2.Prometheus邮件报警配置

2024-01-06 22:11:22

1、安装配置 Alertmanager

wget https://github.com/prometheus/alertmanager/releases/download/v0.20.0/alertmanager-0.20.0.linux-amd64.tar.gz
tar -zxv -f alertmanager-0.20.0.linux-amd64.tar.gz -C /usr/local
cd /usr/local
mv alertmanager-0.20.0.linux-amd64/ alertmanager

2，创建启动文件

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/alert-test.yml --storage.path=/usr/local/alertmanager/data
Restart=on-failure

[Install]
WantedBy=multi-user.target

Alertmanager 安装目录下默认有 alertmanager.yml 配置文件，可以创建新的配置文件，在启动时指定即可。

cd /usr/local/alertmanager
vim alert-test.yml
global:
  smtp_smarthost: 'smtp.qiye.aliyun.com:25'
  smtp_from: 'jump@tongchuangkeji.net'
  smtp_auth_username: 'jump@tongchuangkeji.net'
  smtp_auth_password: 'xxxx'
  smtp_require_tls: false
  
templates:
  - '/alertmanager/template/*.tmpl'
  
route:
  group_by: ['alertname', 'cluster', 'service']
  group_wait: 30s
  group_interval: 5m
  repeat_interval: 10m
  receiver: default-receiver
  
receivers:
- name: 'default-receiver'
  email_configs:
  - to: 'liqilong@edspay.com'
    html: ''
    headers: {Subject: "[WARN] 报警邮件 test"}

邮箱一开始使用的是公司的邮箱，结果在后边验证的时候，总是会报错level=error ts=2019-01-26T06:21:59.062483579Z caller=notify.go:332 component=dispatcher msg="Error on notify" err="*smtp.plainAuth failed: unencrypted connection"，也在这里看了一些人踩坑的报告，试验了25、465、587端口，发现均无效果，最后改成163邮箱，直接就生效了。

smtp_smarthost：是用于发送邮件的邮箱的 SMTP 服务器地址+端口；
smtp_auth_password：是发送邮箱的授权码而不是登录密码；
smtp_require_tls：不设置的话默认为 true，当为 true 时会有 starttls 错误，为了简单这里设置为 false；
templates：指出邮件的模板路径；
receivers 下 html 指出邮件内容模板名，这里模板名为 “alert.html”，在模板路径中的某个文件中定义。
headers：为邮件标题；

使用阿里云企业邮箱无法发送邮件，报错如下：

Jun 11 13:40:52 worker alertmanager: level=error ts=2020-06-11T05:40:52.638Z caller=notify.go:372 component=dispatcher msg="Error on notify" err="*smtp.plainAuth auth: unencrypted connection" context_err="context deadline exceeded"
Jun 11 13:40:52 worker alertmanager: level=error ts=2020-06-11T05:40:52.638Z caller=dispatch.go:301 component=dispatcher msg="Notify for alerts failed" num_alerts=1 err="*smtp.plainAuth auth: unencrypted connection"

3，配置告警规则

配置 rule.yml。

cd /usr/local/prometheus
vim rule.yml
groups:
- name: alert-rules.yml
  rules:
  - alert: InstanceStatus # alert 名字
    expr: up{job="192.168.75.10"} == 0 # 判断条件,job是指prometheus.yml文件中的job_name
    for: 10s # 条件保持 10s 才会发出 alter
    labels: # 设置 alert 的标签
      severity: "critical"
    annotations:  # alert 的其他标签，但不用于标识 alert
      description: 服务器  已当机超过 20s
      summary: 服务器  运行状态

在 prometheus.yml 中指定 rule.yml 的路径

cat prometheus.yml

# my global config
global:
  scrape_interval:     15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
  # scrape_timeout is set to the global default (10s).

# Alertmanager configuration
alerting:
  alertmanagers:
  - static_configs:
    - targets:
      - localhost:9093 # 这里修改为 localhost，# 新增

# Load rules once and periodically evaluate them according to the global 'evaluation_interval'.
rule_files:
  # - "first_rules.yml"
  # - "second_rules.yml"
  - "/usr/local/prometheus/rule.yml" # 新增

# A scrape configuration containing exactly one endpoint to scrape:
# Here it's Prometheus itself.
scrape_configs:
  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
  - job_name: '192.168.75.11'

    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.

    static_configs:
    - targets: ['localhost:9090','localhost:9100']

  - job_name: '192.168.75.10'
    scrape_interval: 5s
    static_configs:
    - targets: ['192.168.75.10:9100']

重启 Prometheus 服务：

chown -R prometheus.prometheus /usr/local/prometheus/rule.yml
systemctl restart prometheus

4，编写邮件模板

注意：文件后缀为 tmpl

mkdir -pv /alertmanager/template/ # 路径跟上面的alertmanager.yml 配置文件保持一致
vim /alertmanager/template/alert.tmpl
<table>
	<tr><td>报警名</td><td>开始时间</td></tr>
	<tr><td></td><td></td></tr>
</table>

注意：启动的时候报错如下：

Jun 11 12:55:44 worker alertmanager: level=error ts=2020-06-11T04:55:44.744Z caller=main.go:236 msg="Unable to create data directory" err="mkdir data/: permission denied"
Jun 11 12:55:44 worker systemd: alertmanager.service: main process exited, code=exited, status=1/FAILURE
Jun 11 12:55:44 worker systemd: Unit alertmanager.service entered failed state.
Jun 11 12:55:44 worker systemd: alertmanager.service failed.
Jun 11 12:55:44 worker systemd: alertmanager.service holdoff time over, scheduling restart.
Jun 11 12:55:44 worker systemd: Stopped alertmanager.

这是因为在新版本中默认情况下存储路径 --storage.path 是相对目录 data/，但是prometheus用户在该路径下没权限创建目录，所以导致启动报错

解决办法：在alertmanager.service文件中指定默认存储路径在当前路径下即可

5，启动 Alertmanager

chown -R prometheus.prometheus /usr/local/alertmanager
systemctl daemon-reload
systemctl start alertmanager.service
systemctl status alertmanager.service
ss -tnl|grep 9093

6，验证效果

此时到管理界面可以看到如下信息：

然后停止192.168.75.10节点上的 node_exporter 服务，然后再看效果。

接着邮箱应该会收到邮件：
略

码农公寓