Amazon EKS Alerting Areas

Amazon EKS Alerting Areas An architecture diagram generated by Archify. API Server · kube-apiserver · Control Plane API Server kube-apiserver etcd · cluster state · Control Plane etcd cluster state Scheduler / Ctrl Mgr · scheduling + controllers · Control Plane Scheduler / Ctrl Mgr scheduling + controllers Node Status · worker nodes · Data Plane Node Status worker nodes Pod Status · phase / restarts · Data Plane Pod Status phase / restarts Container Status · restarts / OOM · Data Plane Container Status restarts / OOM VPC CNI · pod networking · Networking VPC CNI pod networking Service / Ingress · traffic entry · Networking Service / Ingress traffic entry CoreDNS · cluster DNS · Networking CoreDNS cluster DNS EBS CSI / EFS CSI · storage drivers · Storage EBS CSI / EFS CSI storage drivers PV / PVC · volume binding · Storage PV / PVC volume binding Control Plane Data Plane Networking Storage Legend Backend Database Cloud

Cluster

  • • up=0: scrape failure, not proof of API outage
  • • Separate missing targets and recent counter increases

Workloads

  • • CrashLoopBackOff, frequent restarts and readiness differ
  • • Exclude completed Pods; compare desired > available

Collection limits

  • • Verify exporter, CSI and managed-service metric availability
  • • Boxes show monitoring scope, not direct traffic paths