Prometheus+Alertmanager配置邮件报警

(1)安装Alertmanager

安装在k8s-node2服务器上

[root@k8s-node2 ~]# tar -zxvf alertmanager-0.21.0.linux-amd64.tar.gz

alertmanager-0.21.0.linux-amd64/

alertmanager-0.21.0.linux-amd64/alertmanager

alertmanager-0.21.0.linux-amd64/amtool

alertmanager-0.21.0.linux-amd64/NOTICE

alertmanager-0.21.0.linux-amd64/LICENSE

alertmanager-0.21.0.linux-amd64/alertmanager.yml

[root@k8s-node2 ~]# ls

alertmanager-0.21.0.linux-amd64 alertmanager-0.21.0.linux-amd64.tar.gz anaconda-ks.cfg disk_free.sh node_exporter-1.0.1.linux-amd64.tar.gz

[root@k8s-node2 ~]# cp alertmanager-0.21.0.linux-amd64/alertmanager /usr/local/bin/

[root@k8s-node2 ~]# cp alertmanager-0.21.0.linux-amd64/amtool /usr/local/bin/

[root@k8s-node2 ~]# alertmanager --version

(2)配置Alertmanager

[root@k8s-node2 ~]# mkdir -pv /etc/alertmanager

mkdir: 已创建目录 "/etc/alertmanager"

[root@k8s-node2 ~]# cp alertmanager-0.21.0.linux-amd64/alertmanager.yml /etc/alertmanager/

[root@k8s-node2 ~]# cat /etc/alertmanager/alertmanager.yml 默认配置文件

（3）重新配置Alertmanager

[root@k8s-node2 alertmanager]# cp alertmanager.yml alertmanager.yml.bak

[root@k8s-node2 alertmanager]# ls

alertmanager.yml alertmanager.yml.bak

[root@k8s-node2 ~]# cat /etc/alertmanager/alertmanager.yml

[root@k8s-node2 alertmanager]# more alertmanager.yml

global:

smtp_smarthost: 'smtp.qq.com:25'

smtp_from: '190235@qq.com'

smtp_auth_username: '190235@qq.com'

smtp_auth_password: '******'

smtp_require_tls: false

route:

receiver: mail

receivers:

- name: 'mail'

email_configs:

- to: '190235@qq.com'

[root@k8s-node2 alertmanager]#

（4）启动Alertmanager

[root@k8s-node2 alertmanager]# alertmanager --config.file alertmanager.yml

（5）在Prometheus上添加Alertmanage

# Alertmanager configuration

alerting:

alertmanagers:

- static_configs:

- targets:

- 192.168.20.43:9093

（6）在Prometheus上添加对Alertmanage监控

- job_name: 'alertmanager'

static_configs:

- targets: ['192.168.20.43:9093']

（7）在Prometheus添加告警规则

groups:

- name: node-up

rules:

- alert: node-up

expr: up{job="k8s-node1"} == 0

for: 15s

labels:

severity: 1

team: node

annotations:

summary: "{{ $labels.instance }} 已停止运行超过 15s！"

（8）把告警规则加入Prometheus配置文件

rule_files:

- "rules/*_rules.yml"

- "rules/*_alerts.yml"

（9) 查看配置规则

(10) 测试：停止k8s-node1节点，是否有报警

提示Pending(1)

(11) 查看报警Alertmanager

(12) 查看邮件报警

Prometheus+Grafana+Alertmanager +邮件告警配置完成

↓↓ 点击"阅读原文" 【加入DevOps运维团】

更多相关文章

随机推荐