Conduct periodic access reviews and certifications. Implement access governance and recertification workflows. Use when managing access compliance.
日本語の概要は準備中です。原文の説明を表示しています。
Implement Istio and Linkerd service meshes. Configure mTLS, traffic management, and observability. Use when managing microservices communication.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Implement service-to-service communication management with mTLS, traffic shaping, observability, and policy enforcement using Istio or Linkerd.
istioctl CLI installed.linkerd CLI installed.# Download istioctl
curl -L https://istio.io/downloadIstio | sh -
cd istio-*
export PATH=$PWD/bin:$PATH
# Install with the production profile
istioctl install --set profile=default -y
# Or use the demo profile (includes all addons, good for learning)
istioctl install --set profile=demo -y
# Verify installation
istioctl verify-install
# Check running components
kubectl get pods -n istio-system
# Enable automatic sidecar injection for a namespace
kubectl label namespace default istio-injection=enabled
# Verify label
kubectl get namespace default --show-labels
# Restart existing pods to inject sidecars
kubectl rollout restart deployment -n default
# Check sidecar status
kubectl get pods -n default -o jsonpath='{range .items[*]}{.metadata.name}{" containers: "}{range .spec.containers[*]}{.name}{" "}{end}{"\n"}{end}'
# Install Kiali, Prometheus, Grafana, Jaeger
kubectl apply -f samples/addons/prometheus.yaml
kubectl apply -f samples/addons/grafana.yaml
kubectl apply -f samples/addons/jaeger.yaml
kubectl apply -f samples/addons/kiali.yaml
# Wait for rollout
kubectl rollout status deployment/kiali -n istio-system
# Access dashboards
istioctl dashboard kiali
istioctl dashboard grafana
istioctl dashboard jaeger
# virtualservice.yaml — canary deployment with traffic split
apiVersion: networking.istio.io/v1beta1
kind: VirtualService
metadata:
name: my-app
namespace: default
spec:
hosts:
- my-app
http:
# Header-based routing (canary testers)
- match:
- headers:
x-canary:
exact: "true"
route:
- destination:
host: my-app
subset: canary
# Percentage-based traffic split
- route:
- destination:
host: my-app
subset: stable
weight: 90
- destination:
host: my-app
subset: canary
weight: 10
timeout: 30s
retries:
attempts: 3
perTryTimeout: 10s
retryOn: gateway-error,connect-failure,refused-stream
# destinationrule.yaml
apiVersion: networking.istio.io/v1beta1
kind: DestinationRule
metadata:
name: my-app
namespace: default
spec:
host: my-app
trafficPolicy:
connectionPool:
tcp:
maxConnections: 100
http:
h2UpgradePolicy: DEFAULT
http1MaxPendingRequests: 100
http2MaxRequests: 1000
maxRequestsPerConnection: 10
outlierDetection:
consecutive5xxErrors: 5
interval: 10s
baseEjectionTime: 30s
maxEjectionPercent: 50
subsets:
- name: stable
labels:
version: v1
- name: canary
labels:
version: v2
# gateway.yaml — expose service to external traffic
apiVersion: networking.istio.io/v1beta1
kind: Gateway
metadata:
name: app-gateway
namespace: default
spec:
selector:
istio: ingressgateway
servers:
- port:
number: 443
name: https
protocol: HTTPS
tls:
mode: SIMPLE
credentialName: app-tls-cert # Kubernetes secret
hosts:
- app.example.com
- port:
number: 80
name: http
protocol: HTTP
hosts:
- app.example.com
tls:
httpsRedirect: true
---
apiVersion: networking.istio.io/v1beta1
kind: VirtualService
metadata:
name: app-external
namespace: default
spec:
hosts:
- app.example.com
gateways:
- app-gateway
http:
- route:
- destination:
host: my-app
port:
number: 8080
# peer-authentication.yaml
apiVersion: security.istio.io/v1beta1
kind: PeerAuthentication
metadata:
name: default
namespace: istio-system # Applies to entire mesh
spec:
mtls:
mode: STRICT
# Allow both plaintext and mTLS during migration
apiVersion: security.istio.io/v1beta1
kind: PeerAuthentication
metadata:
name: default
namespace: legacy-apps
spec:
mtls:
mode: PERMISSIVE
# Check mTLS status for a namespace
istioctl x describe pod <pod-name> -n default
# View TLS configuration
istioctl proxy-config cluster <pod-name>.default --fqdn my-app.default.svc.cluster.local -o json | grep -A5 "tlsContext"
# Verify with istioctl authn
istioctl authn tls-check <pod-name>.default my-app.default.svc.cluster.local
# authz-policy.yaml — only allow frontend to call API
apiVersion: security.istio.io/v1beta1
kind: AuthorizationPolicy
metadata:
name: api-access
namespace: default
spec:
selector:
matchLabels:
app: my-api
action: ALLOW
rules:
- from:
- source:
principals:
- "cluster.local/ns/default/sa/frontend"
to:
- operation:
methods: ["GET", "POST"]
paths: ["/api/*"]
---
# Deny all other traffic to api
apiVersion: security.istio.io/v1beta1
kind: AuthorizationPolicy
metadata:
name: deny-all
namespace: default
spec:
selector:
matchLabels:
app: my-api
action: DENY
rules:
- from:
- source:
notPrincipals:
- "cluster.local/ns/default/sa/frontend"
# circuit-breaker.yaml
apiVersion: networking.istio.io/v1beta1
kind: DestinationRule
metadata:
name: my-api-circuit-breaker
spec:
host: my-api
trafficPolicy:
connectionPool:
tcp:
maxConnections: 50
http:
http1MaxPendingRequests: 50
http2MaxRequests: 100
maxRetries: 3
outlierDetection:
consecutive5xxErrors: 3
interval: 15s
baseEjectionTime: 60s
maxEjectionPercent: 100
# Install Linkerd CLI
curl --proto '=https' --tlsv1.2 -sSfL https://run.linkerd.io/install | sh
export PATH=$HOME/.linkerd2/bin:$PATH
# Validate cluster prerequisites
linkerd check --pre
# Install Linkerd CRDs
linkerd install --crds | kubectl apply -f -
# Install Linkerd control plane
linkerd install | kubectl apply -f -
# Verify installation
linkerd check
# Inject sidecar into a namespace
kubectl get deploy -n my-app -o yaml | linkerd inject - | kubectl apply -f -
# Or annotate namespace for auto-injection
kubectl annotate namespace my-app linkerd.io/inject=enabled
# View live traffic dashboard
linkerd viz install | kubectl apply -f -
linkerd viz dashboard
# traffic-split.yaml
apiVersion: split.smi-spec.io/v1alpha4
kind: TrafficSplit
metadata:
name: my-app-split
namespace: default
spec:
service: my-app
backends:
- service: my-app-stable
weight: 900
- service: my-app-canary
weight: 100
# Istio: check proxy configuration
istioctl proxy-config routes <pod-name>.default
istioctl proxy-config clusters <pod-name>.default
istioctl proxy-config listeners <pod-name>.default
# Istio: analyze configuration for issues
istioctl analyze -n default
# Istio: proxy debug logs
istioctl proxy-config log <pod-name>.default --level debug
# Linkerd: check proxy stats
linkerd viz stat deploy -n default
linkerd viz top deploy/my-app -n default
linkerd viz edges deploy -n default
| Symptom | Cause | Fix |
|---|---|---|
| Sidecar not injected | Missing namespace label | Add istio-injection=enabled label; restart pods |
| 503 errors between services | mTLS mismatch (one side plaintext) | Set PeerAuthentication to PERMISSIVE during migration |
| High latency after mesh install | Sidecar resource limits too low | Increase sidecar CPU/memory limits in mesh config |
| VirtualService not routing | Missing DestinationRule subsets | Create matching DestinationRule with subset labels |
upstream connect error | Circuit breaker tripped | Check outlier detection settings; increase thresholds |
| Authorization policy blocks everything | Default deny without matching allow rule | Add explicit ALLOW rule before DENY-all |
| Kiali shows "Unknown" traffic | Missing sidecar on calling service | Inject sidecar into all communicating services |
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Conduct periodic access reviews and certifications. Implement access governance and recertification workflows. Use when managing access compliance.
日本語の概要は準備中です。原文の説明を表示しています。
Build automated evaluation suites for AI agents using golden datasets, rubrics, and regression gates. Use when shipping agent features, validating prompt changes, or gating deployments on quality.
日本語の概要は準備中です。原文の説明を表示しています。
Instrument AI agents with tracing, token metrics, latency, and cost visibility. Use for reliability and debugging.
日本語の概要は準備中です。原文の説明を表示しています。
Secure AI agents against prompt injection, tool abuse, and data exfiltration with defense-in-depth controls. Use when building, deploying, or hardening agentic AI systems that invoke tools, access data, or interact with production infrastructure.
日本語の概要は準備中です。原文の説明を表示しています。
Secure AI coding agents (Claude Code, Cursor, Codex, Copilot) with permission boundaries, secret protection, code review gates, and safe sandbox configurations for team environments.
日本語の概要は準備中です。原文の説明を表示しています。
Use service mesh patterns for AI inference traffic management, mTLS, canary releases, policy enforcement, and cross-cluster resilience.
日本語の概要は準備中です。原文の説明を表示しています。