\n| \u5927\u578b<\/td>\n | 5-10 \u8282\u70b9<\/td>\n | 2TB+<\/td>\n | 32GB+<\/td>\n | >2000 \u76d1\u63a7\u76ee\u6807<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
\n4. \u4e09\u5957\u90e8\u7f72\u65b9\u6848：\u88f8\u673a\u90e8\u7f72\u3001Docker Compose \u90e8\u7f72\u3001Kubernetes \u90e8\u7f72<\/h3>\n4.1 \u88f8\u673a\u4e8c\u8fdb\u5236\u90e8\u7f72（\u751f\u4ea7\u63a8\u8350）<\/h4>\n\u4f18\u52bf：\u6027\u80fd\u6700\u4f18\u3001\u8d44\u6e90\u5f00\u9500\u6700\u5c0f\u3001\u4fbf\u4e8e\u6df1\u5ea6\u8c03\u4f18<\/p>\n \u9002\u7528\u573a\u666f：\u7269\u7406\u673a\u96c6\u7fa4\u3001\u5bf9\u6027\u80fd\u8981\u6c42\u4e25\u82db\u7684\u751f\u4ea7\u73af\u5883<\/p>\n # \u521b\u5efa\u76ee\u5f55\u7ed3\u6784
\nmkdir -p \/opt\/aiops\/{prometheus,alertmanager,grafana,thanos,consul}
\ncd \/opt\/aiops\/prometheus<\/p>\n # \u4e0b\u8f7d\u4e8c\u8fdb\u5236\u5305
\nwget https:\/\/github.com\/prometheus\/prometheus\/releases\/download\/v2.45.0\/prometheus-2.45.0.linux-amd64.tar.gz
\ntar -xzf prometheus-2.45.0.linux-amd64.tar.gz
\nmv prometheus-2.45.0.linux-amd64\/* .<\/p>\n # \u521b\u5efa systemd \u670d\u52a1
\ncat > \/etc\/systemd\/system\/prometheus.service <<EOF
\n[Unit]
\nDescription=Prometheus Server
\nAfter=network.target<\/p>\n [Service]
\nUser=prometheus
\nGroup=prometheus
\nExecStart=\/opt\/aiops\/prometheus\/prometheus \\\\
\n –config.file=\/opt\/aiops\/prometheus\/prometheus.yml \\\\
\n –storage.tsdb.path=\/opt\/aiops\/prometheus\/data \\\\
\n –storage.tsdb.retention.time=15d \\\\
\n –web.listen-address=:9090
\nRestart=always<\/p>\n [Install]
\nWantedBy=multi-user.target
\nEOF<\/p>\n systemctl daemon-reload
\nsystemctl enable prometheus
\nsystemctl start prometheus<\/p>\n 4.2 Docker Compose \u90e8\u7f72（\u6d4b\u8bd5\/\u5f00\u53d1\u63a8\u8350）<\/h4>\n\u4f18\u52bf：\u90e8\u7f72\u7b80\u5355\u3001\u4fbf\u4e8e\u8fc1\u79fb\u3001\u73af\u5883\u9694\u79bb<\/p>\n \u9002\u7528\u573a\u666f：\u6d4b\u8bd5\u73af\u5883\u3001\u5f00\u53d1\u73af\u5883\u3001\u5feb\u901f\u9a8c\u8bc1<\/p>\n # docker-compose.yml
\nversion: '3.8'
\nservices:
\n prometheus:
\n image: prom\/prometheus:v2.45.0
\n container_name: prometheus
\n ports:
\n – "9090:9090"
\n volumes:
\n – .\/prometheus.yml:\/etc\/prometheus\/prometheus.yml
\n – prometheus_data:\/prometheus
\n command:
\n – '–config.file=\/etc\/prometheus\/prometheus.yml'
\n – '–storage.tsdb.path=\/prometheus'
\n – '–storage.tsdb.retention.time=15d'
\n restart: always<\/p>\n alertmanager:
\n image: prom\/alertmanager:v0.26.0
\n container_name: alertmanager
\n ports:
\n – "9093:9093"
\n volumes:
\n – .\/alertmanager.yml:\/etc\/alertmanager\/alertmanager.yml
\n restart: always<\/p>\n grafana:
\n image: grafana\/grafana:10.0.0
\n container_name: grafana
\n ports:
\n – "3000:3000"
\n volumes:
\n – grafana_data:\/var\/lib\/grafana
\n environment:
\n – GF_SECURITY_ADMIN_PASSWORD=admin123
\n restart: always<\/p>\n volumes:
\n prometheus_data:
\n grafana_data:<\/p>\n # \u542f\u52a8\u670d\u52a1
\ndocker-compose up -d<\/p>\n # \u67e5\u770b\u72b6\u6001
\ndocker-compose ps<\/p>\n # \u67e5\u770b\u65e5\u5fd7
\ndocker-compose logs -f prometheus<\/p>\n 4.3 Kubernetes \u96c6\u7fa4\u90e8\u7f72（\u4e91\u539f\u751f\u63a8\u8350）<\/h4>\n\u4f18\u52bf：\u9ad8\u53ef\u7528\u3001\u5f39\u6027\u4f38\u7f29\u3001\u81ea\u52a8\u6545\u969c\u6062\u590d<\/p>\n \u9002\u7528\u573a\u666f：K8s \u751f\u4ea7\u73af\u5883\u3001\u4e91\u539f\u751f\u67b6\u6784<\/p>\n # prometheus-deployment.yaml
\napiVersion: apps\/v1
\nkind: Deployment
\nmetadata:
\n name: prometheus
\n namespace: monitoring
\nspec:
\n replicas: 2
\n selector:
\n matchLabels:
\n app: prometheus
\n template:
\n metadata:
\n labels:
\n app: prometheus
\n spec:
\n containers:
\n – name: prometheus
\n image: prom\/prometheus:v2.45.0
\n ports:
\n – containerPort: 9090
\n volumeMounts:
\n – name: config
\n mountPath: \/etc\/prometheus
\n – name: data
\n mountPath: \/prometheus
\n args:
\n – –config.file=\/etc\/prometheus\/prometheus.yml
\n – –storage.tsdb.path=\/prometheus
\n – –storage.tsdb.retention.time=15d
\n volumes:
\n – name: config
\n configMap:
\n name: prometheus-config
\n – name: data
\n persistentVolumeClaim:
\n claimName: prometheus-pvc
\n—
\napiVersion: v1
\nkind: Service
\nmetadata:
\n name: prometheus
\n namespace: monitoring
\nspec:
\n selector:
\n app: prometheus
\n ports:
\n – port: 9090
\n targetPort: 9090
\n type: ClusterIP<\/p>\n # \u90e8\u7f72\u5230 K8s \u96c6\u7fa4
\nkubectl apply -f prometheus-deployment.yaml<\/p>\n # \u9a8c\u8bc1\u90e8\u7f72
\nkubectl get pods -n monitoring
\nkubectl get svc -n monitoring<\/p>\n # \u7aef\u53e3\u8f6c\u53d1\u8bbf\u95ee
\nkubectl port-forward svc\/prometheus 9090:9090 -n monitoring<\/p>\n
\n5. \u6838\u5fc3\u914d\u7f6e\u6587\u4ef6\u8be6\u89e3（\u5168\u4e2d\u6587\u6ce8\u91ca）<\/h3>\n5.1 Prometheus \u4e3b\u914d\u7f6e\u6587\u4ef6<\/h4>\n# prometheus.yml – Prometheus \u6838\u5fc3\u914d\u7f6e\u6587\u4ef6<\/p>\n # \u5168\u5c40\u914d\u7f6e
\nglobal:
\n scrape_interval: 15s # \u9ed8\u8ba4\u6293\u53d6\u95f4\u9694 15 \u79d2
\n evaluation_interval: 15s # \u89c4\u5219\u8bc4\u4f30\u95f4\u9694 15 \u79d2
\n external_labels:
\n monitor: 'aiops-monitor' # \u5916\u90e8\u6807\u7b7e，\u7528\u4e8e Thanos \u533a\u5206\u96c6\u7fa4
\n region: 'cn-beijing' # \u533a\u57df\u6807\u7b7e<\/p>\n # \u544a\u8b66\u7ba1\u7406\u5668\u914d\u7f6e
\nalerting:
\n alertmanagers:
\n – static_configs:
\n – targets:
\n – alertmanager:9093 # Alertmanager \u5730\u5740<\/p>\n # \u89c4\u5219\u6587\u4ef6\u8def\u5f84
\nrule_files:
\n – \/opt\/aiops\/prometheus\/rules\/*.yml<\/p>\n # \u6293\u53d6\u914d\u7f6e
\nscrape_configs:
\n # Prometheus \u81ea\u76d1\u63a7
\n – job_name: 'prometheus'
\n static_configs:
\n – targets: ['localhost:9090']
\n labels:
\n instance_type: 'monitoring'<\/p>\n # Node Exporter \u670d\u52a1\u5668\u76d1\u63a7
\n – job_name: 'node_exporter'
\n static_configs:
\n – targets: ['node1:9100', 'node2:9100', 'node3:9100']
\n labels:
\n instance_type: 'server'<\/p>\n # Blackbox \u62e8\u6d4b\u76d1\u63a7
\n – job_name: 'blackbox_http'
\n metrics_path: \/probe
\n params:
\n module: [http_2xx]
\n static_configs:
\n – targets:
\n – https:\/\/www.example.com
\n relabel_configs:
\n – source_labels: [__address__]
\n target_label: __param_target
\n – source_labels: [__param_target]
\n target_label: instance
\n – target_label: __address__
\n replacement: blackbox:9115<\/p>\n # Consul \u670d\u52a1\u53d1\u73b0
\n – job_name: 'consul_services'
\n consul_sd_configs:
\n – server: consul:8500
\n services: []
\n relabel_configs:
\n – source_labels: [__meta_consul_service]
\n target_label: job<\/p>\n 5.2 Alertmanager \u914d\u7f6e\u6587\u4ef6<\/h4>\n# alertmanager.yml – \u544a\u8b66\u7ba1\u7406\u5668\u914d\u7f6e<\/p>\n global:
\n smtp_smarthost: 'smtp.example.com:587'
\n smtp_from: 'alertmanager@example.com'
\n smtp_auth_username: 'alertmanager@example.com'
\n smtp_auth_password: 'password123'<\/p>\n # \u8def\u7531\u6811\u914d\u7f6e
\nroute:
\n receiver: 'default-receiver' # \u9ed8\u8ba4\u63a5\u6536\u5668
\n group_by: ['alertname', 'cluster'] # \u544a\u8b66\u5206\u7ec4\u7ef4\u5ea6
\n group_wait: 30s # \u7ec4\u7b49\u5f85\u65f6\u95f4
\n group_interval: 5m # \u7ec4\u95f4\u9694\u65f6\u95f4
\n repeat_interval: 4h # \u91cd\u590d\u901a\u77e5\u95f4\u9694<\/p>\n # \u5b50\u8def\u7531 – \u6309\u4e25\u91cd\u7a0b\u5ea6\u5206\u6d41
\n routes:
\n – match:
\n severity: critical
\n receiver: 'critical-receiver'
\n group_wait: 10s
\n repeat_interval: 1h
\n – match:
\n severity: warning
\n receiver: 'warning-receiver'
\n repeat_interval: 4h<\/p>\n # \u63a5\u6536\u5668\u914d\u7f6e
\nreceivers:
\n– name: 'default-receiver'
\n email_configs:
\n – to: 'ops-team@example.com'
\n send_resolved: true<\/p>\n – name: 'critical-receiver'
\n webhook_configs:
\n – url: 'http:\/\/n8n:5678\/webhook\/alertmanager'
\n send_resolved: true
\n email_configs:
\n – to: 'oncall@example.com'
\n send_resolved: true<\/p>\n – name: 'warning-receiver'
\n email_configs:
\n – to: 'ops-team@example.com'
\n send_resolved: true<\/p>\n # \u6291\u5236\u89c4\u5219 – \u9632\u6b62\u544a\u8b66\u98ce\u66b4
\ninhibit_rules:
\n– source_match:
\n severity: 'critical'
\n target_match:
\n severity: 'warning'
\n equal: ['alertname', 'cluster', 'service']<\/p>\n # \u9759\u9ed8\u89c4\u5219
\nsilencers:
\n– matchers:
\n – name: alertname
\n value: 'TestAlert'
\n startsAt: '2024-01-01T00:00:00Z'
\n endsAt: '2024-01-01T23:59:59Z'<\/p>\n
\n6. \u529f\u80fd\u9a8c\u8bc1\u3001\u6307\u6807\u6293\u53d6\u9a8c\u8bc1\u3001\u670d\u52a1\u53ef\u7528\u6027\u9a8c\u8bc1<\/h3>\n6.1 Prometheus \u670d\u52a1\u9a8c\u8bc1<\/h4>\n# \u68c0\u67e5\u670d\u52a1\u72b6\u6001
\nsystemctl status prometheus<\/p>\n # \u68c0\u67e5\u7aef\u53e3\u76d1\u542c
\nnetstat -tlnp | grep 9090<\/p>\n # \u68c0\u67e5 API \u5065\u5eb7\u72b6\u6001
\ncurl -s http:\/\/localhost:9090\/api\/v1\/status\/config | jq .<\/p>\n # \u68c0\u67e5\u76ee\u6807\u6293\u53d6\u72b6\u6001
\ncurl -s http:\/\/localhost:9090\/api\/v1\/targets | jq '.data.activeTargets[].health'<\/p>\n 6.2 \u6307\u6807\u6293\u53d6\u9a8c\u8bc1<\/h4>\n# \u67e5\u8be2 Prometheus \u81ea\u8eab\u6307\u6807
\ncurl -s 'http:\/\/localhost:9090\/api\/v1\/query?query=prometheus_tsdb_head_samples_appended_total' | jq .<\/p>\n # \u67e5\u8be2 Node Exporter \u6307\u6807
\ncurl -s 'http:\/\/localhost:9090\/api\/v1\/query?query=node_cpu_seconds_total' | jq .<\/p>\n # \u9a8c\u8bc1\u6307\u6807\u65f6\u95f4\u8303\u56f4
\ncurl -s 'http:\/\/localhost:9090\/api\/v1\/query_range?query=up&start=2024-01-01T00:00:00Z&end=2024-01-01T23:59:59Z&step=1m' | jq .<\/p>\n 6.3 Grafana \u6570\u636e\u6e90\u9a8c\u8bc1<\/h4>\n\u767b\u5f55 Grafana（http:\/\/grafana:3000，\u9ed8\u8ba4\u8d26\u53f7 admin\/admin123）<\/li>\n\u8fdb\u5165 Configuration \u2192 Data Sources<\/li>\n\u70b9\u51fb Prometheus \u6570\u636e\u6e90<\/li>\n\u70b9\u51fb "Save & Test"，\u663e\u793a "Data source is working" \u8868\u793a\u6210\u529f<\/li>\n6.4 \u544a\u8b66\u94fe\u8def\u9a8c\u8bc1<\/h4>\n# \u624b\u52a8\u89e6\u53d1\u6d4b\u8bd5\u544a\u8b66
\ncat > \/tmp\/test_alert.yml <<EOF
\ngroups:
\n– name: test_alert
\n rules:
\n – alert: TestAlert
\n expr: vector(1)
\n for: 10s
\n labels:
\n severity: critical
\n annotations:
\n summary: "\u6d4b\u8bd5\u544a\u8b66"
\n description: "\u9a8c\u8bc1\u544a\u8b66\u94fe\u8def"
\nEOF<\/p>\n # \u91cd\u8f7d Prometheus \u914d\u7f6e
\ncurl -X POST http:\/\/localhost:9090\/-\/reload<\/p>\n # \u67e5\u770b\u544a\u8b66\u72b6\u6001
\ncurl -s http:\/\/localhost:9090\/api\/v1\/rules | jq .<\/p>\n
\n7. \u7ec4\u4ef6\u81ea\u76d1\u63a7\u914d\u7f6e\u3001\u5bf9\u5e94 Exporter \u6307\u6807\u91c7\u96c6\u3001\u544a\u8b66\u89c4\u5219\u914d\u7f6e<\/h3>\n7.1 Prometheus \u81ea\u76d1\u63a7\u6307\u6807<\/h4>\n# prometheus_self_monitor.yml
\ngroups:
\n– name: prometheus_self
\n rules:
\n – alert: PrometheusDown
\n expr: absent(up{job="prometheus"})
\n for: 1m
\n labels:
\n severity: critical
\n annotations:
\n summary: "Prometheus \u670d\u52a1\u5b95\u673a"
\n description: "Prometheus \u5b9e\u4f8b {{ $labels.instance }} \u5df2\u5b95\u673a\u8d85\u8fc7 1 \u5206\u949f"<\/p>\n – alert: PrometheusTargetMissing
\n expr: up == 0
\n for: 5m
\n labels:
\n severity: warning
\n annotations:
\n summary: "\u76d1\u63a7\u76ee\u6807\u4e22\u5931"
\n description: "\u76ee\u6807 {{ $labels.instance }} \u5df2\u5931\u8054\u8d85\u8fc7 5 \u5206\u949f"<\/p>\n – alert: PrometheusStoragePressure
\n expr: prometheus_tsdb_storage_blocks_bytes_total \/ prometheus_tsdb_storage_blocks_bytes_max > 0.9
\n for: 10m
\n labels:
\n severity: warning
\n annotations:
\n summary: "Prometheus \u5b58\u50a8\u538b\u529b\u8fc7\u5927"
\n description: "\u5b58\u50a8\u4f7f\u7528\u7387\u8d85\u8fc7 90%"<\/p>\n 7.2 Node Exporter \u6838\u5fc3\u544a\u8b66\u89c4\u5219<\/h4>\n# node_exporter_alerts.yml
\ngroups:
\n– name: node_alerts
\n rules:
\n – alert: HighCPUUsage
\n expr: 100 – (avg by(instance) (rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 80
\n for: 10m
\n labels:
\n severity: warning
\n annotations:
\n summary: "CPU \u4f7f\u7528\u7387\u8fc7\u9ad8"
\n description: "\u5b9e\u4f8b {{ $labels.instance }} CPU \u4f7f\u7528\u7387\u8d85\u8fc7 80%"<\/p>\n – alert: HighMemoryUsage
\n expr: (1 – (node_memory_MemAvailable_bytes \/ node_memory_MemTotal_bytes)) * 100 > 85
\n for: 10m
\n labels:
\n severity: warning
\n annotations:
\n summary: "\u5185\u5b58\u4f7f\u7528\u7387\u8fc7\u9ad8"
\n description: "\u5b9e\u4f8b {{ $labels.instance }} \u5185\u5b58\u4f7f\u7528\u7387\u8d85\u8fc7 85%"<\/p>\n – alert: DiskSpaceLow
\n expr: (node_filesystem_avail_bytes \/ node_filesystem_size_bytes) * 100 < 15
\n for: 30m
\n labels:
\n severity: critical
\n annotations:
\n summary: "\u78c1\u76d8\u7a7a\u95f4\u4e0d\u8db3"
\n description: "\u5b9e\u4f8b {{ $labels.instance }} \u78c1\u76d8\u53ef\u7528\u7a7a\u95f4\u4f4e\u4e8e 15%"<\/p>\n – alert: NodeDown
\n expr: up{job="node_exporter"} == 0
\n for: 5m
\n labels:
\n severity: critical
\n annotations:
\n summary: "\u670d\u52a1\u5668\u5b95\u673a"
\n description: "\u670d\u52a1\u5668 {{ $labels.instance }} \u5df2\u5b95\u673a\u8d85\u8fc7 5 \u5206\u949f"<\/p>\n
\n8. \u5de5\u5177\u9ad8\u7ea7\u7528\u6cd5\u3001\u751f\u4ea7\u6700\u4f73\u5b9e\u8df5<\/h3>\n8.1 PromQL \u9ad8\u7ea7\u67e5\u8be2\u6280\u5de7<\/h4>\n# \u8ba1\u7b97 5 \u5206\u949f\u5e73\u5747 CPU \u4f7f\u7528\u7387
\n100 – (avg by(instance) (rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)<\/p>\n # \u8ba1\u7b97\u5185\u5b58\u4f7f\u7528\u7387
\n(1 – (node_memory_MemAvailable_bytes \/ node_memory_MemTotal_bytes)) * 100<\/p>\n # \u8ba1\u7b97\u78c1\u76d8 IO \u7b49\u5f85\u65f6\u95f4
\nrate(node_disk_io_time_seconds_total[5m]) * 100<\/p>\n # \u8ba1\u7b97\u7f51\u7edc\u5e26\u5bbd\u4f7f\u7528
\nrate(node_network_receive_bytes_total{device="eth0"}[5m]) * 8<\/p>\n # \u9884\u6d4b\u78c1\u76d8\u4f55\u65f6\u5199\u6ee1
\npredict_linear(node_filesystem_avail_bytes{mountpoint="\/"}[6h], 24*3600)<\/p>\n # \u540c\u6bd4\/\u73af\u6bd4\u5206\u6790
\n# \u672c\u5468 vs \u4e0a\u5468\u540c\u4e00\u65f6\u95f4
\nrate(http_requests_total[5m]) \/ rate(http_requests_total[5m] offset 7d)<\/p>\n 8.2 \u751f\u4ea7\u73af\u5883\u6700\u4f73\u5b9e\u8df5<\/h4>\n\n\u5b9e\u8df5\u9879\u63a8\u8350\u65b9\u6848\u8bf4\u660e<\/tr>\n \n\n| \u6570\u636e\u4fdd\u7559<\/td>\n | \u672c\u5730 15 \u5929+Thanos \u957f\u671f\u5b58\u50a8<\/td>\n | \u5e73\u8861\u67e5\u8be2\u6027\u80fd\u4e0e\u5b58\u50a8\u6210\u672c<\/td>\n<\/tr>\n | \n| \u6293\u53d6\u95f4\u9694<\/td>\n | \u6838\u5fc3\u6307\u6807 15s，\u6b21\u8981\u6307\u6807 60s<\/td>\n | \u964d\u4f4e Prometheus \u8d1f\u8f7d<\/td>\n<\/tr>\n | \n| \u544a\u8b66\u5206\u7ea7<\/td>\n | Critical\/Warning\/Info \u4e09\u7ea7<\/td>\n | \u907f\u514d\u544a\u8b66\u75b2\u52b3<\/td>\n<\/tr>\n | \n| \u6807\u7b7e\u89c4\u8303<\/td>\n | \u7edf\u4e00 instance\/job\/cluster\/region<\/td>\n | \u4fbf\u4e8e\u805a\u5408\u67e5\u8be2<\/td>\n<\/tr>\n | \n| \u914d\u7f6e\u7ba1\u7406<\/td>\n | Git \u7248\u672c\u63a7\u5236+CI\/CD \u81ea\u52a8\u90e8\u7f72<\/td>\n | \u4fdd\u8bc1\u914d\u7f6e\u4e00\u81f4\u6027<\/td>\n<\/tr>\n | \n| \u5907\u4efd\u7b56\u7565<\/td>\n | \u6bcf\u65e5\u5feb\u7167+\u5f02\u5730\u5907\u4efd<\/td>\n | \u9632\u6b62\u6570\u636e\u4e22\u5931<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n8.3 \u6027\u80fd\u8c03\u4f18\u53c2\u6570<\/h4>\n# Prometheus \u542f\u52a8\u53c2\u6570\u4f18\u5316
\n–storage.tsdb.retention.time=15d # \u6570\u636e\u4fdd\u7559\u65f6\u95f4
\n–storage.tsdb.min-block-duration=2h # \u6700\u5c0f\u5757\u65f6\u957f
\n–storage.tsdb.max-block-duration=2h # \u6700\u5927\u5757\u65f6\u957f
\n–storage.tsdb.wal-compression=true # WAL \u538b\u7f29
\n–web.max-connections=512 # \u6700\u5927\u8fde\u63a5\u6570
\n–query.max-concurrency=20 # \u67e5\u8be2\u5e76\u53d1\u6570
\n–query.timeout=2m # \u67e5\u8be2\u8d85\u65f6\u65f6\u95f4<\/p>\n
\n9. \u751f\u4ea7\u8e29\u5751\u6e05\u5355\u3001\u6545\u969c\u6392\u67e5\u3001\u67b6\u6784\u4f18\u5316\u65b9\u6848<\/h3>\n9.1 \u5e38\u89c1\u8e29\u5751\u6e05\u5355<\/h4>\n\n\u95ee\u9898\u73b0\u8c61\u89e3\u51b3\u65b9\u6848<\/tr>\n \n\n| \u5185\u5b58\u6ea2\u51fa<\/td>\n | Prometheus OOM \u91cd\u542f<\/td>\n | \u589e\u52a0\u5185\u5b58\u9650\u5236，\u542f\u7528 WAL \u538b\u7f29，\u51cf\u5c11\u9ad8\u57fa\u6570\u6807\u7b7e<\/td>\n<\/tr>\n | \n| \u67e5\u8be2\u8d85\u65f6<\/td>\n | Grafana \u9762\u677f\u52a0\u8f7d\u6162<\/td>\n | \u4f18\u5316 PromQL，\u6dfb\u52a0\u7d22\u5f15，\u4f7f\u7528 Thanos Query \u7f13\u5b58<\/td>\n<\/tr>\n | \n| \u544a\u8b66\u98ce\u66b4<\/td>\n | \u77ed\u65f6\u95f4\u5185\u6536\u5230\u6570\u767e\u6761\u544a\u8b66<\/td>\n | \u914d\u7f6e\u5206\u7ec4\u3001\u6291\u5236\u3001\u9759\u9ed8\u89c4\u5219<\/td>\n<\/tr>\n | \n| \u6570\u636e\u4e22\u5931<\/td>\n | \u78c1\u76d8\u5199\u6ee1\u5bfc\u81f4\u6570\u636e\u4e22\u5931<\/td>\n | \u76d1\u63a7\u78c1\u76d8\u4f7f\u7528\u7387，\u8bbe\u7f6e\u81ea\u52a8\u6e05\u7406\u7b56\u7565<\/td>\n<\/tr>\n | \n| \u914d\u7f6e\u5931\u6548<\/td>\n | \u4fee\u6539\u914d\u7f6e\u540e\u672a\u751f\u6548<\/td>\n | \u68c0\u67e5\u8bed\u6cd5，\u6267\u884c\u70ed\u91cd\u8f7d，\u67e5\u770b\u65e5\u5fd7<\/td>\n<\/tr>\n | \n| \u76ee\u6807\u5931\u8054<\/td>\n | Exporter \u65e0\u6cd5\u6293\u53d6<\/td>\n | \u68c0\u67e5\u7f51\u7edc\u8fde\u901a\u6027，\u9632\u706b\u5899\u89c4\u5219，\u670d\u52a1\u72b6\u6001<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n9.2 \u6545\u969c\u6392\u67e5\u6d41\u7a0b<\/h4>\n1. \u68c0\u67e5\u670d\u52a1\u72b6\u6001
\n systemctl status prometheus
\n docker-compose ps<\/p>\n 2. \u67e5\u770b\u670d\u52a1\u65e5\u5fd7
\n journalctl -u prometheus -f
\n docker-compose logs prometheus<\/p>\n 3. \u68c0\u67e5\u914d\u7f6e\u6587\u4ef6\u8bed\u6cd5
\n promtool check config prometheus.yml
\n promtool check rules rules\/*.yml<\/p>\n 4. \u9a8c\u8bc1 API \u5065\u5eb7
\n curl http:\/\/localhost:9090\/api\/v1\/status\/config<\/p>\n 5. \u68c0\u67e5\u76ee\u6807\u6293\u53d6
\n curl http:\/\/localhost:9090\/api\/v1\/targets<\/p>\n 6. \u6d4b\u8bd5 PromQL \u67e5\u8be2
\n curl 'http:\/\/localhost:9090\/api\/v1\/query?query=up'<\/p>\n 9.3 \u67b6\u6784\u4f18\u5316\u65b9\u6848<\/h4>\n\u5c0f\u578b\u73af\u5883（<500 \u76ee\u6807）<\/p>\n | | |