1. 系统架构

2. 基础环境部署

2.1 硬件要求

2.2 软件依赖

# CentOS 8 基础环境
yum install -y epel-release
yum install -y httpd php gcc glibc glibc-common gd gd-devel \
               net-snmp net-snmp-utils python3-devel

3. Nagios Core 安装

3.1 编译安装

wget https://assets.nagios.com/downloads/nagioscore/releases/nagios-4.4.12.tar.gz
tar xzf nagios-4.4.12.tar.gz
cd nagios-4.4.12

./configure --with-command-group=nagcmd \
            --with-httpd-conf=/etc/httpd/conf.d
make all
make install-groups-users
make install
make install-daemoninit
make install-commandmode
make install-config
make install-webconf

3.2 Web界面配置

# 设置管理员密码
htpasswd -c /usr/local/nagios/etc/htpasswd.users nagiosadmin

# 防火墙规则
firewall-cmd --add-port=80/tcp --permanent
firewall-cmd --reload

# 启动服务
systemctl restart httpd
systemctl enable nagios

4. 各层实现

4.1 NRPE 客户端配置（被监控节点）

# 安装NRPE
yum install -y nrpe nagios-plugins-all

# 配置允许的Nagios服务器IP
echo "allowed_hosts=192.168.1.100" >> /etc/nagios/nrpe.cfg

# 示例：添加自定义内存检查
echo "command[check_mem]=/usr/lib64/nagios/plugins/check_memory -w 80 -c 90" >> /etc/nagios/nrpe.cfg

systemctl restart nrpe

Nagios 服务端配置

# 定义NRPE检查命令
echo '
define command{
    command_name check_nrpe
    command_line $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$
}' >> /usr/local/nagios/etc/objects/commands.cfg

# 定义主机服务
echo '
define service{
    use                 generic-service
    host_name           web01
    service_description Memory Usage
    check_command       check_nrpe!check_mem
}' >> /usr/local/nagios/etc/services.cfg

4.2 Elasticsearch 集成

# 安装ELK Stack
yum install -y elasticsearch kibana logstash

# Nagios事件转发配置（/etc/logstash/conf.d/nagios.conf）
input {
  file {
    path => "/usr/local/nagios/var/nagios.log"
    start_position => "beginning"
  }
}
output {
  elasticsearch {
    hosts => ["localhost:9200"]
    index => "nagios-events-%{+YYYY.MM.dd}"
  }
}

Neo4j 关系存储

# 事件关系分析脚本（/opt/nagios/event_analyzer.py）
from py2neo import Graph

graph = Graph("bolt://localhost:7687", auth=("neo4j", "password"))

def create_relation(host, service, symptom):
    query = """
    MERGE (h:Host {name: $host})
    MERGE (s:Service {name: $service})
    MERGE (sym:Symptom {desc: $symptom})
    MERGE (h)-[r:AFFECTS]->(s)
    MERGE (s)-[c:CAUSES]->(sym)
    """
    graph.run(query, host=host, service=service, symptom=symptom)

4.3 Ansible 自愈脚本

# /etc/ansible/playbooks/restart_httpd.yml
- hosts: "{{ target_host }}"
  tasks:
    - name: Check httpd status
      command: systemctl status httpd
      register: httpd_status
      ignore_errors: yes

    - name: Restart httpd if failed
      systemd:
        name: httpd
        state: restarted
      when: "'active (running)' not in httpd_status.stdout"

Nagios 事件处理器

# /usr/local/nagios/etc/objects/eventhandlers.cfg
define command {
    command_name handle_web_failure
    command_line /usr/local/nagios/libexec/eventhandlers/web_failure.sh "$HOSTADDRESS#34; "$SERVICESTATE#34;
}

# 自动修复脚本示例（web_failure.sh）
#!/bin/bash
ansible-playbook /etc/ansible/playbooks/restart_httpd.yml -e "target_host=$1"
echo "`date` - Restarted httpd on $1" >> /var/log/nagios_autofix.log

5. 验证与调优

5.1 配置检查

/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg

5.2 性能优化参数

# /usr/local/nagios/etc/nagios.cfg
check_result_reaper_frequency=2
max_concurrent_checks=30
service_check_timeout=60

6. 维护与监控

6.1 日常维护

# 日志轮转配置（/etc/logrotate.d/nagios）
/usr/local/nagios/var/nagios.log {
    daily
    rotate 30
    compress
    missingok
    postrotate
        /bin/kill -HUP `cat /usr/local/nagios/var/nagios.lock 2>/dev/null` 2>/dev/null
    endscript
}

6.2 监控看板

Grafana 配置：

• ini复制下载

[inputs.influxdb]
urls = ["http://localhost:8086"]
database = "nagios_metrics"

附录A：故障排查指南