Skip to content

Latest commit

 

History

History
274 lines (202 loc) · 9.44 KB

README.md

File metadata and controls

274 lines (202 loc) · 9.44 KB

Hieracles

Gem Version Downloads Build Status Coverage Status Dependency Status Code Climate

Hieracles is a command-line tool for analysis and deep examination of Hiera parameters in a Puppet setup. It can be used to quickly visualize, from a local puppet (typically on a developers environment), all the Hiera params related to a specific node.

It's used internally at Gandi and its first incarnation is strongly tied to Gandi puppet architecture. But Hieracles tends to become, in time, a generic Hiera overlay visualization tool.

Have a look at the Changelog for details about the evolution.

Prerequisite

There are many ways to setup puppet and use Hiera. This tool is designed to match a certain kind of setup, including:

As the development is going on, more generic options will be provided, but for now, we mainly make it robust to fit the context we have.

Install

At this stage, it's to early to even think about installing anything. The internal code from Gandi is still in progress of transition towards total freedom and generic usage.

Despite this warning, you can

gem install hieracles

or add in your Gemfile:

gem 'hieracles'

Configuration

At first launch it will create a configuration file in ~/.config/hieracles/config.yml

Configuration variables are:

  • basepath (alias localpath)
    This is where your puppet repo is cloned
  • classpath
    where to find classes defined in the ENC
  • modulepath
    where to find modules called in the classes
  • encpath
    where to read information about each nodes
  • hierafile
    where to find a hierafile customized for your local puppet installation
  • format
    can be plain, console, csv, yaml, rawyaml, json
  • defaultscope
    a hash defining scope variables that will be used if not defined by a facts file or by params passed as arguments

For an example setup you can check in spec/files.

If you don't specify the basepath, your current location will be used as a base path.

Usage

Usage: hc <fqdn> <command> [extra_args]

Available commands:
  info        provides the farm, datacenter, country
              associated to the given fqdn
              An extra param can be added for filtering
              eg. hc <fqdn> info timestamp
              eg. hc <fqdn> info farm
  facts       lists facts, either provided as a fact file
              or grabbed from puppetdb.
              An extra param can be added for filtering
              eg. hc <fqdn> facts architecture
              eg. hc <fqdn> facts 'memory.*mb'
  files       list all files containing params affecting this fqdn
              (in more than commons)
  paths       list all file paths for files with params
  modules     list modules included in the farm where the node is
  params      list params for the node matching the fqdn
              An extra filter string can be added to limit the list
              use ruby regexp without the enclosing slashes
              eg. hc <fqdn> params postfix.*version
              eg. hc <fqdn> params '^postfix'
              eg. hc <fqdn> params 'version$'
  allparams   same as params but including the common.yaml params (huge)
              Also accepts a search string

Extra args:
  -f <plain|console|csv|yaml|rawyaml|json> - default console
  -p extraparam=what,anotherparam=this 
  -c <configfile>
  -h <hierafile>
  -b <basepath> default ./
  -e <encdir>
  -v - displays version
  -y <fact_file> - facts in yaml format
  -j <fact_file> - facts in json format
  -i - interactive mode
  -db - query puppetdb
  -nodb - do not query puppetdb

About facts aka. scope

Like with Hiera CLI you can use hieracles with defined top-scope variables. Those top-scope vars can be defined with:

  • -p extraparam=what;anotherparam=this
  • -y <fact_file> which takes the fact file from a yaml source created by facter -y on your node for example, but it can be written manually for experimentation purposes.
  • -j <fact_file> same as above, but with output of facter -j

You can define a default scope in your configuration file defaultscope in ~/.confg/hieracles/config.yml. For example:

---
classpath: farm_modules/%s/manifests/init.pp
hierafile: dev/hiera-local.yaml
encpath: enc 
defaultscope:
    operatingsystem: Debian
    lsbdistcodename: Jessie

In order the scope with be built from:

  • the config file
  • if -y <file> option (or -j) is present the defaultscope in the config file will be totally ignored
  • the -p key=value option with overide variable per variable

Note that if the scope var is not defined or if the file declared in hiera config is not found, this entry is silently ignored.

An option, -i enables the interactive mode in which you are prompted to fill up for undefined scope variables encountered in the hiera config file. This behavior can be made systematic by enabling interactive: true in hieracles configuration file.

Optionnaly connecting to a puppetDB

Configuration

When adding to the configuration file:

usedb: false
puppetdb:
  usessl: false
  host: puppetdb.example.com
  port: 8080

or for a ssl setup:

usedb: false
puppetdb:
  usessl: true
  host: puppetdb.example.com
  port: 8081
  key: path/to/key
  key_password: somepassword
  cert: path/to/cert
  ca_file: path/to/ca_file
  verify_peer: false

Note: the SSL config was not tested yet.

Usage

If you set usedb: false the hc commands will not query the puppetdb unless you pass the -db options.

If you set usedb: true the hc command will query the puppetdb by default and display extra informations for the queried node. This default behavior can be changed by passing the -nodb option on the commandline.

Impact

When usedb is true, a call to puppetdb will be made for all commands to retrieve facts if they are present for the queried node.

Extra commandline tool: ppdb

When hieracles is configured with parameters to connect to PuppetDB, you also can use the ppdb commandline to send direct queries to the database. Check man ppdb for more information.

Completion

There is a simple zsh completion file in tools/completion.

If you use oh-my-zsh put it in ~/.oh-my-zsh/completions

wget -O ~/.oh-my-zsh/completions/_hc https://raw.githubusercontent.com/Gandi/hieracles/master/tools/completion/zsh/_hc
echo 'compdef _hc hc "bundle exec hc"' >> ~/.zshrc
echo 'autoload -U _hc' >> ~/.zshrc

Otherwise

mkdir ~/.zsh-completions
wget -O ~/.zsh-completions/_hc https://raw.githubusercontent.com/Gandi/hieracles/master/tools/completion/zsh/_hc
echo 'fpath=(~/.zsh-completions $fpath)' >> ~/.zshrc
echo 'compdef _hc hc "bundle exec hc"' >> ~/.zshrc
echo 'autoload -U _hc' >> ~/.zshrc

Note: ppdb also has a completion file https://raw.githubusercontent.com/Gandi/hieracles/master/tools/completion/zsh/_ppdb

Debian packaging

A debian/ dir is included you can just use sbuild in here and it will build the .deb.

For new releases:

  • update the debian/changelog file

FreeBSD packaging

For new releases:

  • update the Makefile with new version number
  • in a FreeBSD jail or machine:
cd hieracles
git pull
cd ports/
make makesum
# test the stuff, get that there is no warning or what
portlint
make stage
make check-orphans
make package
make install
make deinstall
make clean
cd ..
shar `find rubygem-hieracles` > rubygem-hieracles.shar

Todo

  • add json format (done)
  • add override information in yaml format (done)
  • add a command to search for node according to a filter (done)
  • add a command to find all nodes that use a given module
  • add a command that finds all nodes for which a params is defined
  • detect unused params
  • create a repl, which at launch reads all data so the queries are blazing fast
  • adapt to other ENCs
  • adapt to PuppetDB storage

Other hiera tools

Authors

Hieracles original code is written by @mose.

License

Hieracles is available under MIT license. See LICENSE file for more details

Copyright

copyright (c) 2015 Gandi http://gandi.net