Auto-sync: update nexus workspace
This commit is contained in:
29
wiki/concepts/Container-Insights.md
Normal file
29
wiki/concepts/Container-Insights.md
Normal file
@@ -0,0 +1,29 @@
|
||||
---
|
||||
title: "Container Insights"
|
||||
type: concept
|
||||
tags: [AWS, EKS, monitoring, metrics, CloudWatch, containers]
|
||||
last_updated: 2026-04-28
|
||||
---
|
||||
|
||||
## Definition
|
||||
|
||||
Container Insights 是 Amazon CloudWatch 的容器监控功能,专门用于收集、整理和汇总 EKS 和 ECS 集群中容器化工作负载的指标和日志。它通过 CloudWatch Agent(DaemonSet)和 CloudWatch Embedded Metric Format 自动收集容器级指标,使运维团队无需额外配置即可获得集群运行状况的可见性。
|
||||
|
||||
## Key Mechanisms
|
||||
|
||||
- **自动指标收集**:容器 CPU/内存/网络/磁盘使用率,Pod 级别聚合
|
||||
- **日志收集**:容器标准输出日志自动采集至 CloudWatch Logs
|
||||
- **预构建仪表板**:提供集群、节点、Pod、服务级别的可视化仪表板
|
||||
- **性能告警**:自动发现高负载资源并生成 CloudWatch 告警
|
||||
- **Embedded Metric Format**:应用可直接输出结构化指标,无需额外 SDK
|
||||
|
||||
## Relationship with CloudWatch Agent
|
||||
|
||||
Container Insights 建立在 CloudWatch Agent 之上:
|
||||
- CloudWatch Agent(DaemonSet)是数据采集代理
|
||||
- Container Insights 是指标组织和聚合逻辑(通过 CloudWatch Agent 配置实现)
|
||||
- 用户启用 Container Insights 后,CloudWatch Agent 自动开始发布容器指标
|
||||
|
||||
## Sources
|
||||
- [[ctp-topic-70-eks-deployment-using-iac]]
|
||||
- [[ctp-topic-42-grafana-observability-dashboard]]
|
||||
Reference in New Issue
Block a user