run-chaos-experiment
について
このスキルは、LitmusまたはChaos Meshを使用して制御された障害注入を通じてシステムの回復力をテストするカオスエンジニアリング実験を実行します。仮説駆動型テストのために設計されており、障害回復の改善を目的としています。ローンチ前検証、アーキテクチャ変更後のテスト、GameDaysなどのシナリオで使用されます。このスキルはKubernetesクラスターを必要とし、SRE成熟度の一環として障害モードの仮定を検証するのに役立ちます。
クイックインストール
Claude Code
推奨npx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/run-chaos-experimentこのコマンドをClaude Codeにコピー&ペーストしてスキルをインストールします
ドキュメント
Run Chaos Experiment
Inject controlled failures → test+improve resilience.
Use When
- Pre-major launch (load test)
- Post-arch change (validate resilience)
- GameDays|DR drills
- Validate failure mode assumptions
- SRE maturity program
In
- Required: K8s cluster (Litmus|Chaos Mesh)
- Required: Steady-state def ("normal")
- Required: Hypothesis ("API up if 1 pod crashes")
- Optional: Observability (Prometheus, Grafana) → measure impact
- Optional: Rollback plan
Do
Step 1: Define Steady State + Hypothesis
## Steady State Definition
### Service: API Gateway
- **Availability**: 99.9% (< 0.1% error rate)
- **Latency**: p95 < 200ms
- **Throughput**: 1000 req/s
- **Dependencies**: Database (Postgres), Cache (Redis), Auth Service
### Metrics
- `rate(http_requests_total{job="api"}[5m])`
- `histogram_quantile(0.95, rate(http_request_duration_seconds_bucket[5m]))`
- `rate(http_requests_total{status=~"5.."}[5m])`
## Hypothesis
**"If one API pod is killed, the remaining pods will handle the load with <5s
disruption and no increase in error rate."**
### Validation Criteria
- Error rate remains <1%
- p95 latency stays <300ms (50ms grace)
- Service recovers within 5 seconds
- No cascading failures to downstream services
→ Clear, measurable normal + success criteria.
If err: can't define steady state → observability insufficient. Add metrics first.
Step 2: Blast Radius Limits
Scope → minimize risk:
# chaos-config.yaml
apiVersion: v1
kind: Namespace
metadata:
name: chaos-testing
---
# Label pods participating in chaos experiments
apiVersion: v1
kind: Pod
metadata:
labels:
chaos-enabled: "true"
environment: staging # NEVER production for first run
Safeguards:
## Blast Radius Controls
### Environment
- **Scope**: Staging only (first 5 runs)
- **Production**: Only after 5 successful staging runs
- **Timing**: Business hours (09:00-17:00 local), never weekends/holidays
### Target Selection
- **Limit**: Max 1 pod per service
- **Percentage**: Max 25% of replicas
- **Exclusions**: Database, payment service, auth service (critical path)
### Auto-Abort Conditions
- Error rate >10% for >30 seconds
- Customer-facing alerts fire
- Manual abort signal from on-call engineer
### Rollback Plan
- Kubernetes will auto-restart killed pods
- Manual rollback: `kubectl rollout undo deployment/api`
- Incident declared if recovery takes >5 minutes
→ Clear bounds, won't take down whole sys.
If err: blast too large → narrow. Start non-critical service.
Step 3: Install Chaos Mesh
# Add Chaos Mesh Helm repo
helm repo add chaos-mesh https://charts.chaos-mesh.org
helm repo update
# Install Chaos Mesh in isolated namespace
helm install chaos-mesh chaos-mesh/chaos-mesh \
--namespace chaos-mesh \
--create-namespace \
--set dashboard.create=true \
--set controllerManager.replicaCount=1
# Verify installation
kubectl get pods -n chaos-mesh
# Access dashboard
kubectl port-forward -n chaos-mesh svc/chaos-dashboard 2333:2333
# Open http://localhost:2333
Alt: Litmus (vendor-neutral):
# Install Litmus
kubectl apply -f https://litmuschaos.github.io/litmus/litmus-operator-v2.14.0.yaml
# Wait for Litmus pods
kubectl get pods -n litmus
# Install Litmus CRDs
kubectl apply -f https://hub.litmuschaos.io/api/chaos/master?file=charts/generic/experiments.yaml
→ Chaos Mesh|Litmus running, dashboard accessible.
If err: check RBAC. Tools need cluster-wide access.
Step 4: Create+Exec Experiment
Pod Kill (Chaos Mesh):
# pod-kill-experiment.yaml
apiVersion: chaos-mesh.org/v1alpha1
kind: PodChaos
metadata:
name: api-pod-kill-test
namespace: chaos-testing
spec:
action: pod-kill
mode: one # Kill one pod only
selector:
namespaces:
- production
labelSelectors:
app: api-gateway
chaos-enabled: "true"
duration: "30s"
scheduler:
cron: "@every 5m" # Repeat every 5 minutes (for sustained testing)
Apply:
# Apply experiment
kubectl apply -f pod-kill-experiment.yaml
# Watch experiment status
kubectl get podchaos -n chaos-testing -w
# View detailed status
kubectl describe podchaos api-pod-kill-test -n chaos-testing
# Check which pods were affected
kubectl get events -n production --sort-by=.metadata.creationTimestamp | grep api-gateway
Monitor Grafana:
# Error rate during experiment
rate(http_requests_total{status=~"5..", job="api"}[1m])
# Latency spike
histogram_quantile(0.95, rate(http_request_duration_seconds_bucket{job="api"}[1m]))
# Pod restarts
rate(kube_pod_container_status_restarts_total{pod=~"api-.*"}[5m])
→ Pod killed, K8s restarts, service continues w/ minor blip.
If err: err spike|service degrades → abort + investigate.
Step 5: Analyze + Iterate
# Chaos Experiment Report: API Pod Kill
**Date**: 2025-02-09
**Hypothesis**: API stays available if one pod crashes
**Tool**: Chaos Mesh
**Environment**: Staging
**Duration**: 30 seconds (pod kill + recovery)
## Results
### Metrics During Experiment
- **Error Rate**: Increased from 0.1% to 2.3% (spike lasted 8 seconds)
- **p95 Latency**: Increased from 180ms to 450ms (spike lasted 12 seconds)
- **Recovery Time**: 8 seconds (pod restart + load balancer update)
### Hypothesis Outcome
**FAILED**: Error rate exceeded 1% threshold, latency spike >300ms
## Root Cause Analysis
- Load balancer continued routing to killed pod for 8 seconds (stale endpoint)
- Readiness probe set to 10s interval (too slow)
- No pre-stop hook to drain connections gracefully
## Improvements Made
1. **Reduced readiness probe interval**: 10s → 2s
2. **Added pre-stop hook**: 5-second sleep for connection draining
3. **Tuned load balancer**: Enabled faster endpoint updates
## Follow-Up Experiment
- Re-run with same parameters in 1 week
- Expected: Error rate <1%, recovery <5s
Track in log:
# chaos-experiment-log.csv
date,experiment,environment,status,error_rate_peak,recovery_time_s,outcome
2025-02-09,pod-kill-api,staging,complete,2.3%,8,failed
2025-02-16,pod-kill-api,staging,complete,0.8%,4,passed
2025-02-23,network-delay-db,staging,aborted,15%,N/A,failed
→ Learnings captured, fixes implemented, follow-up scheduled.
If err: no action post-exp = chaos theater. Prioritize fixes.
Step 6: Graduate to Prod (Carefully)
After consistent staging passes:
# Production pod-kill experiment (more conservative)
apiVersion: chaos-mesh.org/v1alpha1
kind: PodChaos
metadata:
name: api-pod-kill-prod
namespace: chaos-testing
spec:
action: pod-kill
mode: one
selector:
namespaces:
- production
labelSelectors:
app: api-gateway
chaos-enabled: "true"
duration: "10s" # Shorter than staging
scheduler:
cron: "0 10 * * 2" # Tuesdays at 10 AM only (predictable, low-risk time)
Prod safeguards:
# Create a kill switch for production chaos
kubectl create configmap chaos-killswitch \
-n chaos-testing \
--from-literal=enabled=true
# Update experiments to check kill switch
# (implementation depends on chaos tool)
→ Prod runs in low-risk windows w/ kill switch ready.
If err: prod exp causes incident → disable immediately + post-mortem.
Check
- Steady state + hypothesis defined
- Blast radius limited (env, scope, timing)
- Tool installed + tested
- Exp runs in staging
- Results documented w/ metrics + analysis
- Improvements implemented
- Follow-up validates fixes
- Prod only after 5+ staging successes
Traps
- No hypothesis: "See what happens" wastes time. Always have one.
- Too broad scope: Kill all pods = DR test, not resilience. Start small.
- Prod-first: Never first run in prod. Staging first, always.
- Ignore results: Chaos w/o action = theater. Fix what you learn.
- Alert fatigue: Exps trigger alerts. Annotate Grafana|silence expected.
- No abort plan: Need kill switch ready.
→
setup-prometheus-monitoring— metrics to measure exp impactconfigure-alerting-rules— alerts during chaos (expected)define-slo-sli-sla— steady state tied to SLOs
GitHub リポジトリ
関連スキル
evaluating-llms-harness
テストこのClaudeスキルは、lm-evaluation-harnessを実行し、MMLUやGSM8Kなど60以上の標準化学術タスクでLLMをベンチマークします。開発者がモデルの品質を比較し、トレーニングの進捗を追跡し、学術的な結果を報告するために設計されています。このツールはHuggingFaceやvLLMモデルを含む様々なバックエンドをサポートしています。
cloudflare-cron-triggers
テストこのスキルは、cron式を使用してWorkersをスケジュールするためのCloudflare Cron Triggersの実装に関する包括的な知識を提供します。定期的なタスクの設定、メンテナンスジョブ、自動化されたワークフローの構築を網羅し、無効なcron式やタイムゾーン問題といった一般的な課題への対処法も含みます。開発者はこれを使用して、スケジュールされたハンドラーの設定、cronトリガーのテスト、WorkflowsやGreen Computeとの連携を構成できます。
webapp-testing
テストこのClaude Skillは、Playwrightベースのツールキットを提供し、Pythonスクリプトを通じてローカルWebアプリケーションのテストを可能にします。フロントエンドの検証、UIデバッグ、スクリーンショット撮影、ログ表示を実現し、サーバーライフサイクルを管理します。ブラウザ自動化タスクにご利用いただけますが、コンテキストの汚染を避けるため、スクリプトのソースコードを読むのではなく直接実行してください。
finishing-a-development-branch
テストこのスキルは、開発者がテストの合格を確認し、構造化された統合オプションを提示することで、完成した作業を仕上げることを支援します。実装が完了した後のマージ、PR作成、ブランチの整理といったワークフローを案内します。コードが準備できてテスト済みの際に使用し、開発プロセスを体系的に完了させましょう。
