打开电脑版钉钉创建机器人
WebHook 接收器
上面我们配置的是 AlertManager 自带的邮件报警模板,我们也说了 AlertManager 支持很多中报警接收器,比如 slack、微信之类的,其中最为灵活的方式当然是使用 webhook 了,我们可以定义一个 webhook 来接收报警信息,然后在 webhook 里面去进行处理,需要发送怎样的报警信息我们自定义就可以,下面的 JSON 数据就是 AlertManager 将报警信息 POST 给 webhook 的数据:
{
"receiver": "webhook",
"status": "firing",
"alerts": [
{
"status": "firing",
"labels": {
"alertname": "NodeMemoryUsage",
"beta_kubernetes_io_arch": "amd64",
"beta_kubernetes_io_os": "linux",
"instance": "node1",
"job": "nodes",
"kubernetes_io_arch": "amd64",
"kubernetes_io_hostname": "node1",
"kubernetes_io_os": "linux",
"team": "node"
},
"annotations": {
"description": "node1: Memory usage is above 30% (current value is: 42.097619438581596)",
"summary": "node1: High Memory usage detected"
},
"startsAt": "2022-03-02T02:13:19.69Z",
"endsAt": "0001-01-01T00:00:00Z",
"generatorURL": "http://prometheus-649968556c-8p4tj:9090/graph?g0.expr=%28node_memory_MemTotal_bytes+-+%28node_memory_MemFree_bytes+%2B+node_memory_Buffers_bytes+%2B+node_memory_Cached_bytes%29%29+%2F+node_memory_MemTotal_bytes+%2A+100+%3E+30\u0026g0.tab=1",
"fingerprint": "8cc4749f998d64dd"
}
],
"groupLabels": { "instance": "node1" },
"commonLabels": {
"alertname": "NodeMemoryUsage",
"beta_kubernetes_io_arch": "amd64",
"beta_kubernetes_io_os": "linux",
"instance": "node1",
"job": "nodes",
"kubernetes_io_arch": "amd64",
"kubernetes_io_hostname": "node1",
"kubernetes_io_os": "linux",
"team": "node"
},
"commonAnnotations": {
"description": "node1: Memory usage is above 30% (current value is: 42.097619438581596)",
"summary": "node1: High Memory usage detected"
},
"externalURL": "http://alertmanager-5774d6f5f4-prdgr:9093",
"version": "4",
"groupKey": "{}/{team=\"node\"}:{instance=\"node1\"}",
"truncatedAlerts": 0
}
我这里实现了一个简单的 webhook 程序,代码仓库地址:https://github.com/cnych/promoter,该程序支持在消息通知中显示报警图表。
首先在钉钉群中选择创建一个自定义的机器人:
这里我们选择添加额外密钥的方式来验证机器人,其他两种方式可以忽略,需要记住该值,下面会使用:
创建完成后会提供一个 webhook 的地址,该地址会带一个 acess_token 的参数,该参数下面也会使用:
接下来我们需要将 webhook 服务部署到集群中,对应的资源清单如下:
# promoter.yaml
apiVersion: v1
kind: ConfigMap
metadata:
name: promoter-conf
namespace: kube-mon
data:
config.yaml: |
global:
prometheus_url: http://192.168.31.31:30104
wechat_api_secret: <secret> # 企业微信 secret
wechat_api_corp_id: <corp_id> # 企业微信 corp_id
dingtalk_api_token: <token> # 钉钉机器人 token
dingtalk_api_secret: <secret> # 钉钉机器人 secret
s3:
access_key: <ak>
secret_key: <sk>
endpoint: oss-cn-beijing.aliyuncs.com
region: cn-beijing
bucket: my-oss-testing
receivers:
- name: test1
wechat_configs:
- agent_id: <agent_id>
to_user: "@all"
message_type: markdown
dingtalk_configs:
- message_type: markdown
at:
isAtAll: true
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: promoter
namespace: kube-mon
labels:
app: promoter
spec:
selector:
matchLabels:
app: promoter
template:
metadata:
labels:
app: promoter
spec:
volumes:
- name: promotercfg
configMap:
name: promoter-conf
containers:
- name: promoter
image: cnych/promoter:main
imagePullPolicy: IfNotPresent
args:
- "--config.file=/etc/promoter/config.yaml"
ports:
- containerPort: 8080
volumeMounts:
- mountPath: "/etc/promoter"
name: promotercfg
---
apiVersion: v1
kind: Service
metadata:
name: promoter
namespace: kube-mon
labels:
app: promoter
spec:
selector:
app: promoter
ports:
- port: 8080
配置完成后,直接创建上面的资源对象即可:
☸ ➜ kubectl apply -f promoter.yaml
☸ ➜ kubectl get pods -n kube-mon -l app=promoter
NAME READY STATUS RESTARTS AGE
promoter-67c5795c4c-7mlvq 1/1 Running 3 (34m ago) 3d16h
部署成功后,现在我们就可以给 AlertManager 配置一个 webhook 了,在上面的配置中增加一个路由接收器。
routes:
- receiver: webhook
group_wait: 10s
group_by: ['instance']
match:
team: node
receivers:
- name: 'webhook'
webhook_configs:
- url: 'http://promoter:8080/test1/send'
send_resolved: true
我们这里配置了一个名为 webhook 的接收器,地址为:http://promoter:8080/test1/send
,这个地址当然就是上面我们部署的钉钉的 webhook 的接收程序的 Service 地址。
然后我们可以更新 AlertManager 和 Prometheus 的 ConfigMap 资源对象,更新完成后,隔一会儿执行 reload 操作是更新生效,如果有报警触发的话,隔一会儿关于这个节点文件系统的报警就会被触发了,由于这个报警信息包含一个team=node
的 label 标签,所以会被路由到 webhook 这个接收器中,也就是上面我们自定义的这个 webhook,触发后可以观察这个 Pod 的日志:
☸ ➜ kubectl logs -f promoter-5dbd47798c-bnjqm -n kube-mon
ts=2022-03-07T01:38:08.001Z caller=main.go:58 level=info msg="Staring Promoter" version="(version=0.2.3, branch=HEAD, revision=0a9cf8fc9bd55d1d2d47d181867135914927c2fc)"
ts=2022-03-07T01:38:08.001Z caller=main.go:59 level=info build_context="(go=go1.17.8, user=root@91adc4eacff7, date=20220305-05:40:54)"
ts=2022-03-07T01:38:08.001Z caller=main.go:127 level=info component=configuration msg="Loading configuration file" file=/etc/promoter/config.yaml
ts=2022-03-07T01:38:08.002Z caller=main.go:138 level=info component=configuration msg="Completed loading of configuration file" file=/etc/promoter/config.yaml
ts=2022-03-07T01:38:08.003Z caller=main.go:88 level=info msg=Listening address=:808
可以看到 POST 请求已经成功了,同时这个时候正常来说就可以收到一条钉钉消息了:
1.创建钉钉机器人
打开电脑版钉钉,创建一个群,创建自定义机器人,按如下步骤创建
我创建的机器人如下:
群设置-->智能群助手-->添加机器人-->自定义-->添加
机器人名称:kubernetes
接收群组:钉钉报警测试
安全设置:
自定义关键词:cluster1
上面配置好之后点击完成即可,这样就会创建一个kubernrtrs的报警机器人,创建机器人成功之后怎么查看webhook,按如下:
https://oapi.dingtalk.com/robot/send?access_token=b2d16db54d6fbd69230f080867d41345e6a883cdb8b929505642483216434f41
点击智能群助手,可以看到刚才创建的kubernetes这个机器人,点击kubernetes,就会进入到kubernetes机器人的设置界面
出现如下内容:
机器人名称:kubernetes
接受群组:钉钉报警测试
消息推送:开启
安全设置:
自定义关键词:cluster1
以上都是在钉钉上面操作完成
2.安装钉钉的webhook插件,在k8s的master1节点操作
tar zxvf prometheus-webhook-dingtalk-0.3.0.linux-amd64.tar.gz
prometheus-webhook-dingtalk-0.3.0.linux-amd64.tar.gz压缩包所在的百度网盘地址如下:
链接:https://pan.baidu.com/s/1_HtVZsItq2KsYvOlkIP9DQ
提取码:d59o
cd prometheus-webhook-dingtalk-0.3.0.linux-amd64
启动钉钉报警插件(关键词不要忘记了,这是上面已经创建好了的 profile="cluster1)
nohup ./prometheus-webhook-dingtalk --web.listen-address="0.0.0.0:8060" --ding.profile="cluster1=https://oapi.dingtalk.com/robot/send?access_token=b2d16db54d6fbd69230f080867d41345e6a883cdb8b929505642483216434f41" &
[root@master prometheus-webhook-dingtalk-0.3.0.linux-amd64]# tail -f nohup.out
level=info ts=2021-11-23T11:47:41.638047928Z caller=main.go:37 msg="Starting prometheus-webhook-dingtalk" version="(version=0.3.0, branch=HEAD, revision=4a7dee0be14073aba1ea2eed80acbb515564f664)"
level=info ts=2021-11-23T11:47:41.63812183Z caller=main.go:57 msg="Using default template"
level=info ts=2021-11-23T11:47:41.638150328Z caller=main.go:62 msg="Using following dingtalk profiles: map[cluster1:https://oapi.dingtalk.com/robot/send?access_token=b2d16db54d6fbd69230f080867d41345e6a883cdb8b929505642483216434f41]"
level=info ts=2021-11-23T11:47:41.638256519Z caller=main.go:83 msg="Listening on address" address=0.0.0.0:8060
对原来的文件做备份
cp alertmanager-cm.yaml alertmanager-cm.yaml.bak
3.alertManager对接webhook插件
[root@master ~]# kubectl get svc -n monitor
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
prometheus-webhook-dingtalk NodePort 10.233.6.147 <none> 8060:30481/TCP 3m57s
[root@master prometheus]# vim alertmanager-dingding-configmap.yaml
receivers:
- name: default-receiver
email_configs:
- to: "xxx@163.com"
- name: dingding
webhook_configs:
- url: http://prometheus-webhook-dingtalk.monitor:8060/dingtalk/dingding_ops/send
send_resolved: true
[root@master prometheus]# kubectl exec -it dns-test sh
/ # nslookup prometheus-webhook-dingtalk.monitor
Server: 169.254.25.10
Address 1: 169.254.25.10
Name: prometheus-webhook-dingtalk.monitor
Address 1: 10.233.6.147 prometheus-webhook-dingtalk.monitor.svc.cluster.local