You can use the Avi Load Balancer API to access throughput data and other metrics for a virtual service.

API calls to /API/analytics/metrics are used to get all metric-related data. The following example shows a query for the throughput of a virtual service for a 24 hours:

https://<Controller VIP>/api/analytics/metrics/virtualservice/virtualservice-<VS UUID>?metric_id=l4_client.avg_bandwidth &step=86400&limit=7&start=2015-11-17T00:00:00.000Z

  • Controller VIP: Replace with the IP address of your Avi Load Balancer Controller or the cluster IP address assigned to the cluster of redundant Controllers.

  • VS UUID: Universal Unique ID (UUID) of the virtual service. As virtual service names might not be unique across multiple tenants, API queries are recommended to include the UUID of the virtual service.

A quick way to find the UUID of a virtual service is to view the virtual service in the Controller web interface, and capture the UUID from the browser URI. To find virtual service UUIDs, see Find Virtual Service UUID topic in the VMware Avi Load BalancerConfiguration Guide.

step

Sampling period, in seconds. For 24 hours, use sampling time 86400.

limit

Maximum number of data points, in days, that will be returned. In this example, 7 data points are returned. Each data point is an average of that day’s client-to-virtual service network throughput.

start

Start time for the query. In this example, the query starts on October 17 and ends after 7 days, on October 23. In the same way, if the starting point is set to 2, only two data points are returned, one for each of the 2 days.

metric_id

Metric to be queried. Here are the metric IDs for some commonly queried metrics:

l4_client.avg_total_rtt

l4_server.avg_total_rtt

l4_client.avg_bandwidth

l4_client.max_open_conns

l4_client.avg_complete_conns

l4_client.avg_lossy_connections

l4_client.avg_errored_connections

l4_client.pct_connection_errors

l7_client.avg_complete_responses

l7_server.avg_resp_4xx_errors

l7_server.avg_resp_5xx_errors

l7_client.avg_resp_4xx_avi_errors

l7_client.avg_resp_5xx_avi_errors

l7_client.pct_response_errors

l7_server.avg_application_response_time

l7_client.avg_client_data_transfer_time

The following is an example of the data returned by the Avi Load Balancer Controller in response to an API query for l4_client.avg_bandwidth.

{
"entity_uuid": "virtualservice-9062c630-a835-4c22-8596-dfa625b9697d", 
"limit": 365, 
"metric_entity": "VSERVER_METRICS_ENTITY", 
"metric_id": "l4_client.avg_bandwidth", 
"series": [
{
"data": [
{
"timestamp": "2015-11-17T00:00:00+00:00", 
"value": 511500369.582942
}, 
{
"timestamp": "2015-11-18T00:00:00+00:00", 
"value": 641335291.443974
}, 
{
"timestamp": "2015-11-19T00:00:00+00:00", 
"value": 642964876.307424
}, 
{
"timestamp": "2015-11-20T00:00:00+00:00", 
"value": 652721668.799788
}, 
{
"timestamp": "2015-11-21T00:00:00+00:00", 
"value": 833373106.654519
}, 
{
"timestamp": "2015-11-22T00:00:00+00:00", 
"value": 670429050.977858
}, 
{
"timestamp": "2015-11-23T00:00:00+00:00", 
"value": 671784391.196803
}, 
], 
"header": {
"derivation_data": {
"derivation_fn": "SUM_BW_GAUGE", 
"metric_ids": "avg_rx_bytes, avg_tx_bytes"
}, 
"entity_uuid": "virtualservice-9062c630-a835-4c22-8596-dfa625b9697d", 
"metric_description": "Average transmit and receive network bandwidth between client and virtual service.", 
"metrics_min_scale": 1000000.0, 
"metrics_sum_agg_invalid": false, 
"name": "l4_client.avg_bandwidth", 
"obj_id_type": "METRICS_OBJ_ID_TYPE_VIRTUALSERVICE", 
"priority": true, 
"statistics": {
"max": 833373106.654519, 
"max_ts": "2015-11-21T00:00:00+00:00", 
"mean": 663407846.526, 
"min": 511500369.582942, 
"min_ts": "2015-11-17T00:00:00+00:00", 
"num_samples": 7, 
"trend": -0.029
}, 
"tenant_uuid": "admin", 
"units": "BITS_PER_SECOND"
}
}
], 
"start": "2015-11-17T00:00:00.000Z", 
"step": 86400, 
"stop": "2015-11-23T00:00:00+00:00"
}