Repository files navigation

This project has been archived and will not recieve any further updates The functionality within this project has been implemented in thepuppet_metrics_dashboard repository. Please use that repository moving forward.

This repository contains a command line tool for generating visualizations of your Puppet metrics data in Docker.

It assumes you have collected the metrics using thepuppetlabs/puppet_metrics_collector module.

It downloads a script from that module, and Grafana dashboards from thepuppetlabs/puppet_metrics_dashboard module.

To use this tool, you will needdocker (and docker-compose) installed.Tip: If you're using a Mac, use the official Mac packages for Docker instead of installing from Brew.(If you figure out how to use this with Docker installed from Brew, let us know.)

With Docker installed, you can run theview-in-grafana.sh script, passing it the directory containing the data files to load.

For example:

./view-in-grafana.sh ~/Downloads/puppet_metrics/puppetserver

You can then view the metrics by visitinghttp://localhost:3000 in a web browser.

• username:admin
• password:admin.

Theview-in-grafana.sh script has several options that can change the behavior of the environment.

By default, the script uses a data retention of 30 days.You can optionally specify a different data retention period.

For example:

./view-in-grafana.sh ~/Downloads/puppet_metrics/puppetserver 10

Note:.json files outside the retention period will be deleted, as the assumption is that they exist in the tar archives.

By default, InfluxDB is used to store data.You can optionally specify Graphite.

For example:

./view-in-grafana.sh -d graphite  ~/Downloads/puppet_metrics/puppetserver

To test Grafana updates, you can specify the-b option to build the localgrafana-puppetserver container.

For example:

./view-in-grafana.sh -b ~/Downloads/puppet_metrics/puppetserver

Thejson2timeseriesdb script frompuppetlabs/puppet_metrics_collector module can be used to transform data in the JSON files into a format that can be imported into any InfluxDB or Graphite database backend.

Usage:

./json2timeseriesdb [--pattern PATTERN] [filename_1 ... filename_n]

Output is in Graphite's plain text input format.The output can be sent to a host running Graphite by passing its hostname to the--netcat flag:

./json2timeseriesdb ~/Downloads/logdump/puppetserver/*.json --netcat localhost

Data will be sent to port 2003.A custom port can be used by sending output tonc:

./json2timeseriesdb ~/Downloads/logdump/puppetserver/*.json | nc localhost 4242

Output in InfluxDB's format can be specified using the--convert-to flag:

./json2timeseriesdb ~/Downloads/logdump/puppetserver/*.json --convert-to influxdb

When--netcat is used with InfluxDB, the--influx-db flag must be used to specify an InfluxDB database:

./json2timeseriesdb ~/Downloads/logdump/puppetserver/*.json --convert-to influxdb --netcat localhost --influx-db pe-metrics

(Data will be sent to port 8086.)

The above examples can be used for small numbers of files.When more files exist than can be referenced as arguments, use--pattern:

./json2timeseriesdb --pattern '~/Downloads/logdump/puppetserver/*.json' --netcat localhost

The--pattern flag accepts a Ruby glob argument, which Ruby will then expand into a list of files to process.