Discovering and scraping data with Kapacitor

Data can be pulled from a dynamic list of remote targets with the discovery and scraping features of Kapacitor. Use those features with TICKscripts to monitor targets, process the data, and write data to InfluxDB. Currently, Kapacitor supports only Prometheus style targets.

Note: Scraping and discovery is currently under technical preview. There may be changes to the configuration and behavior in subsequent releases.

Content

• Overview
• Configuring Scrapers and Discoverers

Overview

The diagram below outlines the infrastructure for discovering and scraping data with Kapacitor.

Image 1 – Scraping and discovery workflow

1. First, Kapacitor implements the discovery process to identify the available targets in your infrastructure. It requests that information at regular intervals and receives that information from an authority. In the diagram, the authority informs Kapacitor of three targets: A, B, and C.
2. Next, Kapacitor implements the scraping process to pull metrics data from the existing targets. It runs the scraping process at regular intervals. Here, Kapacitor requests metrics from targets A, B, and C. The application running on A, B, and C exposes a /metrics endpoint on its HTTP API which returns application-specific statistics.
3. Finally, Kapacitor processes the data according to configured TICKscripts. Use TICKscripts to filter, transform, and perform other tasks on the metrics data. In addition, if the data should be stored, configure a TICKscript to send it to InfluxDB.

Pushing vs. pulling metrics

By combining discovery with scraping, Kapacitor enables a metrics gathering infrastructure to pull metrics off of targets instead of requiring them to push metrics out to InfluxDB. Pulling metrics has several advantages in dynamic environments where a target may have a short lifecycle.

Configuring scrapers and discoverers

A single scraper scrapes the targets from a single discoverer. Configuring both scrapers and discoverers comes down to configuring each individually and then informing the scraper about the discoverer.

Below are all the configuration options for a scraper.

Example 1 – Scraper configuration

[[scraper]]
  enabled = false
  name = "myscraper"
  # ID of the discoverer to use
  discoverer-id = ""
  # The kind of discoverer to use
  discoverer-service = ""
  db = "mydb"
  rp = "myrp"
  type = "prometheus"
  scheme = "http"
  metrics-path = "/metrics"
  scrape-interval = "1m0s"
  scrape-timeout = "10s"
  username = ""
  password = ""
  bearer-token = ""
  ssl-ca = ""
  ssl-cert = ""
  ssl-key = ""
  ssl-server-name = ""
  insecure-skip-verify = false

Available discoverers

Kapacitor supports the following services for discovery:

See the example configuration file for details on configuring each discoverer.