide This plugin cannot be uninstalled if an Azure AKS cluster has been created through the platform.
Google GKE Cluster Management Component Platform extension component that supports Google integration for creating and managing GKE clusters. global Deployment Guide This plugin cannot be uninstalled if a Google GKE cluster has been created through the platform.
Backup Restore Plugin Plugin based on Velero that provides backup and restore capabilities for cluster business resource data (e.g., applications) using backup files. All types of clusters managed by the platform Deployment Guide -
Online operations Enables the automatic synchronization of monitoring data from all clusters on the platform to the Customer Portal system when Online operations are enabled and the platform can connect to the Internet. global Deployment Guide -
Vertical Pod Autoscaler Based on the Vertical Pod Autoscaler open-source component, this plugin analyzes the historical resource usage of Pods to recommend appropriate resource quotas and improve resource utilization. - Self-built clusters
- Standard Kubernetes clusters connected to the platform
- AWS EKS clusters connected to the platform		 Deployment Guide -
Warpgate Platform-developed component that provides image proxy and image P2P distribution capabilities for massive clusters in edge computing scenarios. Self-built clusters Deployment Guide -
Argo CD A tool for declarative GitOps pipeline (CD) for Kubernetes. It uses Git as its core and supports the rapid deployment of Kubernetes resources to multiple clusters by defining various objects declaratively. This makes it easy to track and deploy the same Kubernetes resources in cross-cluster scenarios. global Deployment Guide -