- k8s
kubectl apply -f deployment.yaml
- docker-compose
docker-compose up -d
- docker:
docker run -d --name=work_wechat_robot -p 8080:8080 daozzg/work_wechat_robot
- 创建robot,获得robot的key,根据自己的alertmanager 规则,修改如下alertmanager的配置:
- : 你的robot的key
- : 部署本服务的地址,如果是同k8s的deployment部署,写service名即可,如: http://http://work-wechat-robot/prometheus_webhook
"global":
"resolve_timeout": "5m"
"route":
"group_by":
- "job"
"group_interval": "5m"
"group_wait": "30s"
"receiver": "default"
"repeat_interval": "12h"
"routes":
- "match":
"alertname": "DeadMansSwitch"
"receiver": "null"
"receivers":
- "name": "default"
"webhook_configs":
- "send_resolved": true
"url": "http://<work-wechat-robot>/prometheus_webhook"
"http_config":
"bearer_token": "<your robot key>"
- 建议配置Prometheus 启动参数
--web.external-url
,这样点击告警名能够跳转到 prometheus查看告警到监控信息(如果是prometheus-operator,增加spec.externalUrl)
为了更好的显示告警信息,请参考告警规范 Kubernetes Alert Runbooks: 如果按照规范,会接收到如下的告警信息
git clone {this_url}
cd work_wechat_robot
# please install pip (ubuntu: `apt-get install python-pip`,mac: `brew install python-pip` )
pip install virtualenv
virtualenv venv
source venv/bin/activate
pip install -r requirements.txt
python app.py
更新依赖