aks-troubleshooting

github:microsoft/azure-skills · first seen 2026-10-30T20:36:17 · installs n/a · source · badge

Debug live Azure Kubernetes Service (AKS) incidents with a read-only, evidence-first investigation. WHEN: pod crashes or Pending, CrashLoopBackOff, OOMKilled, ImagePullBackOff, node NotReady, DNS or ingress failure, connectivity timeout, network policy, SNAT exhaustion, node-pool scaling blocked by QuotaExceeded or InsufficientVCPUQuota, upgrade stuck, spot or zone disruption, a bare VMExtensionPr

Works with

aks-gpu-inference
Shares aks, incidents, gpu, pod0.24

aks-network-capture
Shares aks, ingress, connectivity, evidence0.23

Woven into by

aks-gpu-inference
Shares aks, incidents, gpu, pod0.24

aks-network-capture
Shares aks, ingress, connectivity, evidence0.23

airunway-aks-setup
Shares aks, gpu, bare, incidents0.22

azure-kubernetes
Shares aks, pod, kubernetes, upgrade0.22

azure-diagnostics
Shares debug, evidence, applens, aks0.21

aks-cost-optimization
Shares aks, spot, incidents, scaling0.2

aks-known-issues
Shares aks, zone, troubleshooting, bare0.17

investigate-first
Shares evidence, causes, investigate, hypotheses0.13

Skill Health 80 estimated by backskills

Freshness 100 · Quality 60 · Efficacy not tested

Efficacy: not tested yet.

Version history

2026-10-30T20:36:17 · sha256 fdf30800f2ce