Shell脚本如何配置容器监控分析

wen 实用脚本 2026-07-16 1

本文目录导读：

Shell脚本如何配置容器监控分析

基础容器监控脚本
高级监控分析脚本
告警与报告生成
主监控循环
使用说明
配置参数说明

我来详细介绍如何配置容器监控分析的Shell脚本。

基础容器监控脚本

#!/bin/bash
# 容器监控分析脚本
# 功能：监控Docker容器资源使用、健康状态和性能分析
# 配置参数
LOG_DIR="/var/log/container-monitor"
INTERVAL=60  # 采集间隔（秒）
ALERT_CPU=80  # CPU告警阈值(%)
ALERT_MEM=90  # 内存告警阈值(%)
ALERT_DISK=85 # 磁盘告警阈值(%)
# 创建日志目录
mkdir -p "$LOG_DIR"
# 时间戳函数
timestamp() {
    date "+%Y-%m-%d %H:%M:%S"
}
# 获取容器列表
get_containers() {
    docker ps --format "{{.ID}}:{{.Names}}:{{.Image}}:{{.Status}}"
}
# 监控容器CPU使用率
monitor_cpu() {
    local container_id=$1
    local container_name=$2
    # 使用docker stats获取CPU使用率
    local cpu_usage=$(docker stats --no-stream --format "{{.CPUPerc}}" "$container_id" | sed 's/%//')
    echo "$cpu_usage"
}
# 监控容器内存使用
monitor_memory() {
    local container_id=$1
    local container_name=$2
    # 获取内存使用率
    local mem_stats=$(docker stats --no-stream --format "{{.MemPerc}}" "$container_id" | sed 's/%//')
    local mem_usage=$(docker stats --no-stream --format "{{.MemUsage}}" "$container_id")
    local mem_limit=$(docker stats --no-stream --format "{{.MemUsage}}" "$container_id")
    echo "$mem_stats|$mem_usage"
}
# 监控容器网络IO
monitor_network() {
    local container_id=$1
    local container_name=$2
    # 获取网络IO
    local network_io=$(docker stats --no-stream --format "{{.NetIO}}" "$container_id")
    echo "$network_io"
}
# 监控容器磁盘IO
monitor_disk() {
    local container_id=$1
    local container_name=$2
    # 获取块IO
    local block_io=$(docker stats --no-stream --format "{{.BlockIO}}" "$container_id")
    echo "$block_io"
}
# 容器健康检查
health_check() {
    local container_id=$1
    local container_name=$2
    # 检查容器运行状态
    local status=$(docker inspect --format '{{.State.Status}}' "$container_id")
    # 检查健康检查状态（如果有配置）
    local health=""
    if docker inspect --format '{{.State.Health.Status}}' "$container_id" 2>/dev/null; then
        health=$(docker inspect --format '{{.State.Health.Status}}' "$container_id")
    else
        health="not_configured"
    fi
    # 检查重启次数
    local restart_count=$(docker inspect --format '{{.RestartCount}}' "$container_id")
    echo "$status|$health|$restart_count"
}

高级监控分析脚本

#!/bin/bash
# 高级容器分析功能
# 日志分析函数
analyze_container_logs() {
    local container_name=$1
    local lines=${2:-100}
    echo "=== 容器日志分析: $container_name ==="
    # 获取最近N行日志
    recent_logs=$(docker logs --tail "$lines" "$container_name" 2>&1)
    # 统计错误级别
    errors=$(echo "$recent_logs" | grep -ci "error\|exception\|fail")
    warnings=$(echo "$recent_logs" | grep -ci "warn\|warning")
    info=$(echo "$recent_logs" | grep -ci "info\|debug")
    echo "Error count: $errors"
    echo "Warning count: $warnings"
    echo "Info count: $info"
    # 提取错误模式
    if [ $errors -gt 0 ]; then
        echo "=== 最近错误 ==="
        echo "$recent_logs" | grep -i "error\|exception\|fail" | tail -10
    fi
}
# 资源趋势分析
analyze_resource_trends() {
    local container_name=$1
    local duration=${2:-3600}  # 默认1小时
    echo "=== 资源趋势分析: $container_name ==="
    # 创建时间序列数据
    local data_file="$LOG_DIR/${container_name}_trend_$(date +%Y%m%d).csv"
    # 采集数据点
    for ((i=0; i<duration; i+=60)); do
        local cpu=$(docker stats --no-stream --format "{{.CPUPerc}}" "$container_name" | sed 's/%//')
        local mem=$(docker stats --no-stream --format "{{.MemPerc}}" "$container_name" | sed 's/%//')
        echo "$(timestamp),$cpu,$mem" >> "$data_file"
        sleep 60
    done &
    echo "数据采集正在后台运行，文件: $data_file"
}
# 容器间通信分析
analyze_network_connections() {
    local container_name=$1
    echo "=== 网络连接分析: $container_name ==="
    # 获取容器IP
    local container_ip=$(docker inspect --format '{{range .NetworkSettings.Networks}}{{.IPAddress}}{{end}}' "$container_name")
    # 检查网络连接
    local connections=$(docker exec "$container_name" ss -tunp 2>/dev/null || docker exec "$container_name" netstat -tunp 2>/dev/null)
    echo "Container IP: $container_ip"
    echo "Active Connections:"
    echo "$connections" | grep -v "Local Address" | head -20
    # 检查监听端口
    echo "Listening Ports:"
    docker port "$container_name" 2>/dev/null
}
# 依赖分析
analyze_dependencies() {
    local container_name=$1
    echo "=== 依赖分析: $container_name ==="
    # 获取容器卷挂载
    echo "Volume Mounts:"
    docker inspect --format '{{range .Mounts}}{{.Source}} -> {{.Destination}}{{"\n"}}{{end}}' "$container_name"
    # 获取网络连接
    echo "Network Links:"
    docker inspect --format '{{range $net,$v := .NetworkSettings.Networks}}{{$.Name}} -> {{$net}}{{"\n"}}{{end}}' "$container_name"
    # 获取环境变量依赖
    echo "Environment Dependencies:"
    docker inspect --format '{{range .Config.Env}}{{.}}{{"\n"}}{{end}}' "$container_name" | grep -i "host\|url\|endpoint\|connection" | head -10
}

告警与报告生成

#!/bin/bash
# 告警和报告功能
# 告警检查函数
check_alerts() {
    local container_name=$1
    local cpu_usage=$2
    local mem_usage=$3
    local alerts=""
    # CPU告警
    if (( $(echo "$cpu_usage > $ALERT_CPU" | bc -l) )); then
        alerts+="CPU使用率过高: ${cpu_usage}% (阈值: ${ALERT_CPU}%) "
    fi
    # 内存告警
    if (( $(echo "$mem_usage > $ALERT_MEM" | bc -l) )); then
        alerts+="内存使用率过高: ${mem_usage}% (阈值: ${ALERT_MEM}%) "
    fi
    if [ -n "$alerts" ]; then
        send_alert "$container_name" "$alerts"
    fi
}
# 发送告警
send_alert() {
    local container=$1
    local message=$2
    echo "[$(timestamp)] ALERT: $container - $message" >> "$LOG_DIR/alerts.log"
    # 可配置的告警方式
    # 1. 系统日志
    logger "Container Monitor Alert: $container - $message"
    # 2. 邮件告警（如果需要）
    # echo "$message" | mail -s "Container Alert: $container" admin@example.com
    # 3. Webhook（如果需要）
    # curl -X POST -H "Content-Type: application/json" \
    #      -d "{\"container\":\"$container\",\"message\":\"$message\"}" \
    #      http://your-webhook-url
}
# 生成报告
generate_report() {
    local container_name=$1
    local report_file="$LOG_DIR/report_${container_name}_$(date +%Y%m%d_%H%M%S).html"
    cat > "$report_file" << HTML
<!DOCTYPE html>
<html>
<head>容器监控报告 - $container_name</title>
    <style>
        body { font-family: Arial, sans-serif; margin: 20px; }
        .header { background: #4CAF50; color: white; padding: 20px; }
        .metric { margin: 20px 0; padding: 10px; border: 1px solid #ddd; }
        .warning { background-color: #fff3cd; }
        .error { background-color: #f8d7da; }
        table { width: 100%; border-collapse: collapse; }
        th, td { padding: 8px; text-align: left; border-bottom: 1px solid #ddd; }
    </style>
</head>
<body>
    <div class="header">
        <h1>容器监控报告</h1>
        <p>容器: $container_name</p>
        <p>时间: $(date)</p>
    </div>
    <div class="metric">
        <h2>当前状态</h2>
        <table>
            <tr><th>指标</th><th>值</th><th>状态</th></tr>
HTML
    # 获取当前状态
    local cpu=$(docker stats --no-stream --format "{{.CPUPerc}}" "$container_name")
    local mem=$(docker stats --no-stream --format "{{.MemPerc}}" "$container_name")
    local status=$(docker inspect --format '{{.State.Status}}' "$container_name")
    # 添加到报告
    cat >> "$report_file" << HTML
            <tr><td>CPU使用率</td><td>$cpu</td><td>$(echo "$cpu > 80" | bc -l) > 0 ? "⚠️" : "✓"</td></tr>
            <tr><td>内存使用率</td><td>$mem</td><td>$(echo "$mem > 90" | bc -l) > 0 ? "⚠️" : "✓"</td></tr>
            <tr><td>运行状态</td><td>$status</td><td>✓</td></tr>
        </table>
    </div>
    <div class="metric">
        <h2>性能历史</h2>
        <p>最近30分钟的资源使用趋势</p>
        <!-- 这里可以集成图表库 -->
    </div>
</body>
</html>
HTML
    echo "报告已生成: $report_file"
}

主监控循环

#!/bin/bash
# 主监控循环
main() {
    echo "启动容器监控系统..."
    echo "日志目录: $LOG_DIR"
    echo "采集间隔: ${INTERVAL}秒"
    echo "告警阈值 - CPU: ${ALERT_CPU}%, 内存: ${ALERT_MEM}%"
    # 初始化日志文件
    local log_file="$LOG_DIR/monitor_$(date +%Y%m%d).log"
    echo "时间,容器名称,CPU(%),内存(%),网络IO,磁盘IO,状态,健康检查" > "$log_file"
    while true; do
        local containers=$(get_containers)
        while IFS= read -r container; do
            if [ -z "$container" ]; then
                continue
            fi
            # 解析容器信息
            local container_id=$(echo "$container" | cut -d: -f1)
            local container_name=$(echo "$container" | cut -d: -f2)
            local container_image=$(echo "$container" | cut -d: -f3)
            local container_status=$(echo "$container" | cut -d: -f4)
            # 采集指标
            local cpu=$(monitor_cpu "$container_id" "$container_name")
            local mem_data=$(monitor_memory "$container_id" "$container_name")
            local mem_percent=$(echo "$mem_data" | cut -d| -f1)
            local net_io=$(monitor_network "$container_id" "$container_name")
            local disk_io=$(monitor_disk "$container_id" "$container_name")
            local health=$(health_check "$container_id" "$container_name")
            # 记录到日志
            echo "$(timestamp),$container_name,$cpu,$mem_percent,$net_io,$disk_io,$container_status,$health" >> "$log_file"
            # 检查告警
            check_alerts "$container_name" "$cpu" "$mem_percent"
            # 显示实时状态
            printf "%-20s CPU: %5s%% 内存: %5s%% 状态: %-10s\n" \
                   "$container_name" "$cpu" "$mem_percent" "$container_status"
        done <<< "$containers"
        sleep "$INTERVAL"
    done
}
# 启动监控
main

使用说明

安装配置

#!/bin/bash
# 安装脚本
install_monitor() {
    # 1. 创建监控目录
    mkdir -p /usr/local/bin/container-monitor
    # 2. 复制脚本
    chmod +x monitor.sh
    cp monitor.sh /usr/local/bin/container-monitor/
    # 3. 创建systemd服务
    cat > /etc/systemd/system/container-monitor.service << SERVICE
[Unit]
Description=Container Monitor Service
After=docker.service
Requires=docker.service
[Service]
Type=simple
ExecStart=/usr/local/bin/container-monitor/monitor.sh
Restart=always
RestartSec=10
[Install]
WantedBy=multi-user.target
SERVICE
    # 4. 创建日志轮转配置
    cat > /etc/logrotate.d/container-monitor << LOGROTATE
/var/log/container-monitor/*.log {
    daily
    rotate 30
    compress
    delaycompress
    missingok
    notifempty
    create 0644 root root
}
LOGROTATE
    # 5. 启用服务
    systemctl daemon-reload
    systemctl enable container-monitor
    systemctl start container-monitor
    echo "安装完成！"
}

配置参数说明

参数	说明	默认值
LOG_DIR	日志存储目录	/var/log/container-monitor
INTERVAL	采集间隔（秒）	60
ALERT_CPU	CPU告警阈值(%)	80
ALERT_MEM	内存告警阈值(%)	90
ALERT_DISK	磁盘告警阈值(%)	85

这个脚本系统提供了完整的容器监控、分析和告警功能,可根据实际需求进行调整和扩展。