Skip to content

http

Generate requests for an HTTP service and and collect latency and error-related metrics.

Usage example

In this experiment, the http task generates requests for https://httpbin.org/get, and collects latency and error-related metrics. The metrics are used by the assess task to validate SLOs.

iter8 launch \
--set "tasks={http,assess}" \
--set http.url=https://httpbin.org/get \
--set assess.SLOs.upper.http/latency-mean=50 \
--set assess.SLOs.upper.http/error-count=0

Parameters

Name Type Description
url string (URL) HTTP URL where requests are sent.
headers map[string]string HTTP headers to use in the requests.
numRequests int Number of requests to be sent to the app. Default value is 100.
duration string Duration of this task. Specified in the Go duration string format (example, 5s). If both duration and numRequests are specified, then duration is ignored.
qps float qps stands for queries-per-second. Number of requests per second sent to the app. Default value is 8.0.
connections int Number of parallel connections used to send requests. Default value is 4.
payloadURL string (URL) URL from which to download the content that will be used as the request payload. If this field is specified, Iter8 will send HTTP POST requests to the app using this content as the payload.
payloadStr string String data to be used as the request payload. If this field is specified, Iter8 will send HTTP POST requests to the app using this string as the payload.
contentType string Content type of the payload. This is intended to be used in conjunction with one of the payload* fields. If this field is specified, Iter8 will send HTTP POST requests to the app using this as the Content-Type header value.

Metrics

This task creates a built-in provider named http. The following metrics are collected by this task:

  • http/request-count: total number of requests sent
  • http/error-count: number of error responses
  • http/error-rate: fraction of error responses
  • http/latency-mean: mean of observed latency values
  • http/latency-stddev: standard deviation of observed latency values
  • http/latency-min: min of observed latency values
  • http/latency-max: max of observed latency values
  • http/latency-pX: Xth percentile latency, for X in [50.0, 75.0, 90.0, 95.0, 99.0, 99.9]

All latency metrics have msec units.

Back to top