新增生产HTTPS与证书续期告警方案

2026-02-28 16:34:36 +08:00
parent c7974e7f40
commit 78f488fbbb
5 changed files with 199 additions and 0 deletions
@@ -0,0 +1,49 @@
+upstream wx_service_upstream {
+    server 127.0.0.1:8080;
+    keepalive 32;
+}
+
+server {
+    listen 80;
+    listen [::]:80;
+    server_name api.example.com;
+
+    # 强制全站 HTTPS
+    return 301 https://$host$request_uri;
+}
+
+server {
+    listen 443 ssl http2;
+    listen [::]:443 ssl http2;
+    server_name api.example.com;
+
+    # 证书路径请按实际域名替换
+    ssl_certificate /etc/letsencrypt/live/api.example.com/fullchain.pem;
+    ssl_certificate_key /etc/letsencrypt/live/api.example.com/privkey.pem;
+    ssl_trusted_certificate /etc/letsencrypt/live/api.example.com/chain.pem;
+
+    ssl_session_timeout 1d;
+    ssl_session_cache shared:SSL:10m;
+    ssl_session_tickets off;
+
+    ssl_protocols TLSv1.2 TLSv1.3;
+    ssl_prefer_server_ciphers off;
+
+    add_header Strict-Transport-Security "max-age=31536000; includeSubDomains; preload" always;
+    add_header X-Content-Type-Options nosniff always;
+    add_header X-Frame-Options SAMEORIGIN always;
+    add_header Referrer-Policy no-referrer-when-downgrade always;
+
+    client_max_body_size 50m;
+
+    location / {
+        proxy_http_version 1.1;
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header X-Forwarded-Proto https;
+        proxy_set_header Connection "";
+        proxy_pass http://wx_service_upstream;
+    }
+}
+
@@ -0,0 +1,44 @@
+# 生产 HTTPS 配置与证书续期策略
+
+对应 issue：`#10 [P1][T7] 生产 HTTPS 配置与证书续期策略`
+
+## 1. Nginx HTTPS 强制
+
+- 参考配置：`deploy/nginx/wx_service_https.conf`
+- 核心策略：
+  - `80 -> 443` 永久重定向
+  - 仅暴露 `443 ssl http2`
+  - 添加 HSTS 与基础安全响应头
+  - 反代到后端 `127.0.0.1:8080`
+
+部署步骤：
+
+```bash
+sudo cp deploy/nginx/wx_service_https.conf /etc/nginx/conf.d/wx_service.conf
+sudo nginx -t
+sudo systemctl reload nginx
+```
+
+## 2. 证书自动续期
+
+脚本：`scripts/ops/renew_cert.sh`
+
+推荐 cron（每天 03:30）：
+
+```bash
+30 3 * * * CERTBOT_CMD=/usr/bin/certbot NGINX_RELOAD_CMD="systemctl reload nginx" OPS_ALERT_WEBHOOK="https://example.com/webhook" /path/to/wx_service/scripts/ops/renew_cert.sh >> /var/log/wx_service-cert-renew.log 2>&1
+```
+
+## 3. 续期失败与过期告警
+
+脚本：`scripts/ops/check_cert_expiry.sh`
+
+推荐 cron（每天 04:00）：
+
+```bash
+0 4 * * * TLS_DOMAIN=api.example.com TLS_CERT_PORT=443 TLS_MIN_DAYS=15 OPS_ALERT_WEBHOOK="https://example.com/webhook" /path/to/wx_service/scripts/ops/check_cert_expiry.sh >> /var/log/wx_service-cert-check.log 2>&1
+```
+
+说明：
+- 当剩余天数 `<= TLS_MIN_DAYS` 时返回非 0 并推送告警。
+- 当无法获取证书信息时也会告警并返回非 0。
@@ -0,0 +1,24 @@
+# HTTPS 与续期策略验证记录（2026-02-28）
+
+对应 issue：`#10 [P1][T7] 生产 HTTPS 配置与证书续期策略`
+
+## 验证内容
+
+1. 配置文件存在且可读
+- `deploy/nginx/wx_service_https.conf`
+
+2. 续期与告警脚本语法检查
+- `scripts/ops/renew_cert.sh`
+- `scripts/ops/check_cert_expiry.sh`
+
+## 验证命令
+
+```bash
+bash -n scripts/ops/renew_cert.sh
+bash -n scripts/ops/check_cert_expiry.sh
+```
+
+## 结果
+
+- 语法检查通过。
+- 已提供 HTTP->HTTPS 强制跳转、自动续期、过期/失败告警方案与定时任务示例。
@@ -0,0 +1,47 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 用法：
+# TLS_DOMAIN=api.example.com TLS_CERT_PORT=443 ./scripts/ops/check_cert_expiry.sh
+
+TLS_DOMAIN="${TLS_DOMAIN:-api.example.com}"
+TLS_CERT_PORT="${TLS_CERT_PORT:-443}"
+TLS_MIN_DAYS="${TLS_MIN_DAYS:-15}"
+OPS_ALERT_WEBHOOK="${OPS_ALERT_WEBHOOK:-}"
+ALERT_TITLE="${ALERT_TITLE:-[wx_service] HTTPS 证书即将过期}"
+
+send_alert() {
+  local message="$1"
+  if [[ -z "${OPS_ALERT_WEBHOOK}" ]]; then
+    echo "ALERT: ${message}" >&2
+    return
+  fi
+
+  curl -fsS -X POST "${OPS_ALERT_WEBHOOK}" \
+    -H "Content-Type: application/json" \
+    -d "{\"title\":\"${ALERT_TITLE}\",\"message\":\"${message}\"}" >/dev/null || true
+}
+
+if ! command -v openssl >/dev/null 2>&1; then
+  echo "openssl is required" >&2
+  exit 1
+fi
+
+expiry_line="$(echo | openssl s_client -servername "${TLS_DOMAIN}" -connect "${TLS_DOMAIN}:${TLS_CERT_PORT}" 2>/dev/null | openssl x509 -noout -enddate || true)"
+if [[ -z "${expiry_line}" ]]; then
+  send_alert "无法获取 ${TLS_DOMAIN}:${TLS_CERT_PORT} 的证书过期时间。"
+  exit 1
+fi
+
+expiry_raw="${expiry_line#notAfter=}"
+expiry_epoch="$(date -d "${expiry_raw}" +%s)"
+now_epoch="$(date +%s)"
+remaining_days="$(( (expiry_epoch - now_epoch) / 86400 ))"
+
+if (( remaining_days <= TLS_MIN_DAYS )); then
+  send_alert "${TLS_DOMAIN} 证书将在 ${remaining_days} 天后过期，请尽快续期。"
+  exit 1
+fi
+
+echo "certificate valid: ${remaining_days} days remaining"
+
@@ -0,0 +1,35 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# 用法：
+# CERTBOT_CMD=/usr/bin/certbot NGINX_RELOAD_CMD="systemctl reload nginx" ./scripts/ops/renew_cert.sh
+
+CERTBOT_CMD="${CERTBOT_CMD:-certbot}"
+NGINX_RELOAD_CMD="${NGINX_RELOAD_CMD:-systemctl reload nginx}"
+OPS_ALERT_WEBHOOK="${OPS_ALERT_WEBHOOK:-}"
+ALERT_TITLE="${ALERT_TITLE:-[wx_service] HTTPS 证书续期失败}"
+
+send_alert() {
+  local message="$1"
+  if [[ -z "${OPS_ALERT_WEBHOOK}" ]]; then
+    echo "ALERT: ${message}" >&2
+    return
+  fi
+
+  curl -fsS -X POST "${OPS_ALERT_WEBHOOK}" \
+    -H "Content-Type: application/json" \
+    -d "{\"title\":\"${ALERT_TITLE}\",\"message\":\"${message}\"}" >/dev/null || true
+}
+
+if ! "${CERTBOT_CMD}" renew --quiet; then
+  send_alert "certbot renew 执行失败，请立即检查生产证书状态。"
+  exit 1
+fi
+
+if ! bash -lc "${NGINX_RELOAD_CMD}"; then
+  send_alert "证书续期后 Nginx reload 失败，请检查服务状态。"
+  exit 1
+fi
+
+echo "certificate renew completed"
+