Skip to content

voxpupuli/puppet-cron

Puppet Cron Module

License Build Status Puppet Forge Puppet Forge - downloads Puppet Forge - scores

Notes

This module manages cronjobs by placing files in /etc/cron.d. rmueller-cron was a detached fork of torrancew/puppet-cron After v1.0.0, the module was migrated to Vox Pupuli where it is now maintained and released under the puppet namespace.

The current version (starting with v1.0.0) of this module requires Puppet 4.9.1 or greater. If you are using an older version of Puppet you can pin the version to v0.2.1 which was still compatible with much older Puppet versions. You can browse the documentation of that version in the v0.2.x branch here.

This module supports configuration of cronjobs via Hiera as well. For that you need to declare the cron class.

This module defines the following types:

  • cron::job - basic job resource
  • cron::job::multiple - basic job resource for multiple jobs per file
  • cron::hourly - wrapper for hourly jobs
  • cron::daily - wrapper for daily jobs
  • cron::weekly - wrapper for weekly jobs
  • cron::monthly - wrapper for monthly jobs

Installation

As usual use puppet module install puppet-cron to install it.

Usage

The title of the job (e.g. cron::job { 'title':) is completely arbitrary. However, there can only be one cron job by that name. The file in /etc/cron.d/ will be created with the $title as the file name. Keep that in mind when choosing the name to avoid overwriting existing system cronjobs and use characters that don't cause problems when used in filenames.

cron

If you want the class to automatically install the correct cron package you can declare the cron class. By default it will then install the right package. If you want to use Hiera to configure your cronjobs, you must declare the cron class.

You can disable the management of the cron package by setting the manage_package parameter to false.

You can also specify a different cron package name via package_name. By default we try to select the right one for your distribution. But in some cases (e.g. Gentoo) you might want to overwrite it here.

Examples:

  include cron

or:

  class { 'cron':
    manage_package => false,
  }

Add custom crontab run-parts

class { 'cron':
  manage_crontab    => true,
  crontab_run_parts => {'5min' => { 'user' => 'root', 'minute' => '*/5' }},
}

cron::job

cron::job creates generic jobs in /etc/cron.d. It allows specifying the following parameters:

  • ensure - optional - defaults to "present"
  • command - required - the command to execute
  • minute - optional - defaults to "*"
  • hour - optional - defaults to "*"
  • date - optional - defaults to "*"
  • month - optional - defaults to "*"
  • weekday - optional - defaults to "*"
  • special - optional - defaults to undef
  • user - optional - defaults to "root"
  • environment - optional - defaults to ""
  • mode - optional - defaults to "0600"
  • description - optional - defaults to undef

Example: This would create the file /etc/cron.d/mysqlbackup and run the command mysqldump -u root mydb as root at 2:40 AM every day:

  cron::job { 'mysqlbackup':
    minute      => '40',
    hour        => '2',
    date        => '*',
    month       => '*',
    weekday     => '*',
    user        => 'root',
    command     => 'mysqldump -u root mydb',
    environment => [ 'MAILTO=root', 'PATH="/usr/bin:/bin"', ],
    description => 'Mysql backup',
  }

Hiera example:

---
cron::job:
  'mysqlbackup':
    command: 'mysqldump -u root mydb'
    minute: 0
    hour: 0
    date: '*'
    month: '*'
    weekday: '*'
    user: root
    environment:
      - 'MAILTO=root'
      - 'PATH="/usr/bin:/bin"'
    description: 'Mysql backup'

cron::job::multiple

cron:job::multiple creates a file in /etc/cron.d with multiple cron jobs configured in it. It allows specifying the following parameters:

  • ensure - optional - defaults to "present"
  • jobs - required - an array of hashes of multiple cron jobs using a similar structure as cron::job-parameters
  • environment - optional - defaults to ""
  • mode - optional - defaults to "0600"

And the keys of the jobs hash are:

  • command - required - the command to execute
  • minute - optional - defaults to "*"
  • hour - optional - defaults to "*"
  • date - optional - defaults to "*"
  • month - optional - defaults to "*"
  • weekday - optional - defaults to "*"
  • special - optional - defaults to undef
  • user - optional - defaults to "root"
  • description - optional - defaults to undef

Example:

cron::job::multiple { 'test_cron_job_multiple':
  jobs => [
    {
      minute      => '55',
      hour        => '5',
      date        => '*',
      month       => '*',
      weekday     => '*',
      user        => 'rmueller',
      command     => '/usr/bin/uname',
      description => 'print system information',
    },
    {
      command     => '/usr/bin/sleep 1',
      description => 'Sleeping',
    },
    {
      command     => '/usr/bin/sleep 10',
      special     => 'reboot',
    },
  ],
  environment => [ 'PATH="/usr/sbin:/usr/bin:/sbin:/bin"' ],
}

Hiera example:

---
cron::job::multiple:
  'test_cron_job_multiple':
    jobs:
      - {
          minute: 55,
          hour: 5,
          date: '*',
          month: '*',
          weekday: '*',
          user: rmueller,
          command: '/usr/bin/uname',
          description: 'print system information',
        }
      - {
          command: '/usr/bin/sleep 1',
          description: 'Sleeping',
        }
      - {
          command: '/usr/bin/sleep 10',
          special: 'reboot',
        }

    environment:
      - 'PATH="/usr/sbin:/usr/bin:/sbin:/bin"'

That will generate the file /etc/cron.d/test_cron_job_multiple with essentially this content:

PATH="/usr/sbin:/usr/bin:/sbin:/bin"

55 5 * * *  rmueller  /usr/bin/uname
* * * * *  root  /usr/bin/sleep 1
@reboot  root  /usr/bin/sleep 10

cron::hourly

cron::hourly creates jobs in /etc/cron.d that run once per hour. It allows specifying the following parameters:

  • ensure - optional - defaults to "present"
  • command - required - the command to execute
  • minute - optional - defaults to "0"
  • user - optional - defaults to "root"
  • environment - optional - defaults to ""
  • mode - optional - defaults to "0600"
  • description - optional - defaults to undef

Example: This would create the file /etc/cron.d/mysqlbackup_hourly and run the command mysqldump -u root mydb as root on the 20th minute of every hour:

  cron::hourly { 'mysqlbackup_hourly':
    minute      => '20',
    user        => 'root',
    command     => 'mysqldump -u root mydb',
    environment => [ 'MAILTO=root', 'PATH="/usr/bin:/bin"', ],
  }

Hiera example:

---
cron::hourly:
  'mysqlbackup_hourly':
    minute: 20
    user: root
    command: 'mysqldump -u root mydb'
    environment:
      - 'MAILTO=root'
      - 'PATH="/usr/bin:/bin"'

cron::daily

cron::daily creates jobs in /etc/cron.d that run once per day. It allows specifying the following parameters:

  • ensure - optional - defaults to "present"
  • command - required - the command to execute
  • minute - optional - defaults to "0"
  • hour - optional - defaults to "0"
  • user - optional - defaults to "root"
  • environment - optional - defaults to ""
  • mode - optional - defaults to "0600"
  • description - optional - defaults to undef

Example: This would create the file /etc/cron.d/mysqlbackup_daily and run the command mysqldump -u root mydb as root at 2:40 AM every day, like the above generic example:

  cron::daily { 'mysqlbackup_daily':
    minute  => '40',
    hour    => '2',
    user    => 'root',
    command => 'mysqldump -u root mydb',
  }

Hiera example:

---
cron::daily:
  'mysqlbackup_daily':
    minute: 40
    hour: 2
    user: root
    command: 'mysqldump -u root mydb'

cron::weekly

cron::weekly creates jobs in /etc/cron.d that run once per week. It allows specifying the following parameters:

  • ensure - optional - defaults to "present"
  • command - required - the command to execute
  • minute - optional - defaults to "0"
  • hour - optional - defaults to "0"
  • weekday - optional - defaults to "0"
  • user - optional - defaults to "root"
  • environment - optional - defaults to ""
  • mode - optional - defaults to "0600"
  • description - optional - defaults to undef

Example: This would create the file /etc/cron.d/mysqlbackup_weekly and run the command mysqldump -u root mydb as root at 4:40 AM every Sunday, like the above generic example:

  cron::weekly { 'mysqlbackup_weekly':
    minute  => '40',
    hour    => '4',
    weekday => '0',
    user    => 'root',
    command => 'mysqldump -u root mydb',
  }

Hiera example:

---
cron::weekly:
  'mysqlbackup_weekly':
    minute: 40
    hour: 4
    weekday: 0
    user: root
    command: 'mysqldump -u root mydb'

cron::monthly

cron::monthly creates jobs in /etc/cron.d that run once per month. It allows specifying the following parameters:

  • ensure - optional - defaults to "present"
  • command - required - the command to execute
  • minute - optional - defaults to "0"
  • hour - optional - defaults to "0"
  • date - optional - defaults to "1"
  • user - optional - defaults to "root"
  • environment - optional - defaults to ""
  • mode - optional - defaults to "0600"
  • description - optional - defaults to undef

Example: This would create the file /etc/cron.d/mysqlbackup_monthly and run the command mysqldump -u root mydb as root at 3:40 AM the 1st of every month, like the above generic example:

  cron::monthly { 'mysqlbackup_monthly':
    minute  => '40',
    hour    => '3',
    date    => '1',
    user    => 'root',
    command => 'mysqldump -u root mydb',
  }

Hiera example:

---
cron::monthly:
  'mysqlbackup_monthly':
    minute: 40
    hour: 3
    date: 1
    user: root
    command: 'mysqldump -u root mydb'

Contributors

  • Kevin Goess (@kgoess) - Environment variable support + fixes
  • Andy Shinn (@andyshinn) - RedHat derivatives package name fix
  • Chris Weyl (@RsrchBoy) - Fixed Puppet 3.2 deprecation warnings
  • Mathew Archibald (@mattyindustries) - Fixed file ownership issues
  • The Community - Continued improvement of this module via bugs and patches