中文 | English
KubeOperator is an open-source light-weighted Kubernetes distribution that focuses on helping enterprises plan, deploy, and operate production-grade Kubernetes clusters in an offline network environment. It has a graphic Web UI that fasten up the process of software lifecycle in this current rapid cloud age.
KubeOperator uses Terraform to auto-build infrastructure on LaaS platform (vSphere, OpenStack, FusionCompute, user can also use their resources, e.g. VMs or On-premise). It also implements automated deployment and allows changing operation through Ansible, supporting Kubernetes clusters a full life-cycle self-defined management from Day 0 planning, Day 1 deployment, to Day 2 operating.
Note: KubeOperator passed the [Certified Kubernetes Conformance Program] (https://landscape.cncf.io/selected=kube-operator) provided by CNCF (Cloud Native Computing Foundation)
- Easy to Use: Using a visible Web UI that significantly lower down the difficulty of K8s deployment and management, built-in with Webkubectl;
- Offline Support: Continue updating Kubernetes and common components of the offline pack;
- Build by demand: Calling cloud platform API, build and deploy Kubernetes cluster in just a click;
- Scale by demand: Swiftly scale Kubernetes clusters and improve resources utilization;
- Patch by demand: rapid update, patch Kubernetes cluster and being up to date with the community version;
- Self Repair: Through rebuilding malfunction node to confirm the usability of the cluster;
- Full-Stack Monitoring: Full record of events, monitoring, warning and journaling from node, pod, to cluster;
- Multi-AZ Support: Master nodes are distributed in different failure domain to make sure the high usability;
- Marketplace: built-in with KubeApps marketplace. Able to deploy and manage common apps quickly;
- GPU Support: Support with GPU nodes which help operating high computation applications such as machine learning.
| Day 0 Planning | Cluster Model | 1 master node with N number of worker nodes : suitable for develop testing purpose |
| 3 master nodes with N number of worker nodes : suitable for production-grade purpose | ||
| Calculation Scheme | Independent Host : support self-prepared VMs, public clouds or physical machines | |
| vSphere Platform : Support auto-build host (using Terraform) | ||
| OpenStack Platform : Support auto-build host (using Terraform) | ||
| Storage Scheme | Independent host : Support NFS / Ceph RBD / Local Volume | |
| vSphere Platform : Support vSphere Datastore (Centralized storage that compatible with vSAN & vSphere) | ||
| OpenStack Platform : Support OpenStack Cinder (Centralized storage that compatible with Ceph & Cinder) | ||
| Network Scheme | Support Flannel / Calico Network Plug-in | |
| Support internet expose service through F5 Big IP | ||
| Support Traefik / Ingress-Nginx | ||
| Support CoreDNS | ||
| GPU Scheme | Support NVIDIA GPU | |
| Operating System | Support RHEL/CentOS 7.4+ | |
| Running on Container | Support Docker / containerd | |
| Day 1 Deploying | Deployment | Provide full installation package in an offline environment |
| Support a visible screen of the deploying process | ||
| Support one-click automation deployment (using Ansible) | ||
| Day 2 Operating | Management | Support project-centralized hierarchical authorization management |
| 3 roles: system admin, project admin and read-access user | ||
| Support docking with LDAP/AD | ||
| Expose with REST API | ||
| Install K8s Dashboard Management app through Kubeapps+ | ||
| Install Weave Scope Management app through Kubeapps+ | ||
| Support Web Kubectl UI | ||
| Built-in with Helm | ||
| Constant updating certificate | ||
| Observable | Built-in with Prometheus, support fully monitoring & alarming of clusters, pods, nodes, and container | |
| Built-in with Loki log system | ||
| Built-in with Grafana for monitoring & logs display | ||
| Support notification center, signaling various cluster unusual events through DingTalk or WeChat | ||
| Upgrade | Support whole cluster promotion | |
| Scale | Support flexible number of worker nodes | |
| Backup | Support periodical backup for etcd | |
| Safety compliance | Support score system for cluster’s health condition | |
| Support CSI Safe Scan | ||
| Kubeapps+ | Support CI/CD tools, e.g. GitLab, Jenkins, Harbor, Argo CD | |
| Support Machine Learning/AI applications like TensorFlow |
- Terraform: Allowing to auto-build VMs；
- Ansible: Using as an automated deployment tool；
- Kubeapps: Creating a marketplace based on Kubeapps.
Copyright (c) 2014-2019 FIT2CLOUD 飞致云
KubeOperator is licensed under the Apache License, Version 2.0.