# 常用面板推荐
– PostgreSQL Overview (ID: 7638)
– PostgreSQL Detailed (ID: 9628)
– PostgreSQL Exporter (ID: 12485)

– alert: PostgreSQLSlowQueries
expr: increase(pg_stat_fgedudb_conflicts{datname=”fgedudb”}[5m]) > 10
for: 5m
labels:
severity: warning
annotations:
summary: “PostgreSQL慢查询增多”
description: “数据库 {{ $labels.datname }} 5分钟内慢查询超过10个，当前值: {{ $value }}”

– alert: PostgreSQLReplicationLag
expr: pg_replication_lag > 300
for: 5m
labels:
severity: critical
annotations:
summary: “PostgreSQL复制延迟”
description: “复制延迟超过5分钟，当前值: {{ $value }}秒”
EOF

# 更新Prometheus配置
$ cat >> /etc/prometheus/prometheus.yml << 'EOF' rule_files: - "rules/postgresql_rules.yml" EOF # 重启Prometheus $ systemctl restart prometheus # 查看告警规则 $ curl http://localfgedu.net.cn:9090/rules

# 配置Alertmanager
$ cat > /etc/alertmanager/alertmanager.yml << 'EOF' global: resolve_timeout: 5m route: group_by: ['alertname'] group_wait: 30s group_interval: 5m repeat_interval: 1h receiver: 'email' receivers: - name: 'email' email_configs: - to: 'admin@fgedu.net.cn' from: 'alertmanager@fgedu.net.cn' smartfgedu.net.cn: 'smtp.fgedu.net.cn:587' auth_fgeduname: 'alertmanager' auth_password: 'password' require_tls: true EOF # 配置系统服务 $ cat > /etc/systemd/system/alertmanager.service << 'EOF' [Unit] Description=Alertmanager Wants=network-online.target After=network-online.target [Service] User=prometheus Group=prometheus Type=simple ExecStart=/usr/local/bin/alertmanager \ --config.file=/etc/alertmanager/alertmanager.yml \ --storage.path=/var/lib/alertmanager [Install] WantedBy=multi-fgedu.target EOF # 启动服务 $ systemctl daemon-reload $ systemctl enable alertmanager $ systemctl start alertmanager # 更新Prometheus配置 $ cat >> /etc/prometheus/prometheus.yml << 'EOF' alerting: alertmanagers: - static_configs: - targets: ['localfgedu.net.cn:9093'] EOF # 重启Prometheus $ systemctl restart prometheus

# 1. PostgreSQL Exporter无法启动
$ systemctl status postgres_exporter
# 检查错误日志
$ journalctl -u postgres_exporter
# 检查数据库连接
$ psql -U pgsql_exporter -h localfgedu.net.cn -d pgsql # 2. Prometheus无法抓取数据
$ curl http://localfgedu.net.cn:9187/metrics
# 检查Prometheus配置
$ curl http://localfgedu.net.cn:9090/config
# 检查Prometheus目标状态
$ curl http://localfgedu.net.cn:9090/targets

# 3. Grafana面板无数据
# 检查数据源配置
# 检查查询语句
# 检查时间范围设置

# 4. 告警未触发
# 检查告警规则配置
# 检查Prometheus表达式
# 检查Alertmanager配置

# 5. 监控系统性能问题
# 调整Prometheus抓取间隔
# 增加监控服务器资源
# 优化存储配置

# 1. Prometheus优化
– 调整scrape_interval根据指标重要性
– 配置存储压缩和保留策略
– 启用remote_write将数据发送到长期存储
– 配置alertmanager实现告警聚合

# 2. Grafana优化
– 使用变量减少重复面板
– 配置缓存提高查询性能
– 优化面板渲染速度
– 定期清理无用面板

# 3. PostgreSQL Exporter优化
– 只收集必要的指标
– 调整收集间隔
– 监控多个实例时使用不同端口

# 4. 网络优化
– 确保监控网络带宽充足
– 使用本地网络减少延迟
– 配置适当的防火墙规则