Skip to content

Files

Latest commit

 

History

History
263 lines (234 loc) · 7.09 KB

_index.md

File metadata and controls

263 lines (234 loc) · 7.09 KB

Notification Manager HTTP API

The Notification Manager Deployment exposes an HTTP API for receiving alerts, sending notifications, and verifing config.

The default port of Notification Manager is 19093.

Receive alerts

Post /api/v2/alerts

This API is used to send alerts to Notification Manager. Notification Manager will then send notifications to the receivers defined in the cluster.

Request:

{
  "receiver": "Critical",
  "status": "firing",
  "alerts": [
    {
      "status": "firing",
      "labels": {
        "alertname": "KubePodCrashLooping",
        "container": "busybox-3jb7u6",
        "instance": "10.233.71.230:8080",
        "job": "kube-state-metrics",
        "namespace": "pp1",
        "pod": "dd1-0",
        "prometheus": "kubesphere-monitoring-system/k8s",
        "severity": "critical"
      },
      "annotations": {
        "message": "Pod pp1/dd1-0 (busybox-3jb7u6) is restarting 1.07 times / 5 minutes.",
        "runbook_url": "https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-kubepodcrashlooping"
      },
      "startsAt": "2020-02-26T07:05:04.989876849Z",
      "endsAt": "0001-01-01T00:00:00Z",
      "generatorURL": "http://prometheus-k8s-0:9090/graph?g0.expr=rate%28kube_pod_container_status_restarts_total%7Bjob%3D%22kube-state-metrics%22%7D%5B15m%5D%29+%2A+60+%2A+5+%3E+0\u0026g0.tab=1",
      "fingerprint": "a4c6c4f7a49ca0ae"
    }
  ],
  "groupLabels": {
    "alertname": "KubePodCrashLooping",
    "namespace": "pp1"
  },
  "commonLabels": {
    "alertname": "KubePodCrashLooping",
    "container": "busybox-3jb7u6",
    "instance": "10.233.71.230:8080",
    "job": "kube-state-metrics",
    "namespace": "pp1",
    "pod": "dd1-0",
    "prometheus": "kubesphere-monitoring-system/k8s",
    "severity": "critical"
  },
  "commonAnnotations": {
    "message": "Pod pp1/dd1-0 (busybox-3jb7u6) is restarting 1.07 times / 5 minutes.",
    "runbook_url": "https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-kubepodcrashlooping"
  },
  "externalURL": "http://alertmanager-main-2:9093"
}

You can find the details about the request at here.

Response:

{
  "Status":200,
  "Message":"Notification request accepted"
}

Status code 200 means Notification Manager has received the alerts and begins to send notifications, it doesn't mean the notifications have been sent successfully.

Send notifications

Post /api/v2/notifications

This API is used to send notifications directly.

Request:

{
  "config":{
    "apiVersion":"notification.kubesphere.io/v2beta2",
    "kind":"Config",
    "metadata":{
      "name":"test-user-config",
      "labels":{
        "app":"notification-manager",
        "type":"default",
      }
    },
    "spec":{
      "email":{
        "authPassword":{
          "value": <password>
        },
        "authUsername": <user>,
        "from": <email>,
        "requireTLS":true,
        "smartHost":{
          "host": <smtp-host>,
          "port": <smtp port>
        }
      }
    }
  },
  "receiver":{
    "apiVersion":"notification.kubesphere.io/v2beta2",
    "kind":"Receiver",
    "metadata":{
      "name":"test-user-receiver",
      "labels":{
        "app":"notification-manager",
        "type":"global",
      }
    },
    "spec":{
      "email":{
        "to":[
          <email>
        ]
      }
    }
  },
  "alert": {
    "receiver": "Critical",
    "status": "firing",
    "alerts": [
      {
        "status": "firing",
        "labels": {
          "alertname": "KubePodCrashLooping",
          "container": "busybox-3jb7u6",
          "instance": "10.233.71.230:8080",
          "job": "kube-state-metrics",
          "namespace": "pp1",
          "pod": "dd1-0",
          "prometheus": "kubesphere-monitoring-system/k8s",
          "severity": "critical"
        },
        "annotations": {
          "message": "Pod pp1/dd1-0 (busybox-3jb7u6) is restarting 1.07 times / 5 minutes.",
          "runbook_url": "https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-kubepodcrashlooping"
        },
        "startsAt": "2020-02-26T07:05:04.989876849Z",
        "endsAt": "0001-01-01T00:00:00Z",
        "generatorURL": "http://prometheus-k8s-0:9090/graph?g0.expr=rate%28kube_pod_container_status_restarts_total%7Bjob%3D%22kube-state-metrics%22%7D%5B15m%5D%29+%2A+60+%2A+5+%3E+0\u0026g0.tab=1",
        "fingerprint": "a4c6c4f7a49ca0ae"
      }
    ],
    "groupLabels": {
      "alertname": "KubePodCrashLooping",
      "namespace": "pp1"
    },
    "commonLabels": {
      "alertname": "KubePodCrashLooping",
      "container": "busybox-3jb7u6",
      "instance": "10.233.71.230:8080",
      "job": "kube-state-metrics",
      "namespace": "pp1",
      "pod": "dd1-0",
      "prometheus": "kubesphere-monitoring-system/k8s",
      "severity": "critical"
    },
    "commonAnnotations": {
      "message": "Pod pp1/dd1-0 (busybox-3jb7u6) is restarting 1.07 times / 5 minutes.",
      "runbook_url": "https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-kubepodcrashlooping"
    },
    "externalURL": "http://alertmanager-main-2:9093"
  }
}
  • config: The corresponding config for the receiver. If config is not set, the receiver will use the default config defined in the cluster. If the config is set, the config type must match the receiver type.
  • receiver: Receiver that receives notifications. It must be set.
  • alert: Alerts that will be sent to the receiver.

Response:

{
  "Status":200,
  "Message":"Send alerts successfully"
}

Verify

Post /api/v2/verify

This API used to verify the config and receiver are correct. If the config and receiver are correct, receiver will receive a test notification.

Request:

{
  "config":{
    "apiVersion":"notification.kubesphere.io/v2beta2",
    "kind":"Config",
    "metadata":{
      "name":"test-user-config",
      "labels":{
        "app":"notification-manager",
        "type":"default",
      }
    },
    "spec":{
      "email":{
        "authPassword":{
          "value": <password>
        },
        "authUsername": <user>,
        "from": <email>,
        "requireTLS":true,
        "smartHost":{
          "host": <smtp-host>,
          "port": <smtp port>
        }
      }
    }
  },
  "receiver":{
    "apiVersion":"notification.kubesphere.io/v2beta2",
    "kind":"Receiver",
    "metadata":{
      "name":"test-user-receiver",
      "labels":{
        "app":"notification-manager",
        "type":"global",
      }
    },
    "spec":{
      "email":{
        "to":[
          <email>
        ]
      }
    }
  }
}
  • config: The corresponding config for the receiver. If config is not set, the receiver will use the default config defined in the cluster. If the config is set, the config type must match the receiver type.
  • receiver: Receiver that receives notifications. It must be set.

Response:

{
  "Status":200,
  "Message":"Verify successfully"
}