Prometheus+Grafana+Alertmanager监控部署
-
1.环境准备
- Prometheus部署
-
node_export部署
编辑Prometheus.yml
Grafana安装部署
部署Alertmanger
环境准备
阿里云ECS实例*1 4C 8G 系统盘100G 数据盘 300G
Prometheus+Grafana+Alermanager
开启防火墙对应端口 :
# systemctl start firewalld
# firewall-cmd --zone=pubic --add-port=9090/tcp --premanent #Prometheus端口
# firewall-cmd --zone=pubic --add-port=3000/tcp --premanent #grafana看板端口
# firewall-cmd --zone=pubic --add-port=9100/tcp --premanent #node-export 端口
# firewall-cmd --zone=pubic --add-port=9093/tcp --premanent # Alermanager 端口
# firewall-cmd --zone=pubic --add-port=9094/tcp --premanent # Alermanager 端口
# firewall-cmd --reload #重载防火墙,生效
————————————————
Prometheus部署
下载Prometheus次新稳定版本 : https://prometheus.io/download/
Prometheus基础环境配置
# yum -y install golang nodejs yarn ## yum安装golang不需要可以不需要再/etc/profile配置环境变量
如果使用编译安装,则需要在/etc/profile 设置环境变量,示例:
export GOROOT=/usr/lib/golang
export GOPATH=/root/Work/programmer/go/gopath/
export PATH=$PATH:$GOROOT/bin:$GOPATH/bin
2、解压Prometheus的tar包,我这里下载的是2.26
# tar -zxvf prometheus-2.26.0.linux-amd64.tar.gz
# mv prometheus-2.26.0.linux-amd64 /usr/local/
3、运行Prometheus有两种方式
1)使用nohup ./prometheus >> out.log &
但是使用nohup 如果需要关闭或重启Prometheus,需要ps -aux | grep prometheus 查询Prometheus进程的PID ,手动kill -9 进行关闭进程
2)编写system文件
# vim /usr/lib/systemd/system/prometheus.service
[Unit]
Description=Prometheus
Documentation=https://prometheus.io/
After=network.target
[Service]
Type=simple
User=prometheus
ExecStart=/usr/local/prometheus-2.26.0.linux-amd64/prometheus --config.file=/usr/local/prometheus-2.26.0.linux-amd64/prometheus.yml --storage.tsdb.path=/var/lib/prometheus --storage.tsdb.retention=180d --web.enable-admin-api
# storage.tsdb.path手动指定Prometheus的存储位置 config.file指定Prometheus的配置文件 ExecStart指定Prometheus运行文件
Restart=on-failure
[Install]
WantedBy=multi-user.target
配置完system文件之后
systemctl start prometheus
systemctl enable prometheus
node-export 部署
开启Prometheus node端
下载 node_exporter-1.1.2.linux-amd64.tar.gz Linux监控node端
下载地址: https://github.com/prometheus/node_exporter
# tar -zxvf node_exporter-1.1.2.linux-amd64.tar.gz
# mv /usr/local/node_exporter-1.1.2.linux-amd64
# cd /usr/local/node_exporter-1.1.2.linux-amd64
# nohup ./node_exporter >> out.log &
编辑Prometheus.yml
编辑Prometheus.yml 文件
# vim /usr/local/prometheus-2.26.0.linux-amd64/prometheus.yml
scrape_configs:
- job_name: 'prometheus'
static_configs:
- targets: ['localhost:9090']
- job_name: Linux_export
static_configs:
- targets: ['localhost:9100']
之后重启Prometheus
systemctl restart prometheus
访问 localhost:9090 查看Prometheus状态
Grafana 安装部署
下载grafana安装包
https://grafana.com/grafana/dow
yum localinstall grafana-7.3.6-1.x86_64.rpm
# systemctl start grafana
# systemctl enable grafana
启动grafana后,访问
localhost:3000 (默认端口3000) 查看grfana面板
grafana默认密码 admin admin 第一次登陆需要修改密码
部署Alertmanger 告警
下载alermanager告警组件
https://prometheus.io/docs/alerting/latest/alertmanager/
这边用的是0.20.0版本
# tar -zxvf alertmanager-0.20.0.linux-amd64.tar.gz
# mv alertmanager-0.20.0.linux-amd64 /usr/local/alertmanager
# vim /usr/local/alertmanager/alertmanager.yml
global:
resolve_timeout: 5m
smtp_smarthost: 'smtp.163.com:465'
smtp_from: 'xxxxxxxxx@163.com'
smtp_auth_username: 'xxxxxxxx@163.com'
smtp_auth_password: 'xxxxxxxxxxxxxxx'
smtp_require_tls: false
templates:
- '/alertmanager/template/*.tmpl'
route:
group_by: ['alertname']
group_wait: 30s
group_interval: 5m
repeat_interval: 10m
receiver: 'mail'
receivers:
- name: 'mail'
email_configs:
- to: 'xxxxxxxxx@xxx.com'
inhibit_rules:
- source_match:
severity: 'critical'
target_match:
severity: 'warning'
equal: ['alertname', 'dev', 'instance']
编辑system文件
# vim /usr/lib/systemd/system/alertmanager.service
[Unit]
Description=alertmanager
Documentation=https://github.com/prometheus/alertmanager
After=network.target
[Service]
Type=simple
User=prometheus
ExecStart=/usr/local/alertmanager/alertmanager --config.file=/usr/local/alertmanager/alertmanager.yml
Restart=on-failure
[Install]
WantedBy=multi-user.target
systemctl start alermanage
systemctl enable alermanage
编辑Prometheus.yml 文件
# vim /usr/local/prometheus-2.26.0.linux-amd64/prometheus.yml
alerting:
alertmanagers:
- static_configs:
- targets:
- localhost:9093 #指定本地alermanage端口
'evaluation_interval'.
rule_files:
- "/usr/local/prometheus-2.26.0.linux-amd64/rule/*.yml" #在Prometheus下定制的规则将用于告警
之后再Prometheus目录下,创建一个rule文件夹,将定义规则的 .yml文件 放入rule文件夹中 实现告警
以下以告警组件状态为例:
# cd /usr/local/prometheus-2.26.0.linux-amd64/rule/
# vim nodestatus_rule.yml
groups:
- name: alert-rules.yml
rules:
- alert: InstanceStatus
expr: up == 0
for: 10s
labels:
severity: "critical"
annotations:
summary: "主机 {{ $labels.instance }} 无响应"
description: "{{ $labels.instance }} 无响应 (当前值: {{ $value }}) "
之后重启Prometheus,就可以正常告警了
systemctl restart prometheus
如果想验证告警是否成功,可以手动关闭 node-export 验证是否有告警邮件发出
ps -aux | grep node
kill -9 PID
感谢观看,您的点赞是对我最大的支持
后续会陆续更新如何监控nginx、tomcat、MySQL、Oracle
以后会继续更新ansible相关教程。