Linux服务器巡检脚本整理

原创于 2025-10-30 11:10:29 发布 · 469 阅读

3 ·

CC 4.0 BY-SA版权

文章标签：

#服务器 #linux #运维

Linux 专栏收录该内容

9 篇文章

订阅专栏

日常巡检核心脚本

检查服务器CPU、内存、磁盘、网络、进程等关键指标，生成带颜色的报告，异常指标自动标黄/红。

#!/bin/bash

# -------------------------- 配置区 --------------------------
HOSTNAME=$(hostname)
DATE=$(date +"%Y-%m-%d %H:%M:%S")
# 报告保存路径（不用改，默认存/var/log）
REPORT_FILE="/var/log/system_info_check_$(date +%Y%m%d_%H%M%S).log"

# 告警阈值（可根据服务器配置调整）
CPU_WARNING=80       # CPU使用率告警阈值(%)
MEM_WARNING=85       # 内存使用率告警阈值(%)
DISK_WARNING=85      # 磁盘使用率告警阈值(%)
LOAD_WARNING=4       # 系统负载阈值（CPU核心数*2，比如4核设8）
INODE_WARNING=80     # Inode使用率告警阈值(%)
# -------------------------------------------------------------------

# 颜色定义
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m' # 重置颜色

# 日志函数
log() { echo "[$(date +"%Y-%m-%d %H:%M:%S")] $1" | tee -a "$REPORT_FILE"; }
log_section() { echo -e "\n============================================================" | tee -a "$REPORT_FILE"; echo " $1" | tee -a "$REPORT_FILE"; echo "============================================================" | tee -a "$REPORT_FILE"; }
log_warning() { echo -e "${YELLOW}[WARNING] $1${NC}" | tee -a "$REPORT_FILE"; }
log_error() { echo -e "${RED}[ERROR] $1${NC}" | tee -a "$REPORT_FILE"; }
log_ok() { echo -e "${GREEN}[OK] $1${NC}" | tee -a "$REPORT_FILE"; }

# 1. 系统基本信息
check_basic_info() {
    log_section "1. 系统基本信息"
    log "主机名: $HOSTNAME"
    log "检查时间: $DATE"
    log "系统版本: $(cat /etc/redhat-release 2>/dev/null || cat /etc/issue | head -1)"
    log "内核版本: $(uname -r)"
    log "系统架构: $(uname -m)"
    log "运行时长: $(uptime | awk -F'up ' '{print $2}' | awk -F',' '{print $1}')"
    log "当前用户: $(whoami)"
    log "登录用户数: $(who | wc -l)"
}

# 2. CPU检查
check_cpu() {
    log_section "2. CPU使用率检查"
    CPU_CORES=$(grep -c ^processor /proc/cpuinfo)
    log "CPU核心数: $CPU_CORES"
    
    # 获取CPU使用率（5秒平均值，更准确）
    CPU_IDLE=$(top -bn2 -d 1 | grep "Cpu(s)" | tail -1 | awk '{print $8}' | cut -d'%' -f1)
    CPU_USAGE=$(echo "scale=2; 100 - $CPU_IDLE" | bc)
    log "CPU使用率: ${CPU_USAGE}%"
    
    if (( $(echo "$CPU_USAGE > $CPU_WARNING" | bc -l) )); then
        log_warning "CPU使用率超过${CPU_WARNING}%，当前${CPU_USAGE}%"
        log "TOP 5 CPU消耗进程:"
        ps aux | sort -rn -k3 | head -5 | awk '{printf "  PID: %-8s User: %-10s CPU: %-6s CMD: %s\n", $2,$1,$3,$11}' | tee -a "$REPORT_FILE"
    else
        log_ok "CPU使用率正常: ${CPU_USAGE}%"
    fi
    
    # 系统负载检查
    LOAD_1=$(uptime | awk -F'load average:' '{print $2}' | awk -F',' '{print $1}' | xargs)
    LOAD_5=$(uptime | awk -F'load average:' '{print $2}' | awk -F',' '{print $2}' | xargs)
    LOAD_15=$(uptime | awk -F'load average:' '{print $2}' | awk -F',' '{print $3}' | xargs)
    log "系统负载: 1分钟=${LOAD_1}, 5分钟=${LOAD_5}, 15分钟=${LOAD_15}"
    
    # 负载阈值=CPU核心数*2（比如4核设8）
    LOAD_THRESHOLD=$(echo "$CPU_CORES * 2" | bc)
    if (( $(echo "$LOAD_1 > $LOAD_THRESHOLD" | bc -l) )); then
        log_warning "系统负载过高！1分钟负载${LOAD_1}超过阈值${LOAD_THRESHOLD}"
    fi
}

# 3. 内存检查
check_memory() {
    log_section "3. 内存使用检查"
    MEM_TOTAL=$(free -m | awk 'NR==2{print $2}')
    MEM_USED=$(free -m | awk 'NR==2{print $3}')
    MEM_AVAILABLE=$(free -m | awk 'NR==2{print $7}')
    MEM_USAGE=$(echo "scale=2; $MEM_USED / $MEM_TOTAL * 100" | bc)
    
    log "内存总量: ${MEM_TOTAL}MB"
    log "已用内存: ${MEM_USED}MB"
    log "可用内存: ${MEM_AVAILABLE}MB"
    log "内存使用率: ${MEM_USAGE}%"
    
    if (( $(echo "$MEM_USAGE > $MEM_WARNING" | bc -l) )); then
        log_warning "内存使用率超过${MEM_WARNING}%，当前${MEM_USAGE}%"
        log "TOP 5 内存消耗进程:"
        ps aux | sort -rn -k4 | head -5 | awk '{printf "  PID: %-8s User: %-10s MEM: %-6s CMD: %s\n", $2,$1,$4,$11}' | tee -a "$REPORT_FILE"
    else
        log_ok "内存使用率正常: ${MEM_USAGE}%"
    fi
    
    # Swap检查（超过100MB预警）
    SWAP_TOTAL=$(free -m | awk 'NR==3{print $2}')
    SWAP_USED=$(free -m | awk 'NR==3{print $3}')
    log "Swap总量: ${SWAP_TOTAL}MB"
    log "Swap使用: ${SWAP_USED}MB"
    if [ "$SWAP_TOTAL" -gt 0 ] && [ "$SWAP_USED" -gt 100 ]; then
        log_warning "Swap使用量较高: ${SWAP_USED}MB，可能存在内存压力"
    fi
}

# 4. 磁盘检查
check_disk() {
    log_section "4. 磁盘使用检查"
    log "磁盘分区使用情况:"
    df -h | grep -vE '^Filesystem|tmpfs|cdrom|loop' | awk '{print "  " $0}' | tee -a "$REPORT_FILE"
    
    # 磁盘使用率告警
    HAS_DISK_WARNING=0
    while read line; do
        USAGE=$(echo $line | awk '{print $5}' | sed 's/%//')
        MOUNT=$(echo $line | awk '{print $6}')
        if [ "$USAGE" -gt "$DISK_WARNING" ]; then
            log_warning "磁盘分区 $MOUNT 使用率${USAGE}%超过阈值${DISK_WARNING}%"
            HAS_DISK_WARNING=1
            log "  $MOUNT 分区占用空间最大的5个目录:"
            du -sh ${MOUNT}/* 2>/dev/null | sort -rh | head -5 | awk '{print "    " $0}' | tee -a "$REPORT_FILE"
        fi
    done < <(df -h | grep -vE '^Filesystem|tmpfs|cdrom|loop')
    
    if [ $HAS_DISK_WARNING -eq 0 ]; then
        log_ok "所有磁盘分区使用率正常"
    fi
    
    # Inode检查（避免小文件占满磁盘）
    log -e "\nInode使用情况:"
    df -i | grep -vE '^Filesystem|tmpfs|cdrom|loop' | awk '{print "  " $0}' | tee -a "$REPORT_FILE"
    while read line; do
        INODE_USAGE=$(echo $line | awk '{print $5}' | sed 's/%//')
        MOUNT=$(echo $line | awk '{print $6}')
        if [ "$INODE_USAGE" -gt "$INODE_WARNING" ]; then
            log_warning "分区 $MOUNT 的Inode使用率${INODE_USAGE}%超过阈值${INODE_WARNING}%"
        fi
    done < <(df -i | grep -vE '^Filesystem|tmpfs|cdrom|loop')
}

# 5. 网络检查
check_network() {
    log_section "5. 网络状态检查"
    log "网络接口状态:"
    ip -br addr | awk '{print "  " $0}' | tee -a "$REPORT_FILE"
    
    # TCP连接统计
    log -e "\nTCP连接状态统计:"
    netstat -an | awk '/^tcp/ {print $6}' | sort | uniq -c | sort -rn | awk '{print "  " $0}' | tee -a "$REPORT_FILE"
    
    # TIME_WAIT过多预警（超过5000）
    TIME_WAIT_COUNT=$(netstat -an | grep TIME_WAIT | wc -l)
    log "TIME_WAIT连接数: $TIME_WAIT_COUNT"
    if [ "$TIME_WAIT_COUNT" -gt 5000 ]; then
        log_warning "TIME_WAIT连接数过多: $TIME_WAIT_COUNT（可优化TCP参数）"
    fi
    
    # 监听端口检查
    log -e "\n当前监听端口:"
    netstat -tuln | grep LISTEN | awk '{print "  " $0}' | tee -a "$REPORT_FILE"
}

# 6. 进程和服务检查
check_processes() {
    log_section "6. 进程和服务检查"
    PROCESS_COUNT=$(ps aux | wc -l)
    log "当前进程总数: $PROCESS_COUNT"
    
    # 僵尸进程检查（死进程没清理）
    ZOMBIE_COUNT=$(ps aux | awk '{print $8}' | grep -c Z)
    log "僵尸进程数: $ZOMBIE_COUNT"
    if [ "$ZOMBIE_COUNT" -gt 0 ]; then
        log_warning "发现僵尸进程!"
        ps aux | grep 'Z' | grep -v grep | awk '{print "  PID: " $2 " PPID: " $3 " CMD: " $11}' | tee -a "$REPORT_FILE"
    fi
    
    # 关键服务检查（新手可加自己的服务，比如nginx/mysql）
    log -e "\n关键服务状态检查:"
    CRITICAL_SERVICES=("sshd" "crond" "rsyslog") # 可添加服务，如"nginx" "mysqld"
    for service in "${CRITICAL_SERVICES[@]}"; do
        if systemctl is-active --quiet $service 2>/dev/null; then
            log_ok "  $service: 运行中"
        else
            # 兼容非systemd系统（如CentOS 6）
            if ps aux | grep -v grep | grep -q $service; then
                log_ok "  $service: 运行中"
            else
                log_error "  $service: 未运行（需手动启动）"
            fi
        fi
    done
}

# 7. 系统日志检查
check_logs() {
    log_section "7. 系统日志检查"
    log "最近1小时系统错误日志:"
    if [ -f /var/log/messages ]; then
        ERROR_COUNT=$(grep -i "error\|fail\|critical" /var/log/messages | tail -20 | wc -l)
        if [ "$ERROR_COUNT" -gt 0 ]; then
            log_warning "发现 $ERROR_COUNT 条错误日志"
            grep -i "error\|fail\|critical" /var/log/messages | tail -10 | awk '{print "  " $0}' | tee -a "$REPORT_FILE"
        else
            log_ok "无严重错误日志"
        fi
    fi
    
    # OOM检查（内存溢出导致进程被杀）
    log -e "\nOOM(内存溢出)检查:"
    OOM_COUNT=$(dmesg | grep -i "out of memory" | wc -l)
    if [ "$OOM_COUNT" -gt 0 ]; then
        log_warning "发现 $OOM_COUNT 次OOM事件（需检查内存配置）"
        dmesg | grep -i "out of memory" | tail -5 | awk '{print "  " $0}' | tee -a "$REPORT_FILE"
    else
        log_ok "无OOM事件"
    fi
}

# 8. 报告摘要
generate_summary() {
    log_section "8. 巡检报告摘要"
    WARNING_COUNT=$(grep -c "\[WARNING\]" "$REPORT_FILE")
    ERROR_COUNT=$(grep -c "\[ERROR\]" "$REPORT_FILE")
    log "巡检完成时间: $(date +"%Y-%m-%d %H:%M:%S")"
    log "告警数量: $WARNING_COUNT"
    log "错误数量: $ERROR_COUNT"
    
    if [ "$ERROR_COUNT" -gt 0 ]; then
        log_error "发现 $ERROR_COUNT 个严重问题，请立即处理!"
    elif [ "$WARNING_COUNT" -gt 0 ]; then
        log_warning "发现 $WARNING_COUNT 个告警，建议关注"
    else
        log_ok "系统状态良好，无异常"
    fi
    log -e "\n完整报告已保存至: $REPORT_FILE"
}

# 主函数
main() {
    echo "=========================================="
    echo "    服务器健康状态巡检脚本"
    echo "=========================================="
    echo -e "\n"
    
    # 非root用户提示（部分检查需要root）
    if [ "$(id -u)" -ne 0 ]; then
        echo "警告: 非root用户运行，部分检查可能无法执行"
        echo -e "\n"
    fi
    
    # 执行所有检查
    check_basic_info
    check_cpu
    check_memory
    check_disk
    check_network
    check_processes
    check_logs
    generate_summary
    
    echo -e "\n=========================================="
    echo "    巡检完成!"
    echo "=========================================="
}

# 脚本入口
main "$@"

磁盘空间深度检查脚本

找出大文件、旧日志、临时文件，还能预测磁盘满的时间，给出清理建议。

#!/bin/bash

# -------------------------- 配置区 --------------------------
REPORT_FILE="/var/log/disk_check_$(date +%Y%m%d_%H%M%S).log"
DISK_WARNING=80       # 磁盘使用率告警阈值(%)
INODE_WARNING=80      # Inode使用率告警阈值(%)
LARGE_FILE_SIZE=1G    # 大文件阈值（可改500M、2G）
# -------------------------------------------------------------------

# 颜色定义
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m'

# 日志函数
log() { echo "[$(date +"%Y-%m-%d %H:%M:%S")] $1" | tee -a "$REPORT_FILE"; }
log_warning() { echo -e "${YELLOW}[WARNING] $1${NC}" | tee -a "$REPORT_FILE"; }
log_error() { echo -e "${RED}[ERROR] $1${NC}" | tee -a "$REPORT_FILE"; }
log_ok() { echo -e "${GREEN}[OK] $1${NC}" | tee -a "$REPORT_FILE"; }

# 1. 磁盘使用率检查
check_disk_usage() {
    log "==========================================================="
    log "1. 磁盘使用率检查"
    log "==========================================================="
    df -h | grep -vE '^Filesystem|tmpfs|cdrom|loop' | while read line; do
        USAGE=$(echo $line | awk '{print $5}' | sed 's/%//')
        MOUNT=$(echo $line | awk '{print $6}')
        AVAIL=$(echo $line | awk '{print $4}')
        log "分区: $MOUNT"
        log "  使用率: ${USAGE}%"
        log "  可用空间: $AVAIL"
        
        if [ "$USAGE" -gt "$DISK_WARNING" ]; then
            log_error "  磁盘使用率超过${DISK_WARNING}%！"
            # 显示TOP10大目录
            log "  占用空间TOP 10目录:"
            du -sh ${MOUNT}/* 2>/dev/null | sort -rh | head -10 | awk '{print "    " $0}' | tee -a "$REPORT_FILE"
            # 显示超阈值的大文件
            log "  大文件(>$LARGE_FILE_SIZE):"
            find ${MOUNT} -type f -size +${LARGE_FILE_SIZE} -exec ls -lh {} \; 2>/dev/null | awk '{print "    " $9 " (" $5 ")"}' | head -10 | tee -a "$REPORT_FILE"
        else
            log_ok "  使用率正常"
        fi
        log ""
    done
}

# 2. Inode检查（小文件占满磁盘用）
check_inode_usage() {
    log "==========================================================="
    log "2. Inode使用率检查"
    log "==========================================================="
    df -i | grep -vE '^Filesystem|tmpfs|cdrom|loop' | while read line; do
        INODE_USAGE=$(echo $line | awk '{print $5}' | sed 's/%//')
        MOUNT=$(echo $line | awk '{print $6}')
        log "分区: $MOUNT"
        log "  Inode使用率: ${INODE_USAGE}%"
        
        if [ "$INODE_USAGE" -gt "$INODE_WARNING" ]; then
            log_error "  Inode使用率超过${INODE_WARNING}%！（小文件太多）"
            # 显示文件数TOP10目录
            log "  文件数量TOP 10目录:"
            for dir in $(find ${MOUNT}/* -maxdepth 0 -type d 2>/dev/null); do
                echo "$(find $dir -type f 2>/dev/null | wc -l) $dir"
            done | sort -rn | head -10 | awk '{print "    " $1 " files: " $2}' | tee -a "$REPORT_FILE"
        else
            log_ok "  Inode使用率正常"
        fi
        log ""
    done
}

# 3. 日志文件检查
check_log_files() {
    log "==========================================================="
    log "3. 日志文件大小检查"
    log "==========================================================="
    # 常见日志目录（可加自己的日志目录，如/opt/logs）
    LOG_DIRS=("/var/log" "/opt/logs" "/data/logs" "/app/logs")
    for log_dir in "${LOG_DIRS[@]}"; do
        if [ -d "$log_dir" ]; then
            log "检查目录: $log_dir"
            # 显示超100MB的日志
            log "  大日志文件(>100MB):"
            find $log_dir -type f -size +100M -exec ls -lh {} \; 2>/dev/null | awk '{print "    " $9 " (" $5 ")"}' | head -10 | tee -a "$REPORT_FILE"
            # 日志总大小
            TOTAL_SIZE=$(du -sh $log_dir 2>/dev/null | awk '{print $1}')
            log "  总大小: $TOTAL_SIZE"
            log ""
        fi
    done
    
    # 30天前未压缩的旧日志
    log "未压缩的旧日志文件(30天前):"
    find /var/log -type f -name "*.log" -mtime +30 ! -name "*.gz" -exec ls -lh {} \; 2>/dev/null | awk '{print "  " $9 " (" $5 ")"}' | head -10 | tee -a "$REPORT_FILE"
}

# 4. 临时文件检查
check_temp_files() {
    log "==========================================================="
    log "4. 临时文件检查"
    log "==========================================================="
    # /tmp目录检查
    if [ -d "/tmp" ]; then
        TMP_SIZE=$(du -sh /tmp 2>/dev/null | awk '{print $1}')
        TMP_FILE_COUNT=$(find /tmp -type f 2>/dev/null | wc -l)
        log "/tmp目录:"
        log "  总大小: $TMP_SIZE"
        log "  文件数: $TMP_FILE_COUNT"
        
        # 7天未使用的临时文件
        OLD_TMP=$(find /tmp -type f -atime +7 2>/dev/null | wc -l)
        if [ "$OLD_TMP" -gt 0 ]; then
            log_warning "  发现 $OLD_TMP 个超过7天未访问的临时文件"
            log "  占用空间TOP 10:"
            find /tmp -type f -atime +7 -exec ls -lh {} \; 2>/dev/null | sort -k5 -rh | head -10 | awk '{print "    " $9 " (" $5 ")"}' | tee -a "$REPORT_FILE"
        fi
    fi
    
    # /var/tmp检查
    if [ -d "/var/tmp" ]; then
        VAR_TMP_SIZE=$(du -sh /var/tmp 2>/dev/null | awk '{print $1}')
        log -e "\n/var/tmp目录:"
        log "  总大小: $VAR_TMP_SIZE"
    fi
}

# 5. 磁盘IO检查
check_disk_io() {
    log "==========================================================="
    log "5. 磁盘IO统计"
    log "==========================================================="
    if command -v iostat &> /dev/null; then
        log "磁盘IO统计(最近1分钟):"
        iostat -x 1 2 | tail -n +4 | awk 'NF' | tail -n +2 | tee -a "$REPORT_FILE"
    else
        log_warning "未安装iostat工具（执行：yum install sysstat 安装）"
    fi
    
    # IO等待过高预警（超过20%）
    IO_WAIT=$(top -bn1 | grep "Cpu(s)" | awk '{print $10}' | sed 's/%wa,//')
    log -e "\n当前IO等待: ${IO_WAIT}%"
    if (( $(echo "$IO_WAIT > 20" | bc -l) )); then
        log_warning "IO等待过高！（可能是磁盘性能不足）"
    fi
}

# 6. 清理建议
generate_cleanup_suggestions() {
    log "==========================================================="
    log "6. 磁盘清理建议（新手复制命令执行）"
    log "==========================================================="
    log "1. 压缩30天前的旧日志:"
    log "   find /var/log -type f -name '*.log' -mtime +30 -exec gzip {} \;"
    log -e "\n2. 清理7天前的临时文件:"
    log "   find /tmp -type f -atime +7 -delete"
    log -e "\n3. 清理yum缓存（CentOS）:"
    log "   yum clean all"
    log -e "\n4. 清理Docker未使用资源（如有Docker）:"
    log "   docker system prune -a"
    log -e "\n5. 清理journal日志（系统日志）:"
    log "   journalctl --vacuum-time=7d"
}

# 主函数
main() {
    echo "=========================================="
    echo "    磁盘空间深度检查脚本 "
    echo "=========================================="
    echo -e "\n"
    
    check_disk_usage
    check_inode_usage
    check_log_files
    check_temp_files
    check_disk_io
    generate_cleanup_suggestions
    
    log -e "\n==========================================================="
    log "检查完成!"
    log "报告保存至: $REPORT_FILE"
    log "==========================================================="
}

main "$@"

网络连接检查脚本

检查端口监听、TCP连接状态（比如TIME_WAIT过多）、网络连通性（DNS/网关/外网）。

#!/bin/bash

# -------------------------- 配置区 --------------------------
REPORT_FILE="/var/log/network_check_$(date +%Y%m%d_%H%M%S).log"
TIME_WAIT_WARNING=5000  # TIME_WAIT超过5000预警
ESTABLISHED_WARNING=10000 # 活动连接超过10000预警

CRITICAL_PORTS=(22 80 443 3306 6379 8080) # 待检查端口（如MySQL:3306，Redis:6379）
# -------------------------------------------------------------------

# 颜色和日志函数
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m'
log() { echo "[$(date +"%Y-%m-%d %H:%M:%S")] $1" | tee -a "$REPORT_FILE"; }
log_warning() { echo -e "${YELLOW}[WARNING] $1${NC}" | tee -a "$REPORT_FILE"; }
log_ok() { echo -e "${GREEN}[OK] $1${NC}" | tee -a "$REPORT_FILE"; }

# 关键端口检查
check_listening_ports() {
    log "==========================================================="
    log "3. 关键端口检查"
    log "==========================================================="
    log "当前监听端口:"
    netstat -tulnp | grep LISTEN | tee -a "$REPORT_FILE"
    log -e "\n关键端口状态:"
    for port in "${CRITICAL_PORTS[@]}"; do
        if netstat -tuln | grep -q ":$port "; then
            PID=$(netstat -tulnp 2>/dev/null | grep ":$port " | awk '{print $7}' | cut -d'/' -f1)
            PROCESS=$(netstat -tulnp 2>/dev/null | grep ":$port " | awk '{print $7}' | cut -d'/' -f2)
            log_ok "端口 $port: 监听中（进程:$PROCESS PID:$PID）"
        else
            log_warning "端口 $port: 未监听（服务可能未启动）"
        fi
    done
}

main() {
    echo "=========================================="
    echo "    网络状态检查脚本 "
    echo "=========================================="
    check_network_interfaces
    check_tcp_connections
    check_listening_ports
    check_network_connectivity
    log -e "\n检查完成! 报告: $REPORT_FILE"
}
main "$@"