環境管理
このガイドでは、設定の構成、環境固有の設定、およびデプロイメント戦略を使用して、複数の環境（開発、ステージング、プロダクション）間でProbeワークフローを管理する方法について説明します。
基本的な環境設定
目指すのは、1つのワークフローファイルを開発、ステージング、本番のいずれに対しても動かし、変わるのは値だけという状態です。
単一ワークフロー、複数環境
複数の環境で動作するベースワークフローを作成します：
base-workflow.yml:
name: Multi-Environment API Test
description: API testing workflow that adapts to different environments
vars:
api_base_url: "{{API_BASE_URL}}"
environment: "{{ENVIRONMENT ?? 'Unknown'}}"
default_timeout: "{{DEFAULT_TIMEOUT ?? '30s'}}"
jobs:
- id: api-health-check
name: API Health Check
defaults:
http:
headers:
User-Agent: "Probe Test Agent"
Accept: "application/json"
steps:
- name: Health Endpoint Test
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health"
test: res.code == 200
outputs:
api_healthy: res.code == 200
response_time: (rt.sec * 1000)
api_version: res.body.version
- name: Database Health Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/database"
test: res.code == 200
outputs:
database_healthy: res.code == 200
db_response_time: (rt.sec * 1000)
- id: environment-report
name: Environment Report
needs: [api-health-check]
defaults:
http:
headers:
User-Agent: "Probe Test Agent"
Accept: "application/json"
steps:
- name: Environment Summary
uses: hello
echo: |
🌍 Environment Test Report
=========================
Environment: {{vars.environment}}
API Base URL: {{vars.api_base_url}}
Health Check Results:
API Health: {{outputs.api_healthy ? "✅ Healthy" : "❌ Down"}} ({{outputs.response_time}}ms)
Database: {{outputs.database_healthy ? "✅ Healthy" : "❌ Down"}} ({{outputs.db_response_time}}ms)
API Version: {{outputs.api_version}}
Environment-Specific Notes:
{{vars.environment == "development" ? "• Development environment - extended timeouts enabled" : ""}}
{{vars.environment == "staging" ? "• Staging environment - production-like testing" : ""}}
{{vars.environment == "production" ? "• Production environment - strict validation" : ""}}
- id: dev-specific-checks
name: Development Environment Checks
needs: [api-health-check]
skipif: vars.environment != "development"
steps:
- name: Hot Reload Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/dev/hot-reload-status"
test: res.code == 200
outputs:
hot_reload_enabled: res.body.enabled
- name: Debug Endpoints Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/debug/info"
test: res.code == 200
outputs:
debug_info_available: res.code == 200
- name: Development Summary
uses: hello
echo: |
🛠️ Development Environment Status:
Hot Reload: {{outputs.hot_reload_enabled ? "✅ Enabled" : "❌ Disabled"}}
Debug Info: {{outputs.debug_info_available ? "✅ Available" : "❌ Not Available"}}
- id: staging-specific-checks
name: Staging Environment Checks
needs: [api-health-check]
skipif: vars.environment != "staging"
defaults:
http:
headers:
X-Environment: staging
steps:
- name: Load Balancer Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/load-balancer"
test: res.code == 200
outputs:
load_balancer_healthy: res.code == 200
backend_count: res.body.active_backends
- name: Cache Layer Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/cache"
test: res.code == 200
outputs:
cache_healthy: res.code == 200
cache_hit_rate: res.body.hit_rate
- name: Integration Tests
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/test/integration"
test: res.code == 200 && res.body.all_tests_passed == true
outputs:
integration_tests_passed: res.body.all_tests_passed
- name: Staging Summary
uses: hello
echo: |
🧪 Staging Environment Status:
Load Balancer: {{outputs.load_balancer_healthy ? "✅ Healthy" : "❌ Issues"}} ({{outputs.backend_count}} backends)
Cache Layer: {{outputs.cache_healthy ? "✅ Healthy" : "❌ Issues"}} ({{(outputs.cache_hit_rate * 100)}}% hit rate)
Integration Tests: {{outputs.integration_tests_passed ? "✅ Passed" : "❌ Failed"}}
- id: production-specific-checks
name: Production Environment Checks
needs: [api-health-check]
skipif: vars.environment != "production"
defaults:
http:
headers:
X-Environment: production
steps:
- name: SSL Certificate Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/ssl"
test: |
res.code == 200 &&
res.body.certificate_valid == true &&
res.body.days_until_expiry > 30
outputs:
ssl_valid: res.body.certificate_valid
ssl_days_remaining: res.body.days_until_expiry
- name: Performance SLA Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/performance"
test: |
res.code == 200 &&
res.body.avg_response_time < 500 &&
res.body.success_rate > 0.999
outputs:
sla_met: res.body.avg_response_time < 500 && res.body.success_rate > 0.999
avg_response_time: res.body.avg_response_time
success_rate: res.body.success_rate
- name: Security Compliance Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/security"
test: |
res.code == 200 &&
res.body.security_score >= 0.95
outputs:
security_compliant: res.body.security_score >= 0.95
security_score: res.body.security_score
- name: Production Summary
uses: hello
echo: |
🏭 Production Environment Status:
SSL Certificate: {{outputs.ssl_valid ? "✅ Valid" : "❌ Invalid"}} ({{outputs.ssl_days_remaining}} days remaining)
Performance SLA: {{outputs.sla_met ? "✅ Met" : "❌ Violated"}}
- Avg Response: {{outputs.avg_response_time}}ms
- Success Rate: {{(outputs.success_rate * 100)}}%
Security Compliance: {{outputs.security_compliant ? "✅ Compliant" : "❌ Non-Compliant"}} ({{(outputs.security_score * 100)}}%)development.yml:
vars:
environment: "{{ENVIRONMENT ?? 'development'}}"
api_base_url: "{{API_BASE_URL ?? 'http://localhost:3000'}}"
default_timeout: "{{DEFAULT_TIMEOUT ?? '60s'}}"staging.yml:
vars:
environment: "{{ENVIRONMENT ?? 'staging'}}"
api_base_url: "{{API_BASE_URL ?? 'https://api.staging.yourcompany.com'}}"
default_timeout: "{{DEFAULT_TIMEOUT ?? '30s'}}"production.yml:
vars:
environment: "{{ENVIRONMENT ?? 'production'}}"
api_base_url: "{{API_BASE_URL ?? 'https://api.yourcompany.com'}}"
default_timeout: "{{DEFAULT_TIMEOUT ?? '10s'}}"使用方法:
# 開発環境
probe base-workflow.yml,development.yml
# ステージング環境
probe base-workflow.yml,staging.yml
# プロダクション環境
probe base-workflow.yml,production.yml環境ごとのファイルにはvarsだけを書きます。後のファイルのトップレベルのキーは前のファイルのキーを置き換えます。そのため、環境ごとのファイルにjobsを書くとベースワークフローのジョブがすべて置き換わり、needsに書いたジョブもなくなります。前のファイルのjobsを置き換えるファイルがあると、Probeは警告を出します。1つの環境のためのジョブはベースワークフローに置き、skipifでほかの環境では飛ばします。
環境ごとの違いがいくつかの値だけなら、ワークフローは1ファイルのままにして、値を実行時に与えられます。--env-fileはvarsで参照する環境変数を、環境にない場合にファイルから読み、--varはワークフローに書かれた値の代わりにvarを与えます：
# 環境ごとの変数を NAME=value の行で書いたファイルに置く
probe --env-file staging.env workflow.yml
# 1回の実行だけ、1つの値を変える
probe --env-file staging.env --var api_url=http://localhost:8080 workflow.yml詳しくは--varと--env-fileを参照してください。
高度な環境管理
環境ごとに異なるのは接続先だけではありません。有効になっている機能、認証情報の取得元、そして環境自体が正しく設定されているかも異なります。
環境固有の機能フラグ
異なる環境でテストされる機能を制御します。このファイルは独自のjobsを持つ1つのワークフローなので、ベースワークフローにはマージせず、別に実行します：
feature-flags.yml:
name: Feature Flag Tests
# 機能フラグ設定
vars:
api_base_url: "{{API_BASE_URL}}"
admin_token: "{{ADMIN_TOKEN}}"
environment: "{{ENVIRONMENT}}"
# コア機能（常に有効）
feature_user_management: true
feature_basic_api: true
# 環境固有の機能（テンプレートの値は文字列の "true" か "false" になる）
feature_beta_api: "{{vars.environment != 'production'}}"
feature_admin_tools: "{{vars.environment == 'development'}}"
feature_performance_testing: "{{vars.environment != 'development'}}"
feature_security_scanning: "{{vars.environment == 'production'}}"
jobs:
- id: core-feature-tests
name: Core Feature Tests
steps:
- name: User Management Test
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/users"
test: res.code == 200
outputs:
user_management_working: res.code == 200
- name: Basic API Test
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/api/basic"
test: res.code == 200
outputs:
basic_api_working: res.code == 200
- id: beta-feature-tests
name: Beta Feature Tests
skipif: vars.feature_beta_api != "true"
steps:
- name: Beta API Test
id: beta-feature-tests
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/api/beta"
test: res.code == 200
outputs:
beta_api_working: res.code == 200
- id: admin-feature-tests
name: Admin Feature Tests
skipif: vars.feature_admin_tools != "true"
steps:
- name: Admin Tools Test
id: admin-feature-tests
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/admin/tools"
headers:
Authorization: "Bearer {{vars.admin_token}}"
test: res.code == 200
outputs:
admin_tools_working: res.code == 200
- id: performance-tests
name: Performance Tests
skipif: vars.feature_performance_testing != "true"
steps:
- name: Load Test
id: performance-tests
uses: http
with:
url: "{{vars.api_base_url}}/test/load"
method: POST
body: |
{
"concurrent_users": {{vars.environment == "staging" ? 10 : 50}},
"duration_seconds": {{vars.environment == "staging" ? 60 : 300}}
}
test: res.code == 200
outputs:
load_test_passed: res.body.success
- id: security-tests
name: Security Tests
skipif: vars.feature_security_scanning != "true"
steps:
- name: Security Scan
id: security-tests
uses: http
with:
url: "{{vars.api_base_url}}/security/scan"
method: POST
test: res.code == 200 && res.body.vulnerabilities_found == 0
outputs:
security_scan_clean: res.body.vulnerabilities_found == 0
- id: feature-summary
name: Feature Test Summary
needs: [core-feature-tests, beta-feature-tests, admin-feature-tests, performance-tests, security-tests]
steps:
- name: Environment Feature Report
uses: hello
echo: |
🚀 Feature Test Summary for {{vars.environment}}:
================================================
CORE FEATURES:
{{vars.feature_user_management == true ? "User Management: " + (outputs.user_management_working ? "✅ Working" : "❌ Failed") : "User Management: ⏸️ Disabled"}}
{{vars.feature_basic_api == true ? "Basic API: " + (outputs.basic_api_working ? "✅ Working" : "❌ Failed") : "Basic API: ⏸️ Disabled"}}
BETA FEATURES:
{{vars.feature_beta_api == "true" ? "Beta API: " + (outputs['beta-feature-tests'].beta_api_working ? "✅ Working" : "❌ Failed") : "Beta API: ⏸️ Disabled"}}
ADMIN FEATURES:
{{vars.feature_admin_tools == "true" ? "Admin Tools: " + (outputs['admin-feature-tests'].admin_tools_working ? "✅ Working" : "❌ Failed") : "Admin Tools: ⏸️ Disabled"}}
PERFORMANCE TESTING:
{{vars.feature_performance_testing == "true" ? "Load Testing: " + (outputs['performance-tests'].load_test_passed ? "✅ Passed" : "❌ Failed") : "Performance Testing: ⏸️ Disabled"}}
SECURITY TESTING:
{{vars.feature_security_scanning == "true" ? "Security Scan: " + (outputs['security-tests'].security_scan_clean ? "✅ Clean" : "❌ Vulnerabilities Found") : "Security Testing: ⏸️ Disabled"}}
Environment Configuration:
Features enabled: {{
(vars.feature_user_management == true ? 1 : 0) +
(vars.feature_basic_api == true ? 1 : 0) +
(vars.feature_beta_api == "true" ? 1 : 0) +
(vars.feature_admin_tools == "true" ? 1 : 0) +
(vars.feature_performance_testing == "true" ? 1 : 0) +
(vars.feature_security_scanning == "true" ? 1 : 0)
}} / 6認証情報とシークレット管理
認証情報はYAMLに書きません。認証情報だけのYAMLファイルにも分けません。コマンドラインでファイルを1つ増やして渡すと、そのvarsが環境ごとのファイルのvarsを置き換えるためです。ワークフローは認証情報をvarsで環境変数から読み、secretsに名前を並べます。こうすると、その値は出力とレポートで隠されます：
secrets:
- API_TOKEN
vars:
api_token: "{{API_TOKEN}}"開発用のマシンでは、値をNAME=valueの行で書いたファイルに置き、コミットせずに--env-fileで渡せます：
development.env:
ENVIRONMENT=development
API_BASE_URL=http://localhost:3000
API_TOKEN=dev_token_12345
ADMIN_TOKEN=dev_admin_tokenprobe --env-file development.env environment-validation.ymlCIでは、ジョブの環境変数にCIのシークレットから値を渡します。下のGitHub Actionsの例を参照してください。
環境検証ワークフロー
テスト実行前に環境設定を検証します：
environment-validation.yml:
name: Environment Validation
description: Validate environment configuration and prerequisites
vars:
environment: "{{ENVIRONMENT ?? ''}}"
api_base_url: "{{API_BASE_URL ?? ''}}"
api_token: "{{API_TOKEN ?? ''}}"
log_level: "{{LOG_LEVEL ?? ''}}"
default_timeout: "{{DEFAULT_TIMEOUT ?? ''}}"
jobs:
- id: environment-validation
name: Environment Configuration Validation
steps:
- name: Required Environment Variables Check
uses: hello
echo: |
🔍 Environment Variables Validation:
Required Variables:
ENVIRONMENT: {{vars.environment != "" ? "✅ Set (" + vars.environment + ")" : "❌ Missing"}}
API_BASE_URL: {{vars.api_base_url != "" ? "✅ Set (" + vars.api_base_url + ")" : "❌ Missing"}}
API_TOKEN: {{vars.api_token != "" ? "✅ Set (***)" : "❌ Missing"}}
Optional Variables:
LOG_LEVEL: {{vars.log_level != "" ? "✅ Set (" + vars.log_level + ")" : "⚠️ Using default"}}
DEFAULT_TIMEOUT: {{vars.default_timeout != "" ? "✅ Set (" + vars.default_timeout + ")" : "⚠️ Using default"}}
Validation Status: {{
vars.environment != "" && vars.api_base_url != "" && vars.api_token != "" ? "✅ Valid" : "❌ Invalid"
}}
- name: Environment-Specific Validation
uses: hello
echo: |
📋 Environment-Specific Validation:
{{vars.environment == "development" ? "Development Environment:" : ""}}
{{vars.environment == "development" ? "• Extended timeouts enabled" : ""}}
{{vars.environment == "development" ? "• Debug features available" : ""}}
{{vars.environment == "development" ? "• Security checks relaxed" : ""}}
{{vars.environment == "staging" ? "Staging Environment:" : ""}}
{{vars.environment == "staging" ? "• Production-like configuration" : ""}}
{{vars.environment == "staging" ? "• Integration testing enabled" : ""}}
{{vars.environment == "staging" ? "• Performance testing included" : ""}}
{{vars.environment == "production" ? "Production Environment:" : ""}}
{{vars.environment == "production" ? "• Strict timeouts enforced" : ""}}
{{vars.environment == "production" ? "• Security scanning enabled" : ""}}
{{vars.environment == "production" ? "• Full monitoring active" : ""}}
- name: Service Connectivity Pre-Check
uses: http
timeout: 10s
with:
method: GET
url: "{{vars.api_base_url}}/health"
test: res.code == 200
outputs:
connectivity_ok: res.code == 200
api_version: res.body.version
environment_confirmed: res.body.environment
- name: Authentication Pre-Check
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/auth/validate"
headers:
Authorization: "Bearer {{vars.api_token}}"
test: res.code == 200
outputs:
auth_valid: res.code == 200
token_expires_in: res.body.expires_in
- name: Validation Summary
uses: hello
echo: |
✅ Environment Validation Results:
Connectivity: {{outputs.connectivity_ok ? "✅ Connected" : "❌ Failed"}}
API Version: {{outputs.api_version}}
Environment Match: {{outputs.environment_confirmed == vars.environment ? "✅ Confirmed" : "⚠️ Mismatch"}}
Authentication: {{outputs.auth_valid ? "✅ Valid" : "❌ Invalid"}}
{{outputs.auth_valid ? "Token Expires In: " + string(outputs.token_expires_in) + " seconds" : ""}}
Environment Ready: {{
outputs.connectivity_ok &&
outputs.auth_valid &&
outputs.environment_confirmed == vars.environment
? "🟢 YES" : "🔴 NO"
}}CI/CD統合
どの環境に対する実行かは、CI側の環境変数にすでに入っています。その情報を渡せば、同じワークフローで値とテストスイートを切り替えられます。
GitHub Actions統合
自動環境テスト用にCI/CDパイプラインと統合します：
.github/workflows/probe-tests.yml:
name: Probe Environment Tests
on:
push:
branches: [main, develop]
pull_request:
branches: [main]
jobs:
development-tests:
name: Development Environment Tests
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- name: Install Probe
run: |
curl -L https://github.com/mozership/probe/releases/latest/download/probe_linux_x86_64.tar.gz | tar -xz probe
chmod +x probe
sudo mv probe /usr/local/bin/
- name: Run Development Tests
run: |
probe workflows/base-workflow.yml,environments/development.yml
staging-tests:
name: Staging Environment Tests
runs-on: ubuntu-latest
if: github.event_name == 'push' && github.ref == 'refs/heads/main'
steps:
- uses: actions/checkout@v3
- name: Install Probe
run: |
curl -L https://github.com/mozership/probe/releases/latest/download/probe_linux_x86_64.tar.gz | tar -xz probe
chmod +x probe
sudo mv probe /usr/local/bin/
- name: Validate Staging Environment
env:
ENVIRONMENT: staging
API_BASE_URL: https://api.staging.yourcompany.com
API_TOKEN: ${{ secrets.STAGING_API_TOKEN }}
run: |
probe workflows/environment-validation.yml
- name: Run Staging Tests
run: |
probe workflows/base-workflow.yml,environments/staging.yml
- name: Run Feature Tests
env:
API_BASE_URL: https://api.staging.yourcompany.com
ADMIN_TOKEN: ${{ secrets.STAGING_ADMIN_TOKEN }}
run: |
probe --var environment=staging features/feature-flags.yml
production-smoke-tests:
name: Production Smoke Tests
runs-on: ubuntu-latest
if: github.event_name == 'push' && github.ref == 'refs/heads/main'
steps:
- uses: actions/checkout@v3
- name: Install Probe
run: |
curl -L https://github.com/mozership/probe/releases/latest/download/probe_linux_x86_64.tar.gz | tar -xz probe
chmod +x probe
sudo mv probe /usr/local/bin/
- name: Production Smoke Tests
env:
API_BASE_URL: https://api.yourcompany.com
run: |
probe workflows/smoke-test.yml環境固有のテストスイート
異なる環境に対応する様々なテストスイートを作成します：
smoke-test.yml（プロダクション用）:
name: Production Smoke Test
description: Minimal smoke test for production environment
vars:
api_base_url: "{{API_BASE_URL}}"
jobs:
- id: critical-endpoints
name: Critical Endpoints Check
steps:
- name: Health Check
id: api
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health"
test: res.code == 200
outputs:
api_healthy: res.code == 200
- name: Authentication Check
id: auth
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/auth/health"
test: res.code == 200
outputs:
auth_healthy: res.code == 200
- name: Database Check
id: database
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/database"
test: res.code == 200
outputs:
db_healthy: res.code == 200
- id: smoke-test-summary
name: Smoke Test Summary
needs: [critical-endpoints]
steps:
- name: Production Health Summary
uses: hello
echo: |
🏭 Production Smoke Test Results:
Critical Systems:
API: {{outputs.api.api_healthy ? "✅ Healthy" : "🚨 DOWN"}}
Authentication: {{outputs.auth.auth_healthy ? "✅ Healthy" : "🚨 DOWN"}}
Database: {{outputs.database.db_healthy ? "✅ Healthy" : "🚨 DOWN"}}
Overall Status: {{
outputs.api.api_healthy &&
outputs.auth.auth_healthy &&
outputs.database.db_healthy
? "🟢 ALL SYSTEMS OPERATIONAL" : "🔴 CRITICAL ISSUES DETECTED"
}}comprehensive-test.yml（ステージング用）:
name: Comprehensive Staging Test
description: Full test suite for staging environment validation
vars:
api_base_url: "{{API_BASE_URL}}"
jobs:
- id: api-tests
name: API Test Suite
steps:
- name: User Management API
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/users"
test: res.code == 200
- name: Order Management API
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/orders"
test: res.code == 200
- name: Product Catalog API
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/products"
test: res.code == 200
- id: integration-tests
name: Integration Tests
needs: [api-tests]
steps:
- name: User-Order Integration
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/test/user-order-flow"
test: res.code == 200 && res.body.test_passed == true
- name: Payment Integration
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/test/payment-flow"
test: res.code == 200 && res.body.test_passed == true
- id: performance-tests
name: Performance Validation
needs: [integration-tests]
steps:
- name: Load Test
uses: http
with:
url: "{{vars.api_base_url}}/test/load"
method: POST
body: |
{
"concurrent_users": 10,
"duration_seconds": 60
}
test: res.code == 200 && res.body.success_rate > 0.95環境モニタリングとアラート
テストで使う環境設定はそのまま監視にも使えます。これにより、各環境をその環境の基準で確認できます。
環境ヘルスモニタリング
各環境のヘルスを継続的に監視します：
environment-monitor.yml:
name: Environment Health Monitor
description: Continuous monitoring of environment health
vars:
api_base_url: "{{API_BASE_URL}}"
environment: "{{ENVIRONMENT}}"
smtp_host: "{{SMTP_HOST}}"
monitoring_interval: "{{MONITORING_INTERVAL ?? '300'}}" # 5分
jobs:
- id: environment-health-check
name: Environment Health Check
steps:
- name: System Resources Check
id: resources
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/resources"
test: |
res.code == 200 &&
res.body.cpu_usage < 80 &&
res.body.memory_usage < 80 &&
res.body.disk_usage < 90
outputs:
resources_healthy: |
res.code == 200 &&
res.body.cpu_usage < 80 &&
res.body.memory_usage < 80 &&
res.body.disk_usage < 90
cpu_usage: res.body.cpu_usage
memory_usage: res.body.memory_usage
disk_usage: res.body.disk_usage
- name: Service Dependencies Check
id: dependencies
uses: http
with:
method: GET
url: "{{vars.api_base_url}}/health/dependencies"
test: |
res.code == 200 &&
res.body.all_dependencies_healthy == true
outputs:
dependencies_healthy: res.body.all_dependencies_healthy
unhealthy_services: res.body.unhealthy_services
- name: Environment-Specific Checks
id: env-checks
uses: hello
echo: |
Environment-specific validation for {{vars.environment}}
outputs:
env_specific_checks: |
vars.environment == "production" ? "SSL, Security, Performance" :
vars.environment == "staging" ? "Integration, Load Testing" :
"Development Tools, Debug Features"
- id: alerting
name: Environment Alerting
needs: [environment-health-check]
always: true
skipif: outputs.resources?.resources_healthy == true && outputs.dependencies?.dependencies_healthy == true
steps:
- name: Environment Alert
uses: smtp
with:
addr: "{{vars.smtp_host}}:587"
from: "environment-alerts@yourcompany.com"
to: "devops@yourcompany.com"
subject: "🚨 Environment Health Alert - {{vars.environment}}"
session: 1
message: 1
length: 500
test: res.code == 0
echo: |
ENVIRONMENT HEALTH ALERT
========================
Environment: {{vars.environment}}
Time: {{unixtime()}}
RESOURCE STATUS:
{{outputs.resources?.resources_healthy == true ? "✅ Resources Healthy" : "❌ Resource Issues"}}
CPU Usage: {{outputs.resources?.cpu_usage ?? "unknown"}}%
Memory Usage: {{outputs.resources?.memory_usage ?? "unknown"}}%
Disk Usage: {{outputs.resources?.disk_usage ?? "unknown"}}%
DEPENDENCIES STATUS:
{{outputs.dependencies?.dependencies_healthy == true ? "✅ All Dependencies Healthy" : "❌ Dependency Issues"}}
Unhealthy Services: {{join(outputs.dependencies?.unhealthy_services ?? [], ", ")}}
ACTION REQUIRED: Investigate {{vars.environment}} environment immediately
- id: health-summary
name: Health Summary
needs: [environment-health-check]
steps:
- name: All Systems Healthy
uses: hello
echo: |
✅ {{vars.environment}} Environment Health Check
All systems operational:
• Resources: CPU {{outputs.resources.cpu_usage}}%, Memory {{outputs.resources.memory_usage}}%, Disk {{outputs.resources.disk_usage}}%
• Dependencies: All healthy
• Environment: {{vars.environment}}
Next check in {{vars.monitoring_interval}} secondsベストプラクティス
以下では、環境を分離すること、変更を環境の順に確認していくこと、設定を検証すること、秘密情報を扱うことを扱います。
1. 環境分離
環境ごとにファイルを分ければ値が混ざらず、差分として違いを確認できます。
# 良い例: 明確な環境分離
environments/
├── development.yml
├── staging.yml
└── production.yml
# 良い例: 認証情報はYAMLに書かず環境変数で渡す
development.env（コミットしない）
CIのシークレット → ジョブの環境変数2. 段階的テスト
各環境にはそこで意味のある確認を置き、重い確認は本番より前の環境で実行します。
# 良い例: テストパイプラインの進行
Development → Unit Tests
Staging → Integration Tests + Performance Tests
Production → Smoke Tests + Monitoring3. 設定検証
設定が不完全な環境に対して実行すると、アプリケーションの障害に見える失敗が出ます。そのため先に設定を確認します。
# 良い例: テスト実行前の検証
vars:
environment: "{{ENVIRONMENT ?? ''}}"
api_base_url: "{{API_BASE_URL ?? ''}}"
api_token: "{{API_TOKEN ?? ''}}"- name: Pre-Test Validation
uses: hello
echo: |
Environment: {{vars.environment != "" ? "✅" : "❌"}}
API URL: {{vars.api_base_url != "" ? "✅" : "❌"}}
Credentials: {{vars.api_token != "" ? "✅" : "❌"}}4. セキュアなシークレット管理
秘密情報は環境変数から読みます。そうすれば環境設定のファイルはコミットでき、値そのものは残りません。
# 良い例: シークレットに環境変数を使用
vars:
api_token: "{{PROD_API_TOKEN}}" # セキュアな環境変数から
# 避ける: ハードコードされたシークレット
vars:
api_token: "hardcoded_secret_123" # これは絶対にしないあわせてsecretsにも変数を列挙します。そうすると、CIのログやアップロードしたアーティファクトに残る端末の出力、アクションのログ、レポートファイルで、値が<secret:NAME>に置き換わります。
secrets:
- PROD_API_TOKEN
vars:
api_token: "{{PROD_API_TOKEN}}"次のステップ
効果的に環境を管理できるようになったので、次を探索してください：
- モニタリングワークフロー - 全環境の包括的モニタリングを構築
- APIテスト - 環境固有のAPIテストスイートを作成
- エラーハンドリング戦略 - 環境固有の障害を処理
環境管理は、ソフトウェア開発ライフサイクル全体にわたる信頼性の高いテストにとって重要です。これらのパターンを使用して、一貫性があり、セキュアで、保守しやすい環境設定を作成しましょう。
更新日時