For users in China experiencing image loading issues, please visit the Gitee mirror site: https://gitee.com/starsl/KubeDoor

• 💠New Architecture
• 💎Feature Description
  • 💥New Version Highlights
  • 📡Microservice Monitoring & Analysis: Multi-K8S Cluster Unified Monitoring Best Practices
  • 🎛K8S Resource Maintenance Management: Enhanced Management Features
  • 🧬Exception Alert Notification & Intelligent Aggregation
  • 💠Peak Period P95 Resource Collection & Analysis
  • 🚧K8S Admission Control Based: Ensuring Strong Consistency Between Microservice Requirements and Peak Resources
  • ✨Others
• 🚀Deployment Instructions
• 🔔KubeDoor Community
• 🙇Contributors
• 🥰Acknowledgments

🌼花折 - KubeDoor is a microservice resource management platform developed with Python + Vue, based on K8S admission control mechanisms. It supports unified remote storage, monitoring, alerting, and display for multiple K8S clusters. Focusing on the resource perspective of microservices during daily peak periods, it implements resource analysis, statistics, and strong governance for microservices, ensuring consistency between microservice resource request rates and actual usage rates.

• 🌊Based on the VictoriaMetrics suite, implementing a one-stop K8S monitoring solution for multi-K8S cluster unified remote storage, monitoring, alerting, and display.
• 🎨Integrated K8S node monitoring dashboard and K8S resource monitoring dashboard, both supporting viewing of various K8S cluster resources in a single dashboard.
• 📐Built-in alert rules for K8S resources, JVM resources, and K8S nodes, supporting unified alert rule management, integration with various IM alerts, and flexible @ mechanisms.

• 🎭K8S workload real-time monitoring management page: Real-time viewing of microservice status and Pod log tracking with automatic multi-color marking for various exception levels, supporting keyword filtering.
• 📀Microservice version updates support designated accounts, specified time periods for authorized operations, and automatic tag retrieval from image repositories. 【K8S Microservice Image Update Configuration Guide】
• ⏱️Supports immediate, scheduled, and periodic execution of microservice isolation, scaling, and restart operations.
• ♨Provides extensive one-click JVM performance analysis operation support for JAVA microservices.
• 🌐Supports management and synchronization of multi-K8S ISTIO VirtualService rules.

• 🦄K8S microservice unified alert analysis and processing page with daily intelligent aggregation display, processing markers, daily cumulative counting for identical alerts, providing clear daily alert overview.
• 🕹️Supports operations on PODs including isolation, deletion, Java dump, jstack, jfr, JVM data collection and analysis, with IM notifications.
• 📺New K8S event collection, analysis & monitoring alerts! Multi-K8S event collection and query analysis with flexible K8S event alert rule configuration. 【K8S Event Alert Rule Configuration Guide】

• 🎨Daily dimension-based collection of P95 resource data during peak periods, enabling excellent observation of long-term resource changes for various microservices, with smooth performance even when viewing 1 year of data.
• 🏅Peak period global resource statistics and various resource TOP10, namespace-level peak period P95 resource usage and resource consumption ratio relative to overall resources
• 🧿Microservice-level peak period overall resource and utilization analysis, microservice and Pod-level resource curve charts (request values, limit values, usage values)

• ♻After enabling admission control: Implements microservice automatic request value management mechanism, supporting unified strong governance page for manual adjustment of microservice limit values and Pod counts.
• ✨Based on admission control mechanism, achieving consistency between actual usage rates and resource request values for K8S microservice resources has very important significance:
  • 🌊K8S scheduler can more precisely schedule Pods to appropriate nodes through real resource request values, avoiding resource fragmentation and achieving node resource balance.
  • ♻K8S auto-scaling also relies on resource request values for judgment, real request values can more accurately trigger scaling operations.
  • 🛡K8S Quality of Service (QoS mechanism) combined with request values, Pods with real request values will be prioritized for retention, ensuring normal operation of critical services.

• 👑Detailed Explanation of Admission Control-Based Governance Capabilities

• ❤️Agent management page: Update and maintain Agent status, configure collection and governance.
• 🔒Based on NGINX basic authentication, supports LDAP, interface-level permission control, and all operation audit logs with notifications.
• 📊All dashboards are created based on Grafana and integrated into the frontend UI, enabling data analysis to quickly achieve more elegant displays.

### 【Download helm package】
wget https://StarsL.cn/kubedoor/kubedoor-1.5.2.tgz
tar -zxvf kubedoor-1.5.2.tgz
cd kubedoor
### 【Master installation】
# Edit values-master.yaml file, please read comments carefully and modify configuration content according to descriptions.
# try
helm upgrade -i kubedoor . --namespace kubedoor --create-namespace --values values-master.yaml --dry-run --debug
# install
helm upgrade -i kubedoor . --namespace kubedoor --create-namespace --values values-master.yaml
### 【Agent installation】
# Edit values-agent.yaml file, please read comments carefully and modify configuration content according to descriptions.
helm upgrade -i kubedoor-agent . --namespace kubedoor --create-namespace --values values-agent.yaml --set tsdb.external_labels_value=xxxxxxxx

# Download installation package
wget https://StarsL.cn/kubedoor/kubedoor-1.5.2.tgz
tar -zxvf kubedoor-1.5.2.tgz

# Updating from older versions to 1.3.0 and above requires adding 2 new database fields
ALTER TABLE kubedoor.k8s_agent_status ADD COLUMN nms_not_confirm Bool DEFAULT false AFTER admission_namespace;
ALTER TABLE kubedoor.k8s_agent_status ADD COLUMN scheduler Bool DEFAULT false AFTER nms_not_confirm;

helm get values kubedoor -n kubedoor
helm get values kubedoor-agent -n kubedoor

Thanks to the following excellent projects, without which KubeDoor would not be possible:

• TRAE Python AIOHTTP VUE Pure Admin Element Plus Kubernetes VictoriaMetrics ClickHouse Grafana Nginx ...

Special Thanks

• CassTime: The birth of KubeDoor is inseparable from the support of 🦄CassTime.