Edit

Get Load Balancer metrics with Azure Monitor CLI

  • Article

In this article, you learn some examples to list Load Balancer metrics using Azure Monitor CLI.

Complete reference documentation and other samples for retrieving metrics using Azure Monitor CLI are available in the az monitor metrics reference.

Table of metric names via CLI

When you use CLI, Load Balancer metrics may use a different metric name for the CLI parameter value. When specifying the metric name via the --metric dimension parameter, use the CLI metric name instead. For example, the metric Data path availability would be used by specifying a parameter of --metric VipAvaialbility.

Here's a table of common Load Balancer metrics, the CLI metric name, and recommend aggregation values for queries:

Metric CLI metric name Recommended aggregation
Data path availability VipAvailability Average
Health probe status DipAvailability Average
SYN (synchronize) count SYNCount Average
SNAT connection count SnatConnectionCount Sum
Allocated SNAT ports AllocatedSnatPorts Average
Used SNAT ports UsedSnatPorts Average
Byte count ByteCount Sum
Packet count PacketCount Sum

For metric definitions and further details, refer to Monitoring load balancer data reference.

CLI examples for Load Balancer metrics

The az monitor metrics command is used to view Azure resource metrics. To see the metric definitions available for a Standard Load Balancer, you run the az monitor metrics list-definitions command.

# Display available metric definitions for a Standard Load Balancer resource

az monitor metrics list-definitions --resource <resource_id>

Note

In the all the following examples, replace <resource_id> with the unique resource id of your Standard Load Balancer.

To retrieve Standard Load Balancer metrics for a resource, you can use the az monitor metrics list command. For example, use the --metric DipAvailability option to collect the Health Probe Status metric from a Standard Load Balancer.


# List the Health Probe Status metric from a Standard Load Balancer

az monitor metrics list --resource <resource_id> --metric DipAvailability

When you run the above command, the output for Health Probe status will be like the following output:

user@Azure:~$ az monitor metrics list --resource <resource_id> --metric DipAvailability
{
  "cost": 59,
  "interval": "0:01:00",
  "namespace": "Microsoft.Network/loadBalancers",
  "resourceregion": "eastus2",
  "timespan": "2022-06-30T15:22:39Z/2022-06-30T16:22:39Z",
  "value": [
    {
      "displayDescription": "Average Load Balancer health probe status per time duration",
      "errorCode": "Success",
      "errorMessage": null,
      "id": "/subscriptions/6a5f35e9-6951-499d-a36b-83c6c6eed44a/resourceGroups/myResourceGroup2/providers/Microsoft.Network/loadBalancers/myLoadBalancer/providers/Microsoft.Insights/metrics/DipAvailability",
      "name": {
        "localizedValue": "Health Probe Status",
        "value": "DipAvailability"
      },
      "resourceGroup": "myResourceGroup2",
      "timeseries": [],
      "type": "Microsoft.Insights/metrics",
      "unit": "Count"
    }
  ]
}
...

You can specify the aggregation type for a metric with the –-aggregation parameter. For recommended aggregations, see Monitoring load balancer data reference](./monitor-load-balancer-reference.md).


# List the average Health Probe Status metric from a Standard Load Balancer

az monitor metrics list --resource <resource_id> --metric DipAvailability --aggregation Average

To specify the interval to metrics, use the --interval parameter and specify a value in ##h##m format. The default interval is 1 m.


# List the average List the average Health Probe Status metric from a Standard Load Balancer in 5 minute intervals

az monitor metrics list --resource <resource_id> --metric DipAvailability --aggregation Average --interval 5m

By default, az monitor metrics list returns the resource’s aggregate metrics from the last hour. You can query metric data over a period of time using --start-time and --end-time with the format of date (yyyy-mm-dd) time (hh:mm:ss.xxxxx) timezone (+/-hh:mm). To list the average Health Probe Status aggregated per day from May 5, 2022 and May 10, 2022, use the following command:

# List average Health Probe Status metric aggregated per day from May 5, 2022 and May 10, 2022. 

az monitor metrics list --resource <resource_id> --metric DipAvailability --start-time 2022-05-01T00:00:00Z --end-time 2022-05-10T00:00:00Z --interval PT24H --aggregation Average

Note

Start and end times are represented using a format of yyyy-mm-dd format. For example, every day between May 5, 2022 and May 10, 2022 would be represented as 2022-05-01 and 2022-05-10.

To split metrics on a dimension, such as “BackendIPAddress”, specify the dimension in the --filter flag. Dimensions of a metric are name/value pairs that include more data to describe the metric value. To learn more about which dimensions are supported for each metric, see Monitoring load balancer data reference.

# List average Health Probe Status metric and filter for all BackendIPAddress dimensions

az monitor metrics list --resource $res --metric DipAvailability --filter "BackendIPAddress eq '*'" --aggregation Average

You can also specify a specific dimension value.

# List average Health Probe Status metric and filter for the 10.1.0.4 BackendIPAddress dimension

az monitor metrics list --resource <resource_id> --metric DipAvailability --filter "BackendIPAddress eq '10.1.0.4'" --aggregation Average

In cases where you need to filter on multiple dimension values, specify the --filter value using and between the values.

# List average Health Probe Status metric and filter for all BackendIPAddress and BackendPort dimensions

az monitor metrics list --resource <resource_id> --metric DipAvailability --filter "BackendIPAddress eq '*' and BackendPort eq '*'" --aggregation Average

Next steps