# check_http 常用参数
check_http -H host [-I IP] [-p port] [-u URL] [-S] [-e code] [-s string]
           [-a user:pass] [-A auth] [-P post] [-m min:max]
           [-w warn] [-c crit] [-t timeout] [--sni] [-C days]

# HTTP 基本检查
check_http -H www.example.com

# HTTPS 检查
check_http -H www.example.com -S -p 443

# 指定 URL 路径
check_http -H www.example.com -u /api/health

# 响应码检查
check_http -H www.example.com -e 200,301,302

# 内容检查
check_http -H www.example.com -s "Welcome"

# 响应时间阈值
check_http -H www.example.com -w 5 -c 10

# SSL 证书过期检查（提前 30 天警告，7 天严重）
check_http -H www.example.com -S -C 30,7

# Basic 认证
check_http -H www.example.com -a user:password

# POST 请求
check_http -H www.example.com -P "key=value" -u /api/submit

# 虚拟主机
check_http -H www.example.com -I 192.168.1.100 --sni

# 磁盘使用率
check_disk -w 20% -c 10% -p /
check_disk -w 20% -c 10% -p /home -p /var -p /tmp
check_disk -w 1000 -c 500 -p /  # 使用 MB 阈值

# CPU 负载（1/5/15 分钟）
check_load -w 5,4,3 -c 10,8,6

# 用户数
check_users -w 5 -c 10

# 进程数
check_procs -w 150 -c 200
check_procs -w 5:10 -c 1:20 -s httpd  # 指定进程名和范围

# Swap 使用率
check_swap -w 50% -c 25%

# 系统运行时间
check_uptime

# 文件句柄
check_file_handles -w 80% -c 95%

# MySQL 检查
check_mysql -H localhost -u monitoring -p password -d database_name
check_mysql_query -H localhost -u monitoring -p password -q "SELECT COUNT(*) FROM users" -w 100 -c 200

# PostgreSQL 检查
check_pgsql -H localhost -l monitoring -p 5432

# Oracle 检查（需要 check_oracle_health 插件）
check_oracle_health --connect database --mode connection-time --user monitor --password secret

# SNMP 获取单个值
check_snmp -H 192.168.1.1 -C public -o sysUpTime.0

# SNMP 带阈值
check_snmp -H 192.168.1.1 -C public -o ifInOctets.1 -w 1000000 -c 2000000

# SNMP 接口状态
check_snmp -H 192.168.1.1 -C public -o ifOperStatus.1 -s 1

# SNMP CPU 使用率（OID 因厂商而异）
check_snmp -H 192.168.1.1 -C public -o 1.3.6.1.4.1.2021.11.9.0 -w 80 -c 95

# SNMP 内存使用率
check_snmp -H 192.168.1.1 -C public -o 1.3.6.1.4.1.2021.4.6.0 -w 80 -c 95

# 方式一：手动下载
cd /usr/local/nagios/libexec/
wget https://example.com/check_custom
chmod +x check_custom

# 方式二：使用包管理器
yum install -y nagios-plugins-all  # EPEL 源

# 方式三：编译安装
cd /tmp
git clone https://github.com/nagios-plugins/nagios-plugins.git
cd nagios-plugins
./tools/setup
./configure
make && sudo make install

#!/bin/bash
# check_custom.sh - 自定义检查插件模板
# 版本: 1.0
# 作者: DevOps Team
# 描述: 检查自定义指标

PROGNAME=$(basename $0)
PROGPATH=$(dirname $0)
VERSION="1.0"

# 退出码
OK=0
WARNING=1
CRITICAL=2
UNKNOWN=3

# 默认参数
HOST=""
METRIC=""
WARN=""
CRIT=""

# 使用帮助
print_usage() {
    echo "Usage: $PROGNAME -H <host> -M <metric> -w <warning> -c <critical>"
    echo "       $PROGNAME -h"
}

print_help() {
    print_usage
    echo ""
    echo "Check custom metric for a host."
    echo ""
    echo "Options:"
    echo "  -H    Host name or IP address"
    echo "  -M    Metric to check (cpu|memory|disk)"
    echo "  -w    Warning threshold"
    echo "  -c    Critical threshold"
    echo "  -h    Print this help"
    echo ""
    echo "Examples:"
    echo "  $PROGNAME -H 192.168.1.1 -M cpu -w 80 -c 95"
    echo "  $PROGNAME -H 192.168.1.1 -M memory -w 80 -c 95"
}

# 参数解析
while getopts "H:M:w:c:h" opt; do
    case $opt in
        H) HOST=$OPTARG ;;
        M) METRIC=$OPTARG ;;
        w) WARN=$OPTARG ;;
        c) CRIT=$OPTARG ;;
        h) print_help; exit $UNKNOWN ;;
        *) print_usage; exit $UNKNOWN ;;
    esac
done

# 参数验证
if [ -z "$HOST" ] || [ -z "$METRIC" ] || [ -z "$WARN" ] || [ -z "$CRIT" ]; then
    echo "UNKNOWN: Missing required parameters"
    print_usage
    exit $UNKNOWN
fi

# 获取指标值
case $METRIC in
    cpu)
        VALUE=$(check_cpu_usage $HOST)
        UNIT="%"
        ;;
    memory)
        VALUE=$(check_memory_usage $HOST)
        UNIT="%"
        ;;
    disk)
        VALUE=$(check_disk_usage $HOST)
        UNIT="%"
        ;;
    *)
        echo "UNKNOWN: Unknown metric '$METRIC'"
        exit $UNKNOWN
        ;;
esac

# 检查指标是否获取成功
if [ -z "$VALUE" ]; then
    echo "UNKNOWN: Unable to get $METRIC value for $HOST"
    exit $UNKNOWN
fi

# 阈值判断
PERFDATA="$METRIC=${VALUE}${UNIT};${WARN};${CRIT}"

if [ $(echo "$VALUE > $CRIT" | bc -l) -eq 1 ]; then
    echo "CRITICAL: $METRIC=${VALUE}${UNIT} (threshold=${CRIT}${UNIT}) | $PERFDATA"
    exit $CRITICAL
elif [ $(echo "$VALUE > $WARN" | bc -l) -eq 1 ]; then
    echo "WARNING: $METRIC=${VALUE}${UNIT} (threshold=${WARN}${UNIT}) | $PERFDATA"
    exit $WARNING
else
    echo "OK: $METRIC=${VALUE}${UNIT} | $PERFDATA"
    exit $OK
fi

#!/usr/bin/env python3
# check_custom.py - Python 自定义插件模板

import sys
import argparse
from enum import IntEnum

class State(IntEnum):
    OK = 0
    WARNING = 1
    CRITICAL = 2
    UNKNOWN = 3

class NagiosPlugin:
    def __init__(self, name, version, description):
        self.name = name
        self.version = version
        self.description = description
        self.args = None
        self.perfdata = []
        self.messages = {State.OK: [], State.WARNING: [], State.CRITICAL: [], State.UNKNOWN: []}

    def add_args(self):
        parser = argparse.ArgumentParser(description=self.description)
        parser.add_argument('-H', '--host', required=True, help='Host address')
        parser.add_argument('-w', '--warning', type=float, help='Warning threshold')
        parser.add_argument('-c', '--critical', type=float, help='Critical threshold')
        parser.add_argument('-t', '--timeout', type=int, default=30, help='Timeout in seconds')
        parser.add_argument('-v', '--verbose', action='store_true', help='Verbose output')
        return parser

    def add_perfdata(self, label, value, uom='', warn='', crit='', min_val='', max_val=''):
        perf = f"{label}={value}{uom};{warn};{crit};{min_val};{max_val}"
        self.perfdata.append(perf)

    def add_message(self, state, message):
        self.messages[state].append(message)

    def check(self):
        """Override this method with actual check logic"""
        raise NotImplementedError

    def exit(self, state):
        all_messages = self.messages[state]
        if not all_messages:
            all_messages = self.messages[State.UNKNOWN]

        output = f"{state.name}: {'; '.join(all_messages)}"
        if self.perfdata:
            output += f" | {' '.join(self.perfdata)}"

        print(output)
        sys.exit(state)

    def run(self):
        parser = self.add_args()
        self.args = parser.parse_args()

        try:
            self.check()
        except Exception as e:
            print(f"UNKNOWN: {e}")
            sys.exit(State.UNKNOWN)


# 使用示例
class CheckHTTP(NagiosPlugin):
    def __init__(self):
        super().__init__("check_http_custom", "1.0", "Custom HTTP check")

    def check(self):
        import requests
        import time

        try:
            start = time.time()
            response = requests.get(
                f"http://{self.args.host}",
                timeout=self.args.timeout
            )
            elapsed = time.time() - start
        except requests.exceptions.Timeout:
            self.add_message(State.CRITICAL, f"Connection timeout after {self.args.timeout}s")
            self.exit(State.CRITICAL)
        except Exception as e:
            self.add_message(State.CRITICAL, f"Connection failed: {e}")
            self.exit(State.CRITICAL)

        self.add_perfdata("time", f"{elapsed:.3f}", "s", 
                         self.args.warning or '', self.args.critical or '', "0")

        if response.status_code >= 500:
            self.add_message(State.CRITICAL, f"HTTP {response.status_code}")
            self.exit(State.CRITICAL)
        elif response.status_code >= 400:
            self.add_message(State.WARNING, f"HTTP {response.status_code}")
            self.exit(State.WARNING)

        if self.args.critical and elapsed >= self.args.critical:
            self.add_message(State.CRITICAL, f"Response time = {elapsed:.3f}s")
            self.exit(State.CRITICAL)
        elif self.args.warning and elapsed >= self.args.warning:
            self.add_message(State.WARNING, f"Response time = {elapsed:.3f}s")
            self.exit(State.WARNING)

        self.add_message(State.OK, f"HTTP {response.status_code}, Response time = {elapsed:.3f}s")
        self.exit(State.OK)


if __name__ == '__main__':
    plugin = CheckHTTP()
    plugin.run()

# 直接执行插件测试
/usr/local/nagios/libexec/check_ping -H 192.168.1.1 -w 100,20% -c 500,60%

# 查看退出码
echo $?

# 查看输出
/usr/local/nagios/libexec/check_http -H www.example.com -w 5 -c 10

# 详细输出模式（部分插件支持）
/usr/local/nagios/libexec/check_disk -v -w 20% -c 10% -p /

# 使用 strace 跟踪插件系统调用
strace -f /usr/local/nagios/libexec/check_http -H www.example.com

# 使用 timeout 测试超时处理
timeout 5 /usr/local/nagios/libexec/check_http -H slow-server.example.com

# 检查插件依赖
ldd /usr/local/nagios/libexec/check_http

# 测试 NRPE 远程插件
/usr/local/nagios/libexec/check_nrpe -H remote-host -c check_disk -a '-w 20% -c 10% -p /'