# Nginx 配置示例：基礎 upstream 配置
upstream backend {
    # 服務器權重配置
    server 127.0.0.1:9000 weight=3 max_fails=2 fail_timeout=30s;
    server 127.0.0.1:9001 weight=2 max_fails=2 fail_timeout=30s;
    server 127.0.0.1:9002 weight=1 max_fails=2 fail_timeout=30s backup;
    
    # 負載均衡算法
    least_conn;
    
    # 健康檢查配置
    keepalive 32;
    keepalive_requests 100;
    keepalive_timeout 60s;
}

server {
    listen 80;
    server_name example.com;
    
    location / {
        proxy_pass http://backend;
        
        # 關鍵超時參數(shù)
        proxy_connect_timeout 5s;
        proxy_send_timeout 60s;
        proxy_read_timeout 60s;
        
        # 錯誤處理
        proxy_next_upstream error timeout invalid_header http_500 http_502 http_503;
        proxy_next_upstream_tries 3;
        proxy_next_upstream_timeout 10s;
    }
}

# 自定義日志格式，包含 upstream 詳細信息
log_format upstream_log '$remote_addr - $remote_user [$time_local] '
                       '"$request" $status $body_bytes_sent '
                       '"$http_referer" "$http_user_agent" '
                       'rt=$request_time uct="$upstream_connect_time" '
                       'uht="$upstream_header_time" urt="$upstream_response_time" '
                       'uaddr="$upstream_addr" ustatus="$upstream_status"';

server {
    access_log /var/log/nginx/upstream.log upstream_log;
    error_log /var/log/nginx/error.log debug;
}

#!/bin/bash
# upstream_analyzer.sh - Nginx upstream 日志分析腳本

LOG_FILE="/var/log/nginx/upstream.log"
ANALYSIS_PERIOD="1h"  # 分析最近1小時的日志

echo "=== Nginx Upstream 502 錯誤分析報告 ==="
echo "分析時間段: 最近 $ANALYSIS_PERIOD"
echo "日志文件: $LOG_FILE"
echo

# 統(tǒng)計 502 錯誤總數(shù)
error_502_count=$(grep " 502 " "$LOG_FILE" | wc -l)
echo "502 錯誤總數(shù): $error_502_count"

# 分析 502 錯誤的 upstream 響應時間分布
echo -e "\n=== 502 錯誤響應時間分析 ==="
grep " 502 " "$LOG_FILE" | \
awk '{
    # 提取 upstream_response_time
    match($0, /urt="([^"]*)"/, arr)
    if (arr[1] != "-") {
        time = arr[1]
        if (time < 1) bucket="<1s"
        else if (time < 5) bucket="1-5s" 
        else if (time < 10) bucket="5-10s"
        else if (time < 30) bucket="10-30s"
        else bucket=">30s"
        count[bucket]++
    }
}
END {
    for (b in count) {
        printf "%-8s: %d 次\n", b, count[b]
    }
}'

# 分析最頻繁出現(xiàn) 502 的后端服務器
echo -e "\n=== 502 錯誤后端服務器分布 ==="
grep " 502 " "$LOG_FILE" | \
awk '{
    match($0, /uaddr="([^"]*)"/, arr)
    if (arr[1] != "-") {
        servers[arr[1]]++
    }
}
END {
    for (server in servers) {
        printf "%-20s: %d 次\n", server, servers[server]
    }
}' | sort -k2 -nr

# 分析 502 錯誤的時間分布
echo -e "\n=== 502 錯誤時間分布（按小時） ==="
grep " 502 " "$LOG_FILE" | \
awk '{
    # 提取時間戳中的小時
    match($0, /\[([^\]]+)\]/, arr)
    split(arr[1], datetime, ":")
    hour = datetime[2]
    hours[hour]++
}
END {
    for (h in hours) {
        printf "%s:00 - %d 次\n", h, hours[h]
    }
}' | sort

; /etc/php/8.1/fpm/pool.d/www.conf
; PHP-FPM 進程池配置優(yōu)化

[www]
; 進程管理器類型
pm = dynamic

; 進程數(shù)量配置
pm.max_children = 50          ; 最大子進程數(shù)
pm.start_servers = 10         ; 啟動時的進程數(shù)
pm.min_spare_servers = 5      ; 最小空閑進程數(shù)
pm.max_spare_servers = 15     ; 最大空閑進程數(shù)

; 進程生命周期管理
pm.max_requests = 1000        ; 每個進程處理的最大請求數(shù)
pm.process_idle_timeout = 60s ; 空閑進程超時時間

; 超時配置 - 關鍵參數(shù)
request_timeout = 300s        ; 單個請求超時時間
request_terminate_timeout = 300s ; 強制終止超時時間

; 慢日志配置
slowlog = /var/log/php8.1-fpm-slow.log
request_slowlog_timeout = 10s ; 慢查詢閾值

; 狀態(tài)監(jiān)控
pm.status_path = /fpm-status
ping.path = /fpm-ping
ping.response = pong

; 安全配置
security.limit_extensions = .php .php3 .php4 .php5 .php7 .php8

; 環(huán)境變量
env[HOSTNAME] = $HOSTNAME
env[PATH] = /usr/local/bin:/usr/bin:/bin
env[TMP] = /tmp
env[TMPDIR] = /tmp
env[TEMP] = /tmp

location ~ \.php$ {
    fastcgi_pass unix:/var/run/php/php8.1-fpm.sock;
    fastcgi_index index.php;
    
    # FastCGI 超時參數(shù) - 核心配置
    fastcgi_connect_timeout 60s;     # 連接超時
    fastcgi_send_timeout 300s;       # 發(fā)送超時
    fastcgi_read_timeout 300s;       # 讀取超時
    
    # 緩沖區(qū)配置
    fastcgi_buffer_size 64k;         # 響應頭緩沖區(qū)
    fastcgi_buffers 4 64k;           # 響應體緩沖區(qū)
    fastcgi_busy_buffers_size 128k;  # 忙碌緩沖區(qū)大小
    
    # 臨時文件配置
    fastcgi_temp_file_write_size 128k;
    fastcgi_max_temp_file_size 256m;
    
    # 錯誤處理
    fastcgi_intercept_errors on;
    fastcgi_ignore_client_abort off;
    
    # 參數(shù)傳遞
    fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
    fastcgi_param QUERY_STRING $query_string;
    fastcgi_param REQUEST_METHOD $request_method;
    fastcgi_param CONTENT_TYPE $content_type;
    fastcgi_param CONTENT_LENGTH $content_length;
    
    # 自定義參數(shù)
    fastcgi_param HTTP_X_REAL_IP $remote_addr;
    fastcgi_param HTTP_X_FORWARDED_FOR $proxy_add_x_forwarded_for;
    fastcgi_param HTTP_X_FORWARDED_PROTO $scheme;
}

#!/bin/bash
# timeout_optimizer.sh - 根據(jù)業(yè)務負載動態(tài)調整超時參數(shù)

# 配置文件路徑
NGINX_CONF="/etc/nginx/sites-available/default"
PHP_FPM_CONF="/etc/php/8.1/fpm/pool.d/www.conf"

# 獲取當前系統(tǒng)負載
get_system_load() {
    local load_1min=$(uptime | awk -F'load average:' '{print $2}' | awk -F',' '{print $1}' | tr -d ' ')
    echo "$load_1min"
}

# 獲取 PHP-FPM 進程狀態(tài)
get_fpm_status() {
    local active_processes=$(curl -s http://localhost/fpm-status | grep "active processes" | awk '{print $3}')
    local total_processes=$(curl -s http://localhost/fpm-status | grep "total processes" | awk '{print $3}')
    echo "$active_processes/$total_processes"
}

# 分析最近的 502 錯誤率
analyze_502_rate() {
    local error_count=$(tail -1000 /var/log/nginx/access.log | grep " 502 " | wc -l)
    local total_requests=$(tail -1000 /var/log/nginx/access.log | wc -l)
    local error_rate=$(echo "scale=4; $error_count / $total_requests * 100" | bc)
    echo "$error_rate"
}

# 動態(tài)調整超時參數(shù)
adjust_timeouts() {
    local load=$(get_system_load)
    local error_rate=$(analyze_502_rate)
    
    echo "當前系統(tǒng)負載: $load"
    echo "當前 502 錯誤率: $error_rate%"
    
    # 根據(jù)負載和錯誤率調整參數(shù)
    if (( $(echo "$load > 2.0" | bc -l) )) || (( $(echo "$error_rate > 5.0" | bc -l) )); then
        echo "檢測到高負載或高錯誤率，增加超時時間..."
        
        # 備份原配置
        cp "$NGINX_CONF" "${NGINX_CONF}.backup.$(date +%Y%m%d_%H%M%S)"
        
        # 調整 Nginx 超時參數(shù)
        sed -i 's/fastcgi_read_timeout [^;]*/fastcgi_read_timeout 600s/' "$NGINX_CONF"
        sed -i 's/fastcgi_send_timeout [^;]*/fastcgi_send_timeout 600s/' "$NGINX_CONF"
        
        # 調整 PHP-FPM 超時參數(shù)
        sed -i 's/request_timeout = [^;]*/request_timeout = 600s/' "$PHP_FPM_CONF"
        
        # 重載配置
        nginx -s reload
        systemctl reload php8.1-fpm
        
        echo "超時參數(shù)已調整為 600s"
        
    elif (( $(echo "$load < 0.5" | bc -l) )) && (( $(echo "$error_rate < 1.0" | bc -l) )); then
        echo "系統(tǒng)負載較低，恢復默認超時時間..."
        
        # 恢復默認配置
        sed -i 's/fastcgi_read_timeout [^;]*/fastcgi_read_timeout 300s/' "$NGINX_CONF"
        sed -i 's/fastcgi_send_timeout [^;]*/fastcgi_send_timeout 300s/' "$NGINX_CONF"
        sed -i 's/request_timeout = [^;]*/request_timeout = 300s/' "$PHP_FPM_CONF"
        
        # 重載配置
        nginx -s reload
        systemctl reload php8.1-fpm
        
        echo "超時參數(shù)已恢復為 300s"
    else
        echo "系統(tǒng)狀態(tài)正常，保持當前配置"
    fi
}

# 主函數(shù)
main() {
    echo "=== Nginx FastCGI 超時參數(shù)動態(tài)優(yōu)化 ==="
    echo "執(zhí)行時間: $(date)"
    echo
    
    adjust_timeouts
    
    echo
    echo "優(yōu)化完成，建議繼續(xù)監(jiān)控系統(tǒng)狀態(tài)"
}

# 執(zhí)行主函數(shù)
main

# nginx-exporter.yml - Nginx 監(jiān)控配置
global:
  scrape_interval: 15s
  evaluation_interval: 15s

rule_files:
  - "nginx_rules.yml"

scrape_configs:
  - job_name: 'nginx'
    static_configs:
      - targets: ['localhost:9113']
    scrape_interval: 5s
    metrics_path: /metrics

  - job_name: 'php-fpm'
    static_configs:
      - targets: ['localhost:9253']
    scrape_interval: 5s

alerting:
  alertmanagers:
    - static_configs:
        - targets:
          - alertmanager:9093

# nginx_rules.yml - 告警規(guī)則配置
groups:
  - name: nginx_alerts
    rules:
      - alert: Nginx502ErrorHigh
        expr: rate(nginx_http_requests_total{status="502"}[5m]) > 0.1
        for: 2m
        labels:
          severity: critical
        annotations:
          summary: "Nginx 502 錯誤率過高"
          description: "502 錯誤率在過去 5 分鐘內超過 10%"

      - alert: FastCGITimeoutHigh
        expr: nginx_http_request_duration_seconds{quantile="0.95"} > 30
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "FastCGI 響應時間過長"
          description: "95% 的請求響應時間超過 30 秒"

      - alert: PHPFPMProcessesHigh
        expr: phpfpm_active_processes / phpfpm_total_processes > 0.8
        for: 3m
        labels:
          severity: warning
        annotations:
          summary: "PHP-FPM 進程使用率過高"
          description: "活躍進程數(shù)占總進程數(shù)的 80% 以上"

#!/bin/bash
# auto_recovery.sh - 502 錯誤自動恢復腳本

LOG_FILE="/var/log/nginx/access.log"
ERROR_THRESHOLD=10  # 5分鐘內502錯誤超過10次觸發(fā)恢復
TIME_WINDOW=300     # 時間窗口：5分鐘

# 檢查 502 錯誤頻率
check_502_frequency() {
    local current_time=$(date +%s)
    local start_time=$((current_time - TIME_WINDOW))
    
    # 統(tǒng)計時間窗口內的 502 錯誤
    local error_count=$(awk -v start="$start_time" '
    {
        # 解析時間戳
        gsub(/\[|\]/, "", $4)
        cmd = "date -d \"" $4 "\" +%s"
        cmd | getline timestamp
        close(cmd)
        
        if (timestamp >= start && $9 == "502") {
            count++
        }
    }
    END {
        print count + 0
    }' "$LOG_FILE")
    
    echo "$error_count"
}

# 重啟 PHP-FPM 服務
restart_php_fpm() {
    echo "[$(date)] 檢測到大量 502 錯誤，重啟 PHP-FPM 服務..."
    
    # 記錄當前進程狀態(tài)
    echo "重啟前 PHP-FPM 狀態(tài):" >> /var/log/auto_recovery.log
    systemctl status php8.1-fpm >> /var/log/auto_recovery.log
    
    # 優(yōu)雅重啟
    systemctl reload php8.1-fpm
    
    # 等待服務穩(wěn)定
    sleep 10
    
    # 驗證服務狀態(tài)
    if systemctl is-active --quiet php8.1-fpm; then
        echo "[$(date)] PHP-FPM 重啟成功" >> /var/log/auto_recovery.log
        
        # 發(fā)送通知
        curl -X POST "https://api.telegram.org/bot$TELEGRAM_BOT_TOKEN/sendMessage" \
             -d chat_id="$TELEGRAM_CHAT_ID" \
             -d text="?? 自動恢復：PHP-FPM 服務已重啟，502 錯誤應該得到緩解"
    else
        echo "[$(date)] PHP-FPM 重啟失敗" >> /var/log/auto_recovery.log
        
        # 發(fā)送緊急通知
        curl -X POST "https://api.telegram.org/bot$TELEGRAM_BOT_TOKEN/sendMessage" \
             -d chat_id="$TELEGRAM_CHAT_ID" \
             -d text="?? 緊急：PHP-FPM 自動重啟失敗，需要人工介入"
    fi
}

# 清理臨時文件和緩存
cleanup_temp_files() {
    echo "[$(date)] 清理臨時文件和緩存..."
    
    # 清理 PHP session 文件
    find /var/lib/php/sessions -name "sess_*" -mtime +1 -delete
    
    # 清理 Nginx 臨時文件
    find /var/cache/nginx -type f -mtime +1 -delete
    
    # 清理應用緩存（根據(jù)實際情況調整）
    if [ -d "/var/www/html/cache" ]; then
        find /var/www/html/cache -name "*.cache" -mtime +1 -delete
    fi
    
    echo "[$(date)] 臨時文件清理完成" >> /var/log/auto_recovery.log
}

# 主監(jiān)控循環(huán)
main_monitor() {
    while true; do
        local error_count=$(check_502_frequency)
        
        if [ "$error_count" -gt "$ERROR_THRESHOLD" ]; then
            echo "[$(date)] 檢測到 $error_count 個 502 錯誤，啟動自動恢復..."
            
            # 執(zhí)行恢復操作
            cleanup_temp_files
            restart_php_fpm
            
            # 等待恢復生效
            sleep 60
        fi
        
        # 每30秒檢查一次
        sleep 30
    done
}

# 啟動監(jiān)控
echo "[$(date)] 啟動 502 錯誤自動恢復監(jiān)控..."
main_monitor

#!/bin/bash
# system_optimization.sh - 系統(tǒng)級性能優(yōu)化腳本

# 內核參數(shù)優(yōu)化
optimize_kernel_params() {
    echo "優(yōu)化內核參數(shù)..."
    
    cat >> /etc/sysctl.conf << EOF
# 網(wǎng)絡連接優(yōu)化
net.core.somaxconn = 65535
net.core.netdev_max_backlog = 5000
net.ipv4.tcp_max_syn_backlog = 65535
net.ipv4.tcp_fin_timeout = 10
net.ipv4.tcp_keepalive_time = 1200
net.ipv4.tcp_max_tw_buckets = 6000

# 文件描述符限制
fs.file-max = 2097152
fs.nr_open = 2097152

# 內存管理
vm.swappiness = 10
vm.dirty_ratio = 15
vm.dirty_background_ratio = 5
EOF
    
    sysctl -p
}

# 文件描述符限制
optimize_file_limits() {
    echo "優(yōu)化文件描述符限制..."
    
    cat >> /etc/security/limits.conf << EOF
* soft nofile 65535
* hard nofile 65535
* soft nproc 65535
* hard nproc 65535
nginx soft nofile 65535
nginx hard nofile 65535
www-data soft nofile 65535
www-data hard nofile 65535
EOF
}

# 執(zhí)行優(yōu)化
optimize_kernel_params
optimize_file_limits

echo "系統(tǒng)優(yōu)化完成，建議重啟系統(tǒng)使所有配置生效"