\u9999\u6e2f\u6578\u64da\u4e2d\u5fc3<\/a>\u7684\u4f3a\u670d\u5668\u7576\u6a5f<\/a>\u5c0dIT\u57fa\u790e\u8a2d\u65bd\u4f86\u8aaa\u662f\u4e00\u500b\u56b4\u5cfb\u7684\u6311\u6230\uff0c\u53ef\u80fd\u6703\u5728\u4e9e\u592a\u5730\u5340\u7684\u6578\u4f4d\u751f\u614b\u7cfb\u7d71\u4e2d\u5f15\u767c\u4e00\u7cfb\u5217\u6280\u8853\u6545\u969c\u3002\u672c\u6280\u8853\u5206\u6790\u63a2\u8a0e\u4f3a\u670d\u5668\u67b6\u69cb\u4e2d\u7684\u8907\u96dc\u4f9d\u8cf4\u95dc\u4fc2\uff0c\u4e26\u63d0\u4f9b\u5f37\u5065\u7684\u6545\u969c\u8f49\u79fb\u6a5f\u5236\u7684\u7a0b\u5f0f\u78bc\u7d1a\u89e3\u6c7a\u65b9\u6848\u3002<\/p>\n \u9999\u6e2f\u7684\u4f3a\u670d\u5668\u57fa\u790e\u8a2d\u65bd\u901a\u5e38\u63a1\u7528\u591a\u5c64\u67b6\u69cb\uff0c\u5305\u62ec\u8ca0\u8f09\u5e73\u8861\u5668\u3001\u61c9\u7528\u4f3a\u670d\u5668\u548c\u8cc7\u6599\u5eab\u96c6\u7fa4\u3002\u7576\u4f3a\u670d\u5668\u767c\u751f\u7576\u6a5f\u6642\uff0c\u4ee5\u4e0b\u6280\u8853\u7d44\u4ef6\u6703\u7acb\u5373\u53d7\u5230\u5f71\u97ff\uff1a<\/p>\n \u8b93\u6211\u5011\u901a\u904e\u7a0b\u5f0f\u78bc\u793a\u4f8b\u4f86\u6aa2\u67e5\u5178\u578b\u4f3a\u670d\u5668\u5806\u758a\u53ca\u5176\u6545\u969c\u9ede\u3002\u4ee5\u4e0b\u662f\u8a31\u591a\u9999\u6e2f\u4f3a\u670d\u5668\u4f7f\u7528\u7684\u57fa\u672cNode.js\u5065\u5eb7\u6aa2\u67e5\u5be6\u73fe\uff1a<\/p>\n \u7576\u9999\u6e2f\u4f3a\u670d\u5668\u767c\u751f\u7576\u6a5f\u6642\uff0c\u9023\u9396\u53cd\u61c9\u901a\u5e38\u8868\u73fe\u70baDNS\u89e3\u6790\u5931\u6557\u3002\u4ee5\u4e0b\u662f\u4e00\u500b\u5c55\u793aDNS\u50b3\u64ad\u554f\u984c\u5982\u4f55\u7d1a\u806f\u7684Python\u8173\u672c\uff1a<\/p>\n \u9999\u6e2f\u4f5c\u70ba\u91d1\u878d\u4e2d\u5fc3\uff0c\u4f3a\u670d\u5668\u53ef\u9760\u6027\u81f3\u95dc\u91cd\u8981\u3002\u4ea4\u6613\u7cfb\u7d71\u901a\u5e38\u5be6\u65bd\u5fc3\u8df3\u6a5f\u5236\u4f86\u6aa2\u6e2c\u4f3a\u670d\u5668\u5065\u5eb7\u72c0\u6cc1\uff1a<\/p>\n \u9999\u6e2f\u57fa\u790e\u8a2d\u65bd\u7684\u4e92\u806f\u6027\u610f\u5473\u8457\u4f3a\u670d\u5668\u7576\u6a5f\u6703\u986f\u8457\u5f71\u97ff\u8de8\u5883\u71df\u904b\u3002\u73fe\u4ee3\u4f3a\u670d\u5668\u79df\u7528\u63d0\u4f9b\u5546\u4f7f\u7528\u5bb9\u5668\u5316\u89e3\u6c7a\u65b9\u6848\u5be6\u65bd\u8907\u96dc\u7684\u76e3\u63a7\u7cfb\u7d71\uff1a<\/p>\n \u4f3a\u670d\u5668\u7576\u6a5f\u901a\u5e38\u6703\u89f8\u767c\u4e00\u7cfb\u5217\u8907\u96dc\u7684\u6280\u8853\u6545\u969c\u3002\u4ee5\u4e0b\u662f\u9999\u6e2f\u4f3a\u670d\u5668\u79df\u7528\u63d0\u4f9b\u5546\u5e38\u7528\u7684\u9ad8\u53ef\u7528\u6027\u914d\u7f6e\u5be6\u4f8b\uff1a<\/p>\n \u5be6\u65bd\u5f37\u5927\u7684\u76e3\u63a7\u548c\u8b66\u5831\u7cfb\u7d71\u81f3\u95dc\u91cd\u8981\u3002\u4ee5\u4e0b\u662f\u53ef\u5e6b\u52a9\u5728\u554f\u984c\u7d1a\u806f\u4e4b\u524d\u6aa2\u6e2c\u6f5b\u5728\u554f\u984c\u7684Prometheus\u8b66\u5831\u898f\u5247\u914d\u7f6e\uff1a<\/p>\n \u70ba\u4e86\u6e1b\u8f15\u4f3a\u670d\u5668\u7576\u6a5f\u7684\u5f71\u97ff\uff0c\u8a31\u591a\u7d44\u7e54\u5be6\u65bd\u591a\u5340\u57df\u90e8\u7f72\u7b56\u7565\u3002\u4ee5\u4e0b\u662f\u5c55\u793a\u6b64\u65b9\u6cd5\u7684Terraform\u914d\u7f6e\uff1a<\/p>\n \u5be6\u65bd\u81ea\u52d5\u5fa9\u539f\u7a0b\u5e8f\u5c0d\u65bc\u6700\u5c0f\u5316\u7576\u6a5f\u6642\u9593\u81f3\u95dc\u91cd\u8981\u3002\u4ee5\u4e0b\u662f\u5c55\u793a\u81ea\u52d5\u6545\u969c\u8f49\u79fb\u904e\u7a0b\u7684bash\u8173\u672c\uff1a<\/p>\n \u9999\u6e2f\u7684\u73fe\u4ee3\u4f3a\u670d\u5668\u79df\u7528\u63d0\u4f9b\u5546\u6b63\u5728\u4f7f\u7528\u5bb9\u5668\u7de8\u6392\u5be6\u65bd\u5148\u9032\u7684\u76e3\u63a7\u89e3\u6c7a\u65b9\u6848\u3002\u4ee5\u4e0b\u662f\u7528\u65bc\u5f48\u6027\u76e3\u63a7\u5806\u758a\u7684Kubernetes\u6e05\u55ae\uff1a<\/p>\n \u9999\u6e2f\u4f3a\u670d\u5668\u7576\u6a5f\u7684\u9023\u9396\u6548\u61c9\u9060\u8d85\u51fa\u76f4\u63a5\u7684\u6280\u8853\u4e2d\u65b7\uff0c\u5f71\u97ff\u6574\u500b\u4e9e\u592a\u5730\u5340\u7684\u6578\u4f4d\u57fa\u790e\u8a2d\u65bd\u3002\u7d44\u7e54\u5fc5\u9808\u5be6\u65bd\u5f37\u5065\u7684\u4f3a\u670d\u5668\u79df\u7528\u89e3\u6c7a\u65b9\u6848\uff0c\u5305\u62ec\u9069\u7576\u7684\u6545\u969c\u8f49\u79fb\u6a5f\u5236\u3001\u6301\u7e8c\u76e3\u63a7\u548c\u707d\u96e3\u5fa9\u539f\u7a0b\u5e8f\u3002\u5b9a\u671f\u6e2c\u8a66\u5099\u4efd\u7cfb\u7d71\u548c\u7dad\u8b77\u66f4\u65b0\u7684\u6587\u4ef6\u5c0d\u65bc\u6700\u5c0f\u5316\u7576\u6a5f\u5f71\u97ff\u81f3\u95dc\u91cd\u8981\u3002<\/p>\n \u5c0d\u65bc\u4f9d\u8cf4\u9999\u6e2f\u4f3a\u670d\u5668\u57fa\u790e\u8a2d\u65bd\u7684\u4f01\u696d\u4f86\u8aaa\uff0c\u6295\u8cc7\u5197\u9918\u7cfb\u7d71\u548c\u5be6\u65bd\u5168\u9762\u7684\u76e3\u63a7\u89e3\u6c7a\u65b9\u6848\u4e0d\u50c5\u662f\u6280\u8853\u8981\u6c42\uff0c\u66f4\u662f\u696d\u52d9\u5fc5\u9700\u3002\u7d50\u5408\u9069\u7576\u7684\u4f3a\u670d\u5668\u8a17\u7ba1\u7b56\u7565\u3001\u81ea\u52d5\u5fa9\u539f\u7a0b\u5e8f\u548c\u591a\u5340\u57df\u4f3a\u670d\u5668\u79df\u7528\u67b6\u69cb\u53ef\u4ee5\u986f\u8457\u964d\u4f4e\u4f3a\u670d\u5668\u7576\u6a5f\u7684\u98a8\u96aa\u548c\u5f71\u97ff\u3002<\/p>\n","protected":false},"excerpt":{"rendered":" \u9999\u6e2f\u6578\u64da\u4e2d\u5fc3\u7684\u4f3a\u670d\u5668\u7576\u6a5f\u5c0dIT\u57fa\u790e\u8a2d\u65bd\u4f86\u8aaa\u662f\u4e00\u500b\u56b4\u5cfb\u7684\u6311\u6230\uff0c\u53ef\u80fd\u6703\u5728\u4e9e\u592a\u5730\u5340\u7684\u6578\u4f4d\u751f\u614b\u7cfb\u7d71\u4e2d\u5f15\u767c\u4e00\u7cfb\u5217\u6280\u8853\u6545\u969c […]<\/p>\n\u7406\u89e3\u6280\u8853\u67b6\u69cb<\/strong><\/h2>\n
\n
\u5373\u6642\u6280\u8853\u5f71\u97ff<\/strong><\/h2>\n
\r\nconst express = require('express');\r\nconst app = express();\r\n\r\nconst healthCheck = {\r\n uptime: process.uptime(),\r\n timestamp: Date.now(),\r\n status: 'OK'\r\n};\r\n\r\napp.get('\/health', (req, res) => {\r\n try {\r\n res.status(200).send(healthCheck);\r\n } catch (error) {\r\n healthCheck.status = 'ERROR';\r\n res.status(503).send();\r\n }\r\n});\r\n<\/code><\/pre>\n\u7cfb\u7d71\u7d1a\u806f\u6545\u969c<\/strong><\/h2>\n
\r\nimport dns.resolver\r\nimport time\r\n\r\ndef check_dns_propagation(domain, nameservers):\r\n resolver = dns.resolver.Resolver()\r\n resolver.nameservers = nameservers\r\n \r\n try:\r\n records = resolver.resolve(domain, 'A')\r\n return [rdata.address for rdata in records]\r\n except dns.resolver.NXDOMAIN:\r\n return \"DNS record not found\"\r\n except dns.resolver.Timeout:\r\n return \"DNS lookup timed out\"\r\n\r\ndef monitor_dns_health():\r\n nameservers = ['8.8.8.8', '1.1.1.1']\r\n domains = ['example.hk', 'backup-server.hk']\r\n \r\n while True:\r\n for domain in domains:\r\n status = check_dns_propagation(domain, nameservers)\r\n if isinstance(status, str):\r\n trigger_failover(domain)\r\n time.sleep(300)\r\n<\/code><\/pre>\n\u91d1\u878d\u7cfb\u7d71\u5f71\u97ff\u5206\u6790<\/strong><\/h2>\n
\r\nclass ServerHealthMonitor:\r\n def __init__(self, threshold_ms=500):\r\n self.threshold_ms = threshold_ms\r\n self.last_heartbeat = time.time()\r\n self.status = \"healthy\"\r\n\r\n def check_latency(self):\r\n current_time = time.time()\r\n latency = (current_time - self.last_heartbeat) * 1000\r\n \r\n if latency > self.threshold_ms:\r\n self.status = \"degraded\"\r\n self.trigger_alert()\r\n return False\r\n return True\r\n\r\n def trigger_alert(self):\r\n # Alert DevOps team\r\n notification = {\r\n \"severity\": \"high\",\r\n \"message\": f\"Server latency exceeded {self.threshold_ms}ms\",\r\n \"timestamp\": datetime.now().isoformat()\r\n }\r\n send_alert(notification)\r\n<\/code><\/pre>\n\u8de8\u5883\u696d\u52d9\u5f71\u97ff<\/strong><\/h2>\n
\r\nversion: '3'\r\nservices:\r\n uptime-monitor:\r\n image: uptimekuma\/uptime-kuma:latest\r\n container_name: uptime-kuma\r\n volumes:\r\n - uptime-kuma:\/app\/data\r\n ports:\r\n - \"3001:3001\"\r\n restart: always\r\n environment:\r\n - UPTIME_KUMA_PORT=3001\r\n - TZ=Asia\/Hong_Kong\r\n\r\nvolumes:\r\n uptime-kuma:\r\n<\/code><\/pre>\n\u6280\u8853\u93c8\u5f0f\u53cd\u61c9\u548c\u7cfb\u7d71\u4f9d\u8cf4\u6027<\/strong><\/h2>\n
\r\nevents {\r\n worker_connections 1024;\r\n}\r\n\r\nhttp {\r\n upstream backend_servers {\r\n server backend1.example.hk:8080 max_fails=3 fail_timeout=30s;\r\n server backend2.example.hk:8080 max_fails=3 fail_timeout=30s;\r\n server backup.example.hk:8080 backup;\r\n }\r\n\r\n server {\r\n listen 80;\r\n server_name example.hk;\r\n\r\n location \/healthcheck {\r\n proxy_pass http:\/\/backend_servers;\r\n proxy_next_upstream error timeout http_500;\r\n proxy_connect_timeout 2s;\r\n proxy_send_timeout 5s;\r\n proxy_read_timeout 5s;\r\n }\r\n }\r\n}\r\n<\/code><\/pre>\n\u9810\u9632\u63aa\u65bd\u548c\u5fa9\u539f\u7b56\u7565<\/strong><\/h2>\n
\r\ngroups:\r\n- name: server_health\r\n rules:\r\n - alert: HighLatency\r\n expr: rate(http_request_duration_seconds_sum[5m]) > 0.5\r\n for: 5m\r\n labels:\r\n severity: warning\r\n annotations:\r\n summary: High latency detected\r\n description: Server response time is above 500ms for 5 minutes\r\n\r\n - alert: ServerDown\r\n expr: up == 0\r\n for: 1m\r\n labels:\r\n severity: critical\r\n annotations:\r\n summary: Server is down\r\n description: \"Server {{ $labels.instance }} has been down for more than 1 minute\"\r\n<\/code><\/pre>\n\u591a\u5340\u57df\u90e8\u7f72\u67b6\u69cb<\/strong><\/h2>\n
\r\nprovider \"aws\" {\r\n region = \"ap-east-1\" # Hong Kong region\r\n}\r\n\r\nresource \"aws_instance\" \"primary_server\" {\r\n ami = \"ami-0123456789\"\r\n instance_type = \"t3.medium\"\r\n \r\n tags = {\r\n Name = \"HK-Primary\"\r\n Environment = \"Production\"\r\n }\r\n}\r\n\r\nresource \"aws_route53_health_check\" \"failover_check\" {\r\n fqdn = aws_instance.primary_server.public_dns\r\n port = 80\r\n type = \"HTTP\"\r\n resource_path = \"\/health\"\r\n failure_threshold = \"3\"\r\n request_interval = \"30\"\r\n \r\n tags = {\r\n Name = \"Primary-Health-Check\"\r\n }\r\n}\r\n\r\nresource \"aws_instance\" \"backup_server\" {\r\n provider = aws.backup_region\r\n ami = \"ami-9876543210\"\r\n instance_type = \"t3.medium\"\r\n\r\n tags = {\r\n Name = \"SG-Backup\"\r\n Environment = \"DR\"\r\n }\r\n}\r\n<\/code><\/pre>\n\u81ea\u52d5\u5fa9\u539f\u7a0b\u5e8f<\/strong><\/h2>\n
\r\n#!\/bin\/bash\r\n\r\n# Configuration\r\nPRIMARY_IP=\"10.0.1.100\"\r\nBACKUP_IP=\"10.0.1.101\"\r\nTHRESHOLD=3\r\nINTERVAL=5\r\n\r\ncheck_server() {\r\n local server_ip=$1\r\n curl -s --connect-timeout 5 http:\/\/${server_ip}\/health > \/dev\/null\r\n return $?\r\n}\r\n\r\ninitiate_failover() {\r\n logger \"Initiating failover procedure to backup server\"\r\n \r\n # Update DNS\r\n aws route53 change-resource-record-sets \\\r\n --hosted-zone-id ${HOSTED_ZONE_ID} \\\r\n --change-batch file:\/\/failover-record.json\r\n \r\n # Notify team\r\n curl -X POST ${SLACK_WEBHOOK} \\\r\n -H 'Content-type: application\/json' \\\r\n --data '{\"text\":\"Failover initiated to backup server\"}'\r\n}\r\n<\/code><\/pre>\n\u9762\u5411\u672a\u4f86\u7684\u57fa\u790e\u8a2d\u65bd<\/strong><\/h2>\n
\r\napiVersion: apps\/v1\r\nkind: Deployment\r\nmetadata:\r\n name: monitoring-stack\r\n namespace: monitoring\r\nspec:\r\n replicas: 3\r\n selector:\r\n matchLabels:\r\n app: monitoring\r\n template:\r\n metadata:\r\n labels:\r\n app: monitoring\r\n spec:\r\n containers:\r\n - name: prometheus\r\n image: prom\/prometheus:v2.30.3\r\n ports:\r\n - containerPort: 9090\r\n volumeMounts:\r\n - name: config\r\n mountPath: \/etc\/prometheus\r\n - name: grafana\r\n image: grafana\/grafana:8.2.0\r\n ports:\r\n - containerPort: 3000\r\n volumes:\r\n - name: config\r\n configMap:\r\n name: prometheus-config\r\n<\/code><\/pre>\n\u7d50\u8ad6\u548c\u6700\u4f73\u5be6\u8e10<\/strong><\/h2>\n