Yace导出Cloudwatch指标至AWS托管Prometheus-AlertManager示例配置文件

告警管理模板

template_files:
  default_template: |
    {
   { define "sns.default.subject" }}[{
   { .Status | toUpper }}{
   { if eq .Status "firing" }}:{
   { .Alerts.Firing | len }}{
   { end }}]{
   { end }}
    {
   { define "__alertmanager" }}AlertManager{
   { end }}
    {
   { define "__alertmanagerURL" }}{
   { .ExternalURL }}/#/alerts?receiver={
   { .Receiver | urlquery }}{
   { end }}
alertmanager_config: |
  global:
    resolve_timeout: 5m
  templates:
    - 'default_template'
  route:
    group_wait: 60s  #在接收到一条新的告警（第一次出现的告警）时，将这条告警发送给 receiver 之前需要等待的时间
    group_interval: 60s #对于一条已经出现过的告警，alertmanager 检查会每隔
    repeat_interval: 60s #对于一条已经出现过的告警，每隔 repeat_interval 会重新发送给 receivers
    receiver: default
  receivers:
    - name: 'default'
      sns_configs:
      - topic_arn: arn:aws:sns:ap-southeast-1:AccontID:AlertManager
        sigv4:
          region: ap-southeast-1
        attributes:
          key: severity
          value: SEV2

RDS 告警规则

groups:
- name: AWS-RDS
  rules:
  - alert