Configuring Alarm and Alert Collection from a Qumulo Cluster

This section explains how to collect alarms and alerts from a Qumulo Cluster by using the Qumulo Alerts Web UI and the alerts CLI.

To Configure Alarm and Alert Collection by Using the Qumulo Alerts Web UI

This section explains how to configure alarm and alert collection by adding your Qumulo cluster to Qumulo Alerts and then configuring alert plugins for your cluster in the Qumulo Alerts Web UI.

Important

When you add a Qumulo cluster to Qumulo Alerts, all available alarm and alert plugins are enabled for that cluster by default.
If you use a floating IP address, you must click Network Load Balancer to ensure that Qumulo Alerts connects to the node that currently uses the floating IP address.
To avoid spreading a plugin's API request load across the nodes of a Qumulo cluster, all alarm and alert plugins communicate with your cluster by using either a network load balancer or floating IP addresses. You can configure one—but not both—of these communication methods.

Step 1: Add a Qumulo Cluster to Qumulo Alerts

Log in to the Qumulo Alerts Web UI.
On the sidebar, under Monitoring, click Clusters.
Click + Add Cluster and enter the following details:
1. For Cluster Name/IP, specify the hostname or IP address for your Qumulo cluster.
  
  Important
  Qumulo Alerts uses this value as the display name. It isn’t possible to change it after saving the cluster configuration.
2. For Access Token, specify the long-lived access token that you created while Installing and Configuring Qumulo Alerts guide.
3. For Port, specify the REST API port (8000 by default).
4. For Polling Frequency (minutes), specify how frequently Qumulo Alerts should poll your Qumulo cluster.
5. (Optional) If your Qumulo cluster is accessible through a floating IP address behind a network load balancer, click Network Load Balancer.
Click Save.

Step 2: Configure Alarm or Alert Plugins for a Qumulo Cluster

Tip
To view all available plugin names and categories before configuring a cluster, click Monitoring > Alert Types on the sidebar

Log in to the Qumulo Alerts Web UI.
On the sidebar, under Monitoring, click Clusters and then click Edit next to the cluster to configure.
In the Plugins section, click individual alarm and alert plugins to enable or disable them for your Qumulo cluster.
Click Save.

To Configure Alarm and Alert Collection by Using the alerts CLI

This section explains how to collect information and specific alarms; all alarms; or all alarms, alerts, and informational messages by using the alerts CLI.

Collecting Information about Specific Alarms

Run the ./alerts cluster_add command and specify the fully qualified domain name (FQDN) of your Qumulo cluster, your long-lived access token for the Qumulo REST API, and the plugins or plugin categories to include or exclude from monitoring.

In the following example, we include the plugins Disks and Nodes.

./alerts cluster_add \
  --name cluster.example.com \
  --token 12345678901234567890 \
  -pi Disks \
  -pi Nodes

The following is example output.

[{
  "frequency": 1,
  "id": 1,
  "name": "cluster.example.com",
  "nlb": false,
  "plugins": [{
    "category": "Alarms",
    "description": "Get Disk State Information",
    "frequency": null,
    "name": "Disks"
  },{
    "category": "Alarms",
    "description": "Get Cluster Node Failures",
    "frequency": null,
    "name": "Nodes"
  }],
  "port": 8000
}]

Note

For the --nlb flag, the false setting requires floating IP address configuration.
To avoid spreading a plugin's API request load across the nodes of a Qumulo cluster, all alarm and alert plugins communicate with your cluster by using either a network load balancer or floating IP addresses. You can configure one—but not both—of these communication methods.

Collecting Information about All Alarms

In the following example, we include the Alarms category.

./alerts cluster_add \
  --name cluster.example.com \
  --token 12345678901234567890 \
  -pc Alarms

The following is example output.

[{
  "frequency": 1,
  "id": 1,
  "name": "cluster.example.com",
  "nlb": false,
  "plugins": [{
    "category": "Alarms",
    "description": "Get Disk State Information",
    "frequency": null,
    "name": "Disks"
  },{
    "category": "Alarms",
    "description": "Get Cluster Node Failures",
    "frequency": null,
    "name": "Nodes"
  },{
    "category": "Alarms",
    "description": "Get Fan Failures",
    "frequency": null,
    "name": "Fans"
  },{
    "category": "Alarms",
    "description": "Get CPU Overtemp",
    "frequency": null,
    "name": "CPU"
  },
  ...
  ],
  "port": 8000
}]

Collecting Information about All Alarms, Alerts, and Informational Messages

In the following example, we include the Alarms, Alerts, and Informational categories.

./alerts cluster_add \
  --name cluster.example.com \
  --token 12345678901234567890 \
  -pc Alarms \
  -pc Alerts \
  -pc Informational

The following is example output.

[{
  "frequency": 1,
  "id": 1,
  "name": "cluster.example.com",
  "nlb": false,
  "plugins": [{
    "category": "Alarms",
    "description": "Get Disk State Information",
    "frequency": null,
    "name": "Disks"
  },{
    "category": "Alarms",
    "description": "Get Cluster Node Failures",
    "frequency": null,
    "name": "Nodes"
  },{
    "category": "Alerts",
    "description": "Get Active Directory State",
    "frequency": null,
    "name": "AD"
  },{
    "category": "Alerts",
    "description": "Get Audit Status",
    "frequency": null,
    "name": "Audit"
  },{
    "category": "Alerts",
    "description": "Get Cluster Volume Capacity",
    "frequency": null,
    "name": "Capacity"
  },
  ...
  ],
  "port": 8000
}]