NOTE: This page was generated from the Puppet source code on 2018-12-20 09:14:50 -0800
This is the documentation for Puppet’s built-in resource types and providers. Additional resource types are distributed in Puppet modules.
You can find and install modules by browsing the Puppet Forge. See each module’s documentation for information on how to use its custom resource types. For more information about creating custom types, see Custom resources.
As of Puppet 6.0, some resource types were removed from Puppet and repackaged as individual modules. These supported type modules are still included in the
puppet-agentpackage, so you don’t have to download them from the Forge. See the complete list of affected types in the supported type modules section.
To manage resources on a target system, declare them in Puppet manifests. For more details, see the resources page of the Puppet language reference.
You can also browse and manage resources interactively using the
puppet resource subcommand; run puppet resource --help for more information.
All types have a special attribute called the namevar. This is the attribute used to uniquely identify a resource on the target system.
Each resource has a specific namevar attribute, which is listed on this page in each resource’s reference. If you don’t specify a value for the namevar, its value defaults to the resource’s title.
Example of a title as a default namevar:
file { '/etc/passwd':
owner => 'root',
group => 'root',
mode => '0644',
}
In this code, /etc/passwd is the title of the file resource.
The file type’s namevar is path. Because we didn’t provide a path value in
this example, the value defaults to the title, /etc/passwd.
Example of a namevar:
file { 'passwords':
path => '/etc/passwd',
owner => 'root',
group => 'root',
mode => '0644',
This example is functionally similar to the previous example. Its path
namevar attribute has an explicitly set value separate from the title, so
its name is still /etc/passwd.
Other Puppet code can refer to this resource as File['/etc/passwd'] to
declare relationships.
The attributes (sometimes called parameters) of a resource determine its
desired state. They either directly modify the system (internally, these are
called “properties”) or they affect how the resource behaves (for instance,
adding a search path for exec resources or controlling directory recursion
on file resources).
Providers implement the same resource type on different kinds of systems. They usually do this by calling out to external commands.
Although Puppet automatically selects an appropriate default provider, you
can override the default with the provider attribute. (For example, package
resources on Red Hat systems default to the yum provider, but you can specify
provider => gem to install Ruby libraries with the gem command.)
Providers often specify binaries that they require. Fully qualified binary paths indicate that the binary must exist at that specific path, and unqualified paths indicate that Puppet searches for the binary using the shell path.
Features are abilities that some providers might not support. Generally, a feature corresponds to some allowed values for a resource attribute.
This is often the case with the ensure attribute. In most types, Puppet
doesn’t create new resources when omitting ensure but still modifies existing
resources to match specifications in the manifest. However, in some types this
isn’t always the case, or additional values provide more granular control. For
example, if a package provider supports the purgeable feature, you can
specify ensure => purged to delete configuration files installed by the
package.
Resource types define the set of features they can use, and providers can declare which features they provide.
In Puppet 6.0, we removed some of Puppet’s built-in types and moved them into individual modules.
puppet-agentThe following types are included in supported modules on the Forge. However, they are also included in the puppet-agent package, so you do not have to install them separately. See each module’s README for detailed information about that type.
augeascronhostmountscheduled_taskselbooleanselmodulessh_authorized_keysshkeyyumrepozfszonezpoolThe following types are contained in modules that are maintained, but are not repackaged into Puppet agent. If you need to use them, you must install the modules separately.
The following types were deprecated with Puppet 6.0.0. They are available in modules, but are not updated. If you need to use them, you must install the modules separately.
computerinterface (Use the updated cisco_ios module instead.macauthorizationmcxrouter (Use the updated cisco_ios module instead.vlan (Use the updated cisco_ios module instead.The following types are located in the core Puppet code base.
Executes external commands.
Any command in an exec resource must be able to run multiple times
without causing harm — that is, it must be idempotent. There are three
main ways for an exec to be idempotent:
apt-get update.)onlyif, unless, or creates attribute, which prevents
Puppet from running the command unless some condition is met.refreshonly => true, which only allows Puppet to run the
command when some other resource is changed. (See the notes on refreshing
below.)A caution: There’s a widespread tendency to use collections of execs to manage resources that aren’t covered by an existing resource type. This works fine for simple tasks, but once your exec pile gets complex enough that you really have to think to understand what’s happening, you should consider developing a custom resource type instead, as it will be much more predictable and maintainable.
Duplication: Even though command is the namevar, Puppet allows multiple exec resources with the same command value.
Refresh: exec resources can respond to refresh events (via
notify, subscribe, or the ~> arrow). The refresh behavior of execs
is non-standard, and can be affected by the refresh and
refreshonly attributes:
refreshonly is set to true, the exec will only run when it receives an
event. This is the most reliable way to use refresh with execs.onlyif, unless, or creates condition
is no longer met after the first run, the second run will not occur.)refresh command, and receives an
event, it will run its normal command, then run its refresh command
(as long as any onlyif, unless, or creates conditions are still met
after the normal command finishes).onlyif, unless, or creates
attribute) and receives an event, it still will not run.noop => true, would otherwise have run, and receives
an event from a non-noop resource, it will run once (or run its refresh
command instead, if it has one).In short: If there’s a possibility of your exec receiving refresh events, it becomes doubly important to make sure the run conditions are restricted.
Autorequires: If Puppet is managing an exec’s cwd or the executable file used in an exec’s command, the exec resource will autorequire those files. If Puppet is managing the user that an exec should run as, the exec resource will autorequire that user.
exec { 'resource title':
command => # (namevar) The actual command to execute. Must either be...
creates => # A file to look for before running the command...
cwd => # The directory from which to run the command. If
environment => # An array of any additional environment variables
group => # The group to run the command as. This seems to...
logoutput => # Whether to log command output in addition to...
onlyif => # A test command that checks the state of the...
path => # The search path used for command execution...
provider => # The specific backend to use for this `exec...
refresh => # An alternate command to run when the `exec...
refreshonly => # The command should only be run as a refresh...
returns => # The expected exit code(s). An error will be...
timeout => # The maximum time the command should take. If...
tries => # The number of times execution of the command...
try_sleep => # The time to sleep in seconds between 'tries'....
umask => # Sets the umask to be used while executing this...
unless => # A test command that checks the state of the...
user => # The user to run the command as. > **Note:*...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The actual command to execute. Must either be fully qualified
or a search path for the command must be provided. If the command
succeeds, any output produced will be logged at the instance’s
normal log level (usually notice), but if the command fails
(meaning its return code does not match the specified code) then
any output is logged at the err log level.
Multiple exec resources can use the same command value; Puppet only uses the resource title to ensure execs are unique.”
The directory from which to run the command. If this directory does not exist, the command will fail.
An array of any additional environment variables you want to set for a
command, such as [ 'HOME=/root', 'MAIL=root@example.com'].
Note that if you use this to set PATH, it will override the path
attribute. Multiple environment variables should be specified as an
array.
The group to run the command as. This seems to work quite haphazardly on different platforms – it is a platform issue not a Ruby or Puppet one, since the same variety exists when running commands as different users in the shell.
Whether to log command output in addition to logging the
exit code. Defaults to on_failure, which only logs the output
when the command has an exit code that does not match any value
specified by the returns attribute. As with any resource type,
the log level can be controlled with the loglevel metaparameter.
Default: on_failure
Allowed values:
truefalseon_failureA test command that checks the state of the target system and restricts
when the exec can run. If present, Puppet runs this test command
first, and only runs the main command if the test has an exit code of 0
(success). For example:
exec { 'logrotate':
path => '/usr/bin:/usr/sbin:/bin',
provider => shell,
onlyif => 'test `du /var/log/messages | cut -f1` -gt 100000',
}
This would run logrotate only if that test returns true.
Note that this test command runs with the same provider, path, cwd, user, and group as the main command. If the path isn’t set, you
must fully qualify the command’s name.
This parameter can also take an array of commands. For example:
onlyif => ['test -f /tmp/file1', 'test -f /tmp/file2'],
This exec would only run if every command in the array has an
exit code of 0 (success).
The search path used for command execution. Commands must be fully qualified if no path is specified. Paths can be specified as an array or as a ‘
An alternate command to run when the exec receives a refresh event
from another resource. By default, Puppet runs the main command again.
For more details, see the notes about refresh behavior above, in the
description for this resource type.
Note that this alternate command runs with the same provider, path,
user, and group as the main command. If the path isn’t set, you
must fully qualify the command’s name.
The command should only be run as a refresh mechanism for when a dependent object is changed. It only makes sense to use this option when this command depends on some other object; it is useful for triggering an action:
# Pull down the main aliases file
file { '/etc/aliases':
source => 'puppet://server/module/aliases',
}
# Rebuild the database, but only when the file changes
exec { newaliases:
path => ['/usr/bin', '/usr/sbin'],
subscribe => File['/etc/aliases'],
refreshonly => true,
}
Note that only subscribe and notify can trigger actions, not require,
so it only makes sense to use refreshonly with subscribe or notify.
Allowed values:
truefalse(Property: This attribute represents concrete state on the target system.)
The expected exit code(s). An error will be returned if the executed command has some other exit code. Can be specified as an array of acceptable exit codes or a single value.
On POSIX systems, exit codes are always integers between 0 and 255.
On Windows, most exit codes should be integers between 0 and 2147483647.
Larger exit codes on Windows can behave inconsistently across different tools. The Win32 APIs define exit codes as 32-bit unsigned integers, but both the cmd.exe shell and the .NET runtime cast them to signed integers. This means some tools will report negative numbers for exit codes above 2147483647. (For example, cmd.exe reports 4294967295 as -1.) Since Puppet uses the plain Win32 APIs, it will report the very large number instead of the negative number, which might not be what you expect if you got the exit code from a cmd.exe session.
Microsoft recommends against using negative/very large exit codes, and you should avoid them when possible. To convert a negative exit code to the positive one Puppet will use, add it to 4294967296.
Default: 0
The maximum time the command should take. If the command takes longer than the timeout, the command is considered to have failed and will be stopped. The timeout is specified in seconds. The default timeout is 300 seconds and you can set it to 0 to disable the timeout.
Default: 300
The number of times execution of the command should be tried. This many attempts will be made to execute the command until an acceptable return code is returned. Note that the timeout parameter applies to each try rather than to the complete set of tries.
Default: 1
The time to sleep in seconds between ‘tries’.
Default: 0
Sets the umask to be used while executing this command
A test command that checks the state of the target system and restricts
when the exec can run. If present, Puppet runs this test command
first, then runs the main command unless the test has an exit code of 0
(success). For example:
exec { '/bin/echo root >> /usr/lib/cron/cron.allow':
path => '/usr/bin:/usr/sbin:/bin',
unless => 'grep root /usr/lib/cron/cron.allow 2>/dev/null',
}
This would add root to the cron.allow file (on Solaris) unless
grep determines it’s already there.
Note that this test command runs with the same provider, path, cwd, user, and group as the main command. If the path isn’t set, you
must fully qualify the command’s name.
This parameter can also take an array of commands. For example:
unless => ['test -f /tmp/file1', 'test -f /tmp/file2'],
This exec would only run if every command in the array has a
non-zero exit code.
Note that if you use this attribute, any error output is not captured due to a bug within Ruby. If you use Puppet to create this user, the exec automatically requires the user, as long as it is specified by name.
The $HOME environment variable is not automatically set when using this attribute.
Executes external binaries directly, without passing through a shell or performing any interpolation. This is a safer and more predictable way to execute most commands, but prevents the use of globbing and shell built-ins (including control logic like “for” and “if” statements).
feature == posix["feature", "posix"] == Passes the provided command through /bin/sh; only available on
POSIX systems. This allows the use of shell globbing and built-ins, and
does not require that the path to a command be fully-qualified. Although
this can be more convenient than the posix provider, it also means that
you need to be more careful with escaping; as ever, with great power comes
etc. etc.
This provider closely resembles the behavior of the exec type
in Puppet 0.25.x.
feature == posixExecute external binaries on Windows systems. As with the posix
provider, this provider directly calls the command with the arguments
given, without passing it through a shell or performing any interpolation.
To use shell built-ins — that is, to emulate the shell provider on
Windows — a command must explicitly invoke the shell:
exec {'echo foo':
command => 'cmd.exe /c echo "foo"',
}
If no extension is specified for a command, Windows will use the PATHEXT
environment variable to locate the executable.
Note on PowerShell scripts: PowerShell’s default restricted
execution policy doesn’t allow it to run saved scripts. To run PowerShell
scripts, specify the remotesigned execution policy as part of the
command:
exec { 'test':
path => 'C:/Windows/System32/WindowsPowerShell/v1.0',
command => 'powershell -executionpolicy remotesigned -file C:/test.ps1',
}
operatingsystem == windows["operatingsystem", "windows"] == Manages files, including their content, ownership, and permissions.
The file type can manage normal files, directories, and symlinks; the
type should be specified in the ensure attribute.
File contents can be managed directly with the content attribute, or
downloaded from a remote source using the source attribute; the latter
can also be used to recursively serve directories (when the recurse
attribute is set to true or local). On Windows, note that file
contents are managed in binary mode; Puppet never automatically translates
line endings.
Autorequires: If Puppet is managing the user or group that owns a file, the file resource will autorequire them. If Puppet is managing any parent directories of a file, the file resource will autorequire them.
Warning: Enabling recurse on directories containing large numbers of files slows agent runs. To manage file attributes for many files, consider using alternative methods such as the chmod_r, chown_r, or recursive_file_permissions modules from the Forge.
file { 'resource title':
path => # (namevar) The path to the file to manage. Must be fully...
backup => # Whether (and how) file content should be backed...
force => # Perform the file operation even if it will...
ignore => # A parameter which omits action on files matching
links => # How to handle links during file actions. During
purge => # Whether unmanaged files should be purged. This...
recurse => # Whether to recursively manage the _contents_ of...
recurselimit => # How far Puppet should descend into...
replace => # Whether to replace a file or symlink that...
show_diff => # Whether to display differences when the file...
sourceselect => # Whether to copy all valid sources, or just the...
validate_cmd => # A command for validating the file's syntax...
validate_replacement => # The replacement string in a `validate_cmd` that...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The path to the file to manage. Must be fully qualified.
On Windows, the path should include the drive letter and should use / as
the separator character (rather than \\).
(Property: This attribute represents concrete state on the target system.)
Whether the file should exist, and if so what kind of file it should be.
Possible values are present, absent, file, directory, and link.
present accepts any form of file existence, and creates a
normal file if the file is missing. (The file will have no content
unless the content or source attribute is used.)absent ensures the file doesn’t exist, and deletes it if necessary.file ensures it’s a normal file, and enables use of the content or
source attribute.directory ensures it’s a directory, and enables use of the source,
recurse, recurselimit, ignore, and purge attributes.link ensures the file is a symlink, and requires that you also
set the target attribute. Symlinks are supported on all Posix
systems and on Windows Vista / 2008 and higher. On Windows, managing
symlinks requires Puppet agent’s user account to have the “Create
Symbolic Links” privilege; this can be configured in the “User Rights
Assignment” section in the Windows policy editor. By default, Puppet
agent runs as the Administrator account, which has this privilege.Puppet avoids destroying directories unless the force attribute is set
to true. This means that if a file is currently a directory, setting
ensure to anything but directory or present will cause Puppet to
skip managing the resource and log either a notice or an error.
There is one other non-standard value for ensure. If you specify the
path to another file as the ensure value, it is equivalent to specifying
link and using that path as the target:
# Equivalent resources:
file { '/etc/inetd.conf':
ensure => '/etc/inet/inetd.conf',
}
file { '/etc/inetd.conf':
ensure => link,
target => '/etc/inet/inetd.conf',
}
However, we recommend using link and target explicitly, since this
behavior can be harder to read and is
deprecated
as of Puppet 4.3.0.
Valid values are absent (also called false), file, present, directory, link. Values can match /./.
Whether (and how) file content should be backed up before being replaced.
This attribute works best as a resource default in the site manifest
(File { backup => main }), so it can affect all file resources.
false, file content won’t be backed up.., such as .puppet-bak, Puppet will
use copy the file in the same directory with that value as the extension
of the backup. (A value of true is a synonym for .puppet-bak.)filebucket resource type for more details.
(This is the preferred method for backup, since it can be centralized
and queried.)Default value: puppet, which backs up to a filebucket of the same name.
(Puppet automatically creates a local filebucket named puppet if one
doesn’t already exist.)
Backing up to a local filebucket isn’t particularly useful. If you want
to make organized use of backups, you will generally want to use the
puppet master server’s filebucket service. This requires declaring a
filebucket resource and a resource default for the backup attribute
in site.pp:
# /etc/puppetlabs/puppet/manifests/site.pp
filebucket { 'main':
path => false, # This is required for remote filebuckets.
server => 'puppet.example.com', # Optional; defaults to the configured puppet master.
}
File { backup => main, }
If you are using multiple puppet master servers, you will want to centralize the contents of the filebucket. Either configure your load balancer to direct all filebucket traffic to a single master, or use something like an out-of-band rsync task to synchronize the content on all masters.
Default: puppet
The checksum type to use when determining whether to replace a file’s contents.
The default checksum type is md5.
Valid values are md5, md5lite, sha224, sha256, sha256lite, sha384, sha512, mtime, ctime, none.
(Property: This attribute represents concrete state on the target system.)
The checksum of the source contents. Only md5, sha256, sha224, sha384 and sha512 are supported when specifying this parameter. If this parameter is set, source_permissions will be assumed to be false, and ownership and permissions will not be read from source.
(Property: This attribute represents concrete state on the target system.)
The desired contents of a file, as a string. This attribute is mutually
exclusive with source and target.
Newlines and tabs can be specified in double-quoted strings using standard escaped syntax — \n for a newline, and \t for a tab.
With very small files, you can construct content strings directly in the manifest…
define resolve($nameserver1, $nameserver2, $domain, $search) {
$str = "search ${search}
domain ${domain}
nameserver ${nameserver1}
nameserver ${nameserver2}
"
file { '/etc/resolv.conf':
content => $str,
}
}
…but for larger files, this attribute is more useful when combined with the template or file function.
(Property: This attribute represents concrete state on the target system.)
A read-only state to check the file ctime. On most modern *nix-like systems, this is the time of the most recent change to the owner, group, permissions, or content of the file.
Perform the file operation even if it will destroy one or more directories.
You must use force in order to:
purge subdirectoriesensure => absentDefault: false
Allowed values:
truefalseyesnoA parameter which omits action on files matching
specified patterns during recursion. Uses Ruby’s builtin globbing
engine, so shell metacharacters such as [a-z]* are fully supported.
Matches that would descend into the directory structure are ignored,
such as */*.
How to handle links during file actions. During file copying,
follow will copy the target file instead of the link and manage
will copy the link itself. When not copying, manage will manage
the link, and follow will manage the file to which the link points.
Default: manage
Allowed values:
followmanage(Property: This attribute represents concrete state on the target system.)
The desired permissions mode for the file, in symbolic or numeric notation. This value must be specified as a string; do not use un-quoted numbers to represent file modes.
If the mode is omitted (or explicitly set to undef), Puppet does not
enforce permissions on existing files and creates new files with
permissions of 0644.
The file type uses traditional Unix permission schemes and translates
them to equivalent permissions for systems which represent permissions
differently, including Windows. For detailed ACL controls on Windows,
you can leave mode unmanaged and use
the puppetlabs/acl module.
Numeric modes should use the standard octal notation of
<SETUID/SETGID/STICKY><OWNER><GROUP><OTHER> (for example, “0644”).
Symbolic modes should be represented as a string of comma-separated
permission clauses, in the form <WHO><OP><PERM>:
Thus, mode "0664" could be represented symbolically as either a=r,ug+w
or ug=rw,o=r. However, symbolic modes are more expressive than numeric
modes: a mode only affects the specified bits, so mode => 'ug+w' will
set the user and group write bits, without affecting any other bits.
See the manual page for GNU or BSD chmod for more details
on numeric and symbolic modes.
On Windows, permissions are translated as follows:
FILE_GENERIC_READ,
FILE_GENERIC_WRITE, and FILE_GENERIC_EXECUTE access rights; a
file’s owner always has the FULL_CONTROL right(Property: This attribute represents concrete state on the target system.)
A read-only state to check the file mtime. On *nix-like systems, this is the time of the most recent change to the content of the file.
(Property: This attribute represents concrete state on the target system.)
The user to whom the file should belong. Argument can be a user name or a user ID.
On Windows, a group (such as “Administrators”) can be set as a file’s owner
and a user (such as “Administrator”) can be set as a file’s group; however,
a file’s owner and group shouldn’t be the same. (If the owner is also
the group, files with modes like "0640" will cause log churn, as they
will always appear out of sync.)
The specific backend to use for this file
resource. You will seldom need to specify this — Puppet will usually
discover the appropriate provider for your platform.
Available providers are:
Whether unmanaged files should be purged. This option only makes
sense when ensure => directory and recurse => true.
source
attribute, purge => true will automatically purge any files
that are not in the source directory.purge => true will purge any files that aren’t being
specifically managed.If you have a filebucket configured, the purged files will be uploaded, but if you do not, this will destroy data.
Unless force => true is set, purging will not delete directories,
although it will delete the files they contain.
If recurselimit is set and you aren’t using force => true, purging
will obey the recursion limit; files in any subdirectories deeper than the
limit will be treated as unmanaged and left alone.
Default: false
Allowed values:
truefalseyesnoWhether to recursively manage the contents of a directory. This attribute
is only used when ensure => directory is set. The allowed values are:
false — The default behavior. The contents of the directory will not be
automatically managed.remote — If the source attribute is set, Puppet will automatically
manage the contents of the source directory (or directories), ensuring
that equivalent files and directories exist on the target system and
that their contents match.
Using remote will disable the purge attribute, but results in faster
catalog application than recurse => true.
The source attribute is mandatory when recurse => remote.
true — If the source attribute is set, this behaves similarly to
recurse => remote, automatically managing files from the source directory.
This also enables the purge attribute, which can delete unmanaged
files from a directory. See the description of purge for more details.
The source attribute is not mandatory when using recurse => true, so you
can enable purging in directories where all files are managed individually.
By default, setting recurse to remote or true manages all
subdirectories. You can use the recurselimit attribute to limit the
recursion depth.
Allowed values:
truefalseremoteHow far Puppet should descend into subdirectories, when using
ensure => directory and either recurse => true or recurse => remote.
The recursion limit affects which files will be copied from the source
directory, as well as which files can be purged when purge => true.
Setting recurselimit => 0 is the same as setting recurse => false —
Puppet will manage the directory, but all of its contents will be treated
as unmanaged.
Setting recurselimit => 1 will manage files and directories that are
directly inside the directory, but will not manage the contents of any
subdirectories.
Setting recurselimit => 2 will manage the direct contents of the
directory, as well as the contents of the first level of subdirectories.
This pattern continues for each incremental value of recurselimit.
Allowed values:
/^[0-9]+$/Whether to replace a file or symlink that already exists on the local system but
whose content doesn’t match what the source or content attribute
specifies. Setting this to false allows file resources to initialize files
without overwriting future changes. Note that this only affects content;
Puppet will still manage ownership and permissions.
Default: true
Allowed values:
true or yesfalse or noIf this is set then Puppet will not ask SELinux (via matchpathcon) to supply defaults for the SELinux attributes (seluser, selrole, seltype, and selrange). In general, you should leave this set at its default and only set it to true when you need Puppet to not try to fix SELinux labels automatically.
Valid values are true, false.
(Property: This attribute represents concrete state on the target system.)
What the SELinux range component of the context of the file should be.
Any valid SELinux range component is accepted. For example s0 or
SystemHigh. If not specified it defaults to the value returned by
matchpathcon for the file, if any exists. Only valid on systems with
SELinux support enabled and that have support for MCS (Multi-Category
Security).
(Property: This attribute represents concrete state on the target system.)
What the SELinux role component of the context of the file should be.
Any valid SELinux role component is accepted. For example role_r.
If not specified it defaults to the value returned by matchpathcon for
the file, if any exists. Only valid on systems with SELinux support
enabled.
(Property: This attribute represents concrete state on the target system.)
What the SELinux type component of the context of the file should be.
Any valid SELinux type component is accepted. For example tmp_t.
If not specified it defaults to the value returned by matchpathcon for
the file, if any exists. Only valid on systems with SELinux support
enabled.
(Property: This attribute represents concrete state on the target system.)
What the SELinux user component of the context of the file should be.
Any valid SELinux user component is accepted. For example user_u.
If not specified it defaults to the value returned by matchpathcon for
the file, if any exists. Only valid on systems with SELinux support
enabled.
Whether to display differences when the file changes, defaulting to
true. This parameter is useful for files that may contain passwords or
other secret data, which might otherwise be included in Puppet reports or
other insecure outputs. If the global show_diff setting
is false, then no diffs will be shown even if this parameter is true.
Default: true
Allowed values:
truefalseyesnoWhether to copy all valid sources, or just the first one. This parameter
only affects recursive directory copies; by default, the first valid
source is the only one used, but if this parameter is set to all, then
all valid sources will have all of their contents copied to the local
system. If a given file exists in more than one source, the version from
the earliest source in the list will be used.
Default: first
Allowed values:
firstall(Property: This attribute represents concrete state on the target system.)
The target for creating a link. Currently, symlinks are the
only type supported. This attribute is mutually exclusive with source
and content.
Symlink targets can be relative, as well as absolute:
# (Useful on Solaris)
file { '/etc/inetd.conf':
ensure => link,
target => 'inet/inetd.conf',
}
Directories of symlinks can be served recursively by instead using the
source attribute, setting ensure to directory, and setting the
links attribute to manage.
Allowed values:
notlink/./(Property: This attribute represents concrete state on the target system.)
A read-only state to check the file type.
A command for validating the file’s syntax before replacing it. If
Puppet would need to rewrite a file due to new source or content, it
will check the new content’s validity first. If validation fails, the file
resource will fail.
This command must have a fully qualified path, and should contain a
percent (%) token where it would expect an input file. It must exit 0
if the syntax is correct, and non-zero otherwise. The command will be
run on the target system while applying the catalog, not on the puppet master.
Example:
file { '/etc/apache2/apache2.conf':
content => 'example',
validate_cmd => '/usr/sbin/apache2 -t -f %',
}
This would replace apache2.conf only if the test returned true.
Note that if a validation command requires a % as part of its text,
you can specify a different placeholder token with the
validate_replacement attribute.
The replacement string in a validate_cmd that will be replaced
with an input file name.
Default: %
Uses POSIX functionality to manage file ownership and permissions.
feature == posixmanages_symlinksUses Microsoft Windows functionality to manage file ownership and permissions.
operatingsystem == windowsmanages_symlinksAvailable features:
manages_symlinks — The provider can manage symbolic links.Provider support:
| Provider | manages symlinks |
|---|---|
| posix | X |
| windows | X |
A repository for storing and retrieving file content by MD5 checksum. Can be local to each agent node, or centralized on a puppet master server. All puppet masters provide a filebucket service that agent nodes can access via HTTP, but you must declare a filebucket resource before any agents will do so.
Filebuckets are used for the following features:
file type’s backup attribute is set to
the name of a filebucket, Puppet will back up the old content whenever
it rewrites a file; see the documentation for the file type for more
details. These backups can be used for manual recovery of content, but
are more commonly used to display changes and differences in a tool like
Puppet Dashboard.To use a central filebucket for backups, you will usually want to declare
a filebucket resource and a resource default for the backup attribute
in site.pp:
# /etc/puppetlabs/puppet/manifests/site.pp
filebucket { 'main':
path => false, # This is required for remote filebuckets.
server => 'puppet.example.com', # Optional; defaults to the configured puppet master.
}
File { backup => main, }
Puppet master servers automatically provide the filebucket service, so
this will work in a default configuration. If you have a heavily
restricted auth.conf file, you may need to allow access to the
file_bucket_file endpoint.
filebucket { 'resource title':
name => # (namevar) The name of the...
path => # The path to the _local_ filebucket; defaults to...
port => # The port on which the remote server is...
server => # The server providing the remote filebucket...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The name of the filebucket.
(↑ Back to filebucket attributes)
The path to the local filebucket; defaults to the value of the
clientbucketdir setting. To use a remote filebucket, you must set
this attribute to false.
(↑ Back to filebucket attributes)
The port on which the remote server is listening. Defaults to the
value of the masterport setting, which is usually 8140.
(↑ Back to filebucket attributes)
The server providing the remote filebucket service. Defaults to the
value of the server setting (that is, the currently configured
puppet master server).
This setting is consulted only if the path attribute is set to false.
(↑ Back to filebucket attributes)
Manage groups. On most platforms this can only create groups. Group membership must be managed on individual users.
On some platforms such as OS X, group membership is managed as an attribute of the group, not the user record. Providers must have the feature ‘manages_members’ to manage the ‘members’ property of a group record.
group { 'resource title':
name => # (namevar) The group name. While naming limitations vary by
ensure => # Create or remove the group. Default: `present`
allowdupe => # Whether to allow duplicate GIDs. Default...
attribute_membership => # AIX only. Configures the behavior of the...
attributes => # Specify group AIX attributes, as an array of...
auth_membership => # Configures the behavior of the `members...
forcelocal => # Forces the management of local accounts when...
gid => # The group ID. Must be specified numerically....
ia_load_module => # The name of the I&A module to use to manage this
members => # The members of the group. For platforms or...
provider => # The specific backend to use for this `group...
system => # Whether the group is a system group with lower...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The group name. While naming limitations vary by operating system, it is advisable to restrict names to the lowest common denominator, which is a maximum of 8 characters beginning with a letter.
Note that Puppet considers group names to be case-sensitive, regardless of the platform’s own rules; be sure to always use the same case when referring to a given group.
(Property: This attribute represents concrete state on the target system.)
Create or remove the group.
Default: present
Allowed values:
presentabsentWhether to allow duplicate GIDs.
Default: false
Allowed values:
truefalseyesnoAIX only. Configures the behavior of the attributes parameter.
minimum (default) — The provided list of attributes is partial, and Puppet
ignores any attributes that aren’t listed there.inclusive — The provided list of attributes is comprehensive, and
Puppet purges any attributes that aren’t listed there.Default: minimum
Allowed values:
inclusiveminimum(Property: This attribute represents concrete state on the target system.)
Specify group AIX attributes, as an array of 'key=value' strings. This
parameter’s behavior can be configured with attribute_membership.
Requires features manages_aix_lam.
Configures the behavior of the members parameter.
false (default) — The provided list of group members is partial,
and Puppet ignores any members that aren’t listed there.true — The provided list of of group members is comprehensive, and
Puppet purges any members that aren’t listed there.Default: false
Allowed values:
truefalseyesnoForces the management of local accounts when accounts are also being managed by some other NSS.
Default: false
Allowed values:
truefalseyesno(Property: This attribute represents concrete state on the target system.)
The group ID. Must be specified numerically. If no group ID is specified when creating a new group, then one will be chosen automatically according to local system standards. This will likely result in the same group having different GIDs on different systems, which is not recommended.
On Windows, this property is read-only and will return the group’s security identifier (SID).
The name of the I&A module to use to manage this user
Requires features manages_aix_lam.
(Property: This attribute represents concrete state on the target system.)
The members of the group. For platforms or directory services where group
membership is stored in the group objects, not the users. This parameter’s
behavior can be configured with auth_membership.
Requires features manages_members.
The specific backend to use for this group
resource. You will seldom need to specify this — Puppet will usually
discover the appropriate provider for your platform.
Available providers are:
Whether the group is a system group with lower GID.
Default: false
Allowed values:
truefalseyesnoGroup management for AIX.
/usr/sbin/lsgroup, /usr/bin/mkgroup, /usr/sbin/rmgroup, /usr/bin/chgroupoperatingsystem == aixoperatingsystem == aixmanages_aix_lam, manages_membersGroup management using DirectoryService on OS X.
/usr/bin/dscloperatingsystem == darwinoperatingsystem == darwinmanages_membersGroup management via groupadd and its ilk. The default for most platforms.
groupadd, groupdel, groupmodsystem_groupsGroup management via LDAP.
This provider requires that you have valid values for all of the
LDAP-related settings in puppet.conf, including ldapbase. You will
almost definitely need settings for ldapuser and ldappassword in order
for your clients to write to LDAP.
Note that this provider will automatically generate a GID for you if you do not specify one, but it is a potentially expensive operation, as it iterates across all existing groups to pick the appropriate next one.
feature == ldap, false == (Puppet[:ldapuser] == "")Group management via pw on FreeBSD and DragonFly BSD.
pwoperatingsystem == [:freebsd, :dragonfly]operatingsystem == freebsd, dragonfly.manages_members.Local group management for Windows. Group members can be both users and groups. Additionally, local groups can contain domain users.
operatingsystem == windowsoperatingsystem == windowsmanages_members.Available features:
libuser — Allows local groups to be managed on systems that also use some other remote NSS method of managing accounts.manages_aix_lam — The provider can manage AIX Loadable Authentication Module (LAM) system.manages_members — For directories where membership is an attribute of groups not users.system_groups — The provider allows you to create system groups with lower GIDs.Provider support:
| Provider | libuser | manages aix lam | manages members | system groups |
|---|---|---|---|---|
| aix | X | X | ||
| directoryservice | X | |||
| groupadd | X | X | ||
| ldap | ||||
| pw | X | |||
| windows_adsi | X |
Sends an arbitrary message to the agent run-time log.
notify { 'resource title':
name => # (namevar) An arbitrary tag for your own reference; the...
message => # The message to be sent to the...
withpath => # Whether to show the full object path. Default...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
An arbitrary tag for your own reference; the name of the message.
(Property: This attribute represents concrete state on the target system.)
The message to be sent to the log.
Whether to show the full object path.
Default: false
Allowed values:
truefalseManage packages. There is a basic dichotomy in package
support right now: Some package types (such as yum and apt) can
retrieve their own package files, while others (such as rpm and sun)
cannot. For those package formats that cannot retrieve their own files,
you can use the source parameter to point to the correct file.
Puppet will automatically guess the packaging format that you are
using based on the platform you are on, but you can override it
using the provider parameter; each provider defines what it
requires in order to function, and you must meet those requirements
to use a given provider.
You can declare multiple package resources with the same name, as long
as they specify different providers and have unique titles.
Note that you must use the title to make a reference to a package
resource; Package[<NAME>] is not a synonym for Package[<TITLE>] like
it is for many other resource types.
Autorequires: If Puppet is managing the files specified as a
package’s adminfile, responsefile, or source, the package
resource will autorequire those files.
package { 'resource title':
name => # (namevar) The package name. This is the name that the...
provider => # (namevar) The specific backend to use for this `package...
ensure => # What state the package should be in. On...
adminfile => # A file containing package defaults for...
allow_virtual => # Specifies if virtual package names are allowed...
allowcdrom => # Tells apt to allow cdrom sources in the...
category => # A read-only parameter set by the...
configfiles => # Whether to keep or replace modified config files
description => # A read-only parameter set by the...
flavor => # OpenBSD supports 'flavors', which are further...
install_options => # An array of additional options to pass when...
instance => # A read-only parameter set by the...
package_settings => # Settings that can change the contents or...
platform => # A read-only parameter set by the...
reinstall_on_refresh => # Whether this resource should respond to refresh...
responsefile => # A file containing any necessary answers to...
root => # A read-only parameter set by the...
source => # Where to find the package file. This is only...
status => # A read-only parameter set by the...
uninstall_options => # An array of additional options to pass when...
vendor => # A read-only parameter set by the...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The package name. This is the name that the packaging system uses internally, which is sometimes (especially on Solaris) a name that is basically useless to humans. If a package goes by several names, you can use a single title and then set the name conditionally:
# In the 'openssl' class
$ssl = $operatingsystem ? {
solaris => SMCossl,
default => openssl
}
package { 'openssl':
ensure => installed,
name => $ssl,
}
...
$ssh = $operatingsystem ? {
solaris => SMCossh,
default => openssh
}
package { 'openssh':
ensure => installed,
name => $ssh,
require => Package['openssl'],
}
(↑ Back to package attributes)
(Secondary namevar: This resource type allows you to manage multiple resources with the same name as long as their providers are different.)
The specific backend to use for this package
resource. You will seldom need to specify this — Puppet will usually
discover the appropriate provider for your platform.
Available providers are:
aixappdmgappleaptaptitudeaptrpmblastwavednfdpkgfinkfreebsdgemhpuxmacportsnimopenbsdopkgpacmanpip3pippkgpkgdmgpkginpkgngpkgutilportageportsportupgradepuppet_gemrpmrugsunsunfreewaretdnfup2dateurpmiwindowsyumzypper(↑ Back to package attributes)
(Property: This attribute represents concrete state on the target system.)
What state the package should be in. On packaging systems that can
retrieve new packages on their own, you can choose which package to
retrieve by specifying a version number or latest as the ensure
value. On packaging systems that manage configuration files separately
from “normal” system files, you can uninstall config files by
specifying purged as the ensure value. This defaults to installed.
Version numbers must match the full version to install, including
release if the provider uses a release moniker. Ranges or semver
patterns are not accepted except for the gem package provider. For
example, to install the bash package from the rpm
bash-4.1.2-29.el6.x86_64.rpm, use the string '4.1.2-29.el6'.
Default: installed
Allowed values:
presentabsentpurgedheldinstalledlatest/./(↑ Back to package attributes)
A file containing package defaults for installing packages.
This attribute is only used on Solaris. Its value should be a path to a
local file stored on the target system. Solaris’s package tools expect
either an absolute file path or a relative path to a file in
/var/sadm/install/admin.
The value of adminfile will be passed directly to the pkgadd or
pkgrm command with the -a <ADMINFILE> option.
(↑ Back to package attributes)
Specifies if virtual package names are allowed for install and uninstall.
Default: true
Allowed values:
truefalseyesno(↑ Back to package attributes)
Tells apt to allow cdrom sources in the sources.list file. Normally apt will bail if you try this.
Allowed values:
truefalse(↑ Back to package attributes)
A read-only parameter set by the package.
(↑ Back to package attributes)
Whether to keep or replace modified config files when installing or
upgrading a package. This only affects the apt and dpkg providers.
Default: keep
Allowed values:
keepreplace(↑ Back to package attributes)
A read-only parameter set by the package.
(↑ Back to package attributes)
OpenBSD supports ‘flavors’, which are further specifications for which type of package you want.
(↑ Back to package attributes)
An array of additional options to pass when installing a package. These
options are package-specific, and should be documented by the software
vendor. One commonly implemented option is INSTALLDIR:
package { 'mysql':
ensure => installed,
source => 'N:/packages/mysql-5.5.16-winx64.msi',
install_options => [ '/S', { 'INSTALLDIR' => 'C:\\mysql-5.5' } ],
}
Each option in the array can either be a string or a hash, where each key and value pair are interpreted in a provider specific way. Each option will automatically be quoted when passed to the install command.
With Windows packages, note that file paths in an install option must
use backslashes. (Since install options are passed directly to the
installation command, forward slashes won’t be automatically converted
like they are in file resources.) Note also that backslashes in
double-quoted strings must be escaped and backslashes in single-quoted
strings can be escaped.
Requires features install_options.
(↑ Back to package attributes)
A read-only parameter set by the package.
(↑ Back to package attributes)
(Property: This attribute represents concrete state on the target system.)
Settings that can change the contents or configuration of a package.
The formatting and effects of package_settings are provider-specific; any provider that implements them must explain how to use them in its documentation. (Our general expectation is that if a package is installed but its settings are out of sync, the provider should re-install that package with the desired settings.)
An example of how package_settings could be used is FreeBSD’s port build options — a future version of the provider could accept a hash of options, and would reinstall the port if the installed version lacked the correct settings.
package { 'www/apache22':
package_settings => { 'SUEXEC' => false }
}
Again, check the documentation of your platform’s package provider to see the actual usage.
Requires features package_settings.
(↑ Back to package attributes)
A read-only parameter set by the package.
(↑ Back to package attributes)
Whether this resource should respond to refresh events (via subscribe,
notify, or the ~> arrow) by reinstalling the package. Only works for
providers that support the reinstallable feature.
This is useful for source-based distributions, where you may want to recompile a package if the build options change.
If you use this, be careful of notifying classes when you want to restart services. If the class also contains a refreshable package, doing so could cause unnecessary re-installs.
Default: false
Allowed values:
truefalse(↑ Back to package attributes)
A file containing any necessary answers to questions asked by the package. This is currently used on Solaris and Debian. The value will be validated according to system rules, but it should generally be a fully qualified path.
(↑ Back to package attributes)
A read-only parameter set by the package.
(↑ Back to package attributes)
Where to find the package file. This is only used by providers that don’t
automatically download packages from a central repository. (For example:
the yum and apt providers ignore this attribute, but the rpm and
dpkg providers require it.)
Different providers accept different values for source. Most providers
accept paths to local files stored on the target system. Some providers
may also accept URLs or network drive paths. Puppet will not
automatically retrieve source files for you, and usually just passes the
value of source to the package installation command.
You can use a file resource if you need to manually copy package files
to the target system.
(↑ Back to package attributes)
A read-only parameter set by the package.
(↑ Back to package attributes)
An array of additional options to pass when uninstalling a package. These options are package-specific, and should be documented by the software vendor. For example:
package { 'VMware Tools':
ensure => absent,
uninstall_options => [ { 'REMOVE' => 'Sync,VSS' } ],
}
Each option in the array can either be a string or a hash, where each key and value pair are interpreted in a provider specific way. Each option will automatically be quoted when passed to the uninstall command.
On Windows, this is the only place in Puppet where backslash separators should be used. Note that backslashes in double-quoted strings must be double-escaped and backslashes in single-quoted strings may be double-escaped.
Requires features uninstall_options.
(↑ Back to package attributes)
A read-only parameter set by the package.
(↑ Back to package attributes)
Installation from an AIX software directory, using the AIX installp
command. The source parameter is required for this provider, and should
be set to the absolute path (on the puppet agent machine) of a directory
containing one or more BFF package files.
The installp command will generate a table of contents file (named .toc)
in this directory, and the name parameter (or resource title) that you
specify for your package resource must match a package name that exists
in the .toc file.
Note that package downgrades are not supported; if your resource specifies a specific version number and there is already a newer version of the package installed on the machine, the resource will fail with an error message.
/usr/bin/lslpp, /usr/sbin/installpoperatingsystem == [ :aix ]operatingsystem == aixinstallable, uninstallable, upgradeable, versionablePackage management which copies application bundles to a target.
/usr/bin/hdiutil, /usr/bin/curl, /usr/bin/dittooperatingsystem == darwin, feature == cfpropertylistinstallablePackage management based on OS X’s built-in packaging system. This is
essentially the simplest and least functional package system in existence –
it only supports installation; no deletion or upgrades. The provider will
automatically add the .pkg extension, so leave that off when specifying
the package name.
/usr/sbin/installeroperatingsystem == darwininstallablePackage management via apt-get.
This provider supports the install_options attribute, which allows command-line flags to be passed to apt-get.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
/usr/bin/apt-get, /usr/bin/apt-cache, /usr/bin/debconf-set-selectionsosfamily == debianholdable, install_options, installable, purgeable, uninstallable, upgradeable, versionablePackage management via aptitude.
/usr/bin/aptitude, /usr/bin/apt-cacheholdable, installable, purgeable, uninstallable, upgradeable, versionablePackage management via apt-get ported to rpm.
apt-get, apt-cache, rpminstallable, purgeable, uninstallable, upgradeable, versionablePackage management using Blastwave.org’s pkg-get command on Solaris.
pkggetosfamily == solarisinstallable, uninstallable, upgradeableSupport via dnf.
Using this provider’s uninstallable feature will not remove dependent packages. To
remove dependent packages with this provider use the purgeable feature, but note this
feature is destructive and should be used with the utmost care.
This provider supports the install_options attribute, which allows command-line flags to be passed to dnf.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
dnf, rpmoperatingsystem == fedora and operatingsystemmajrelease == 22, 23, 24, 25, 26, 27, 28, 29, 30.install_options, installable, purgeable, uninstallable, upgradeable, versionable, virtual_packagesPackage management via dpkg. Because this only uses dpkg
and not apt, you must specify the source of any packages you want
to manage.
/usr/bin/dpkg, /usr/bin/dpkg-deb, /usr/bin/dpkg-queryholdable, installable, purgeable, uninstallable, upgradeablePackage management via fink.
/sw/bin/fink, /sw/bin/apt-get, /sw/bin/apt-cache, /sw/bin/dpkg-queryholdable, installable, purgeable, uninstallable, upgradeable, versionableThe specific form of package management on FreeBSD. This is an extremely quirky packaging system, in that it freely mixes between ports and packages. Apparently all of the tools are written in Ruby, so there are plans to rewrite this support to directly use those libraries.
/usr/sbin/pkg_info, /usr/sbin/pkg_add, /usr/sbin/pkg_deleteoperatingsystem == freebsdinstallable, purgeable, uninstallable, upgradeable.Ruby Gem support. If a URL is passed via source, then that URL is
appended to the list of remote gem repositories; to ensure that only the
specified source is used, also pass --clear-sources via install_options.
If source is present but is not a valid URL, it will be interpreted as the
path to a local gem file. If source is not present, the gem will be
installed from the default gem repositories. Note that to modify this for Windows, it has to be a valid URL.
This provider supports the install_options and uninstall_options attributes,
which allow command-line flags to be passed to the gem command.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
geminstall_options, installable, uninstall_options, uninstallable, upgradeable, versionableHP-UX’s packaging system.
/usr/sbin/swinstall, /usr/sbin/swlist, /usr/sbin/swremoveoperatingsystem == hp-ux["operatingsystem", "hp-ux"] == Package management using MacPorts on OS X.
Supports MacPorts versions and revisions, but not variants. Variant preferences may be specified using the MacPorts variants.conf file.
When specifying a version in the Puppet DSL, only specify the version, not the revision. Revisions are only used internally for ensuring the latest version/revision of a port.
/opt/local/bin/portoperatingsystem == darwininstallable, uninstallable, upgradeable, versionableInstallation from an AIX NIM LPP source. The source parameter is required
for this provider, and should specify the name of a NIM lpp_source resource
that is visible to the puppet agent machine. This provider supports the
management of both BFF/installp and RPM packages.
Note that package downgrades are not supported; if your resource specifies a specific version number and there is already a newer version of the package installed on the machine, the resource will fail with an error message.
/usr/sbin/nimclient, /usr/bin/lslpp, rpmexists == /etc/niminfoinstallable, uninstallable, upgradeable, versionableOpenBSD’s form of pkg_add support.
This provider supports the install_options and uninstall_options
attributes, which allow command-line flags to be passed to pkg_add and pkg_delete.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
pkg_info, pkg_add, pkg_deleteoperatingsystem == openbsd["operatingsystem", "openbsd"] == Opkg packaging support. Common on OpenWrt and OpenEmbedded platforms
opkgoperatingsystem == openwrt["operatingsystem", "openwrt"] == Support for the Package Manager Utility (pacman) used in Archlinux.
This provider supports the install_options attribute, which allows command-line flags to be passed to pacman.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
/usr/bin/pacmanoperatingsystem == [:archlinux, :manjarolinux]["operatingsystem", "[:archlinux, :manjarolinux]"] == install_options, installable, uninstall_options, uninstallable, upgradeable, virtual_packagesPython packages via pip.
This provider supports the install_options attribute, which allows command-line flags to be passed to pip.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
install_options, installable, uninstallable, upgradeable, versionablePython packages via pip3.
This provider supports the install_options attribute, which allows command-line flags to be passed to pip3.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
install_options, installable, uninstallable, upgradeable, versionableOpenSolaris image packaging system. See pkg(5) for more information.
/usr/bin/pkgosfamily == solaris["osfamily", "solaris"] == ["kernelrelease", "['5.11', '5.12']"]holdable, installable, uninstallable, upgradeable, versionablePackage management based on Apple’s Installer.app and DiskUtility.app.
This provider works by checking the contents of a DMG image for Apple pkg or mpkg files. Any number of pkg or mpkg files may exist in the root directory of the DMG file system, and Puppet will install all of them. Subdirectories are not checked for packages.
This provider can also accept plain .pkg (but not .mpkg) files in addition to .dmg files.
Notes:
source attribute is mandatory. It must be either a local disk path
or an HTTP, HTTPS, or FTP URL to the package.name of the resource must be the filename (without path) of the DMG file./var/db/.puppet_pkgdmg_installed_NAME. If that file is present,
Puppet assumes all packages from that DMG are already installed.This provider is not versionable and uses DMG filenames to determine whether a package has been installed. Thus, to install new a version of a package, you must create a new DMG with a different filename.
/usr/sbin/installer, /usr/bin/hdiutil, /usr/bin/curloperatingsystem == darwin, feature == cfpropertylist["operatingsystem", "darwin"] == Package management using pkgin, a binary package manager for pkgsrc.
pkgin["operatingsystem", "[ :smartos, :netbsd ]"] == A PkgNG provider for FreeBSD and DragonFly.
/usr/local/sbin/pkgoperatingsystem == [:freebsd, :dragonfly]["operatingsystem", "[:freebsd, :dragonfly]"] == Package management using Peter Bonivart’s pkgutil command on Solaris.
pkgutilosfamily == solarisinstallable, uninstallable, upgradeableProvides packaging support for Gentoo’s portage system.
This provider supports the install_options and uninstall_options attributes, which allows command-line
flags to be passed to emerge. These options should be specified as a string (e.g. ‘–flag’), a hash
(e.g. {‘–flag’ => ‘value’}), or an array where each element is either a string or a hash.
/usr/bin/eix-update, /usr/bin/eix, /usr/bin/emerge, /usr/bin/qatomoperatingsystem == gentoo["operatingsystem", "gentoo"] == Support for FreeBSD’s ports. Note that this, too, mixes packages and ports.
/usr/local/sbin/portupgrade, /usr/local/sbin/portversion, /usr/local/sbin/pkg_deinstall, /usr/sbin/pkg_infoinstallable, purgeable, uninstallable, upgradeableSupport for FreeBSD’s ports using the portupgrade ports management software. Use the port’s full origin as the resource name. eg (ports-mgmt/portupgrade) for the portupgrade port.
/usr/local/sbin/portupgrade, /usr/local/sbin/portinstall, /usr/local/sbin/portversion, /usr/local/sbin/pkg_deinstall, /usr/sbin/pkg_infoinstallable, uninstallable, upgradeablePuppet Ruby Gem support. This provider is useful for managing gems needed by the ruby provided in the puppet-agent package.
/opt/puppetlabs/puppet/bin/geminstall_options, installable, uninstall_options, uninstallable, upgradeable, versionableRPM packaging support; should work anywhere with a working rpm
binary.
This provider supports the install_options and uninstall_options
attributes, which allow command-line flags to be passed to rpm.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
rpminstall_options, installable, uninstall_options, uninstallable, upgradeable, versionable, virtual_packagesSupport for suse rug package manager.
/usr/bin/rug, rpmoperatingsystem == [:suse, :sles]installable, uninstallable, upgradeable, versionableSun’s packaging system. Requires that you specify the source for the packages you’re managing.
This provider supports the install_options attribute, which allows command-line flags to be passed to pkgadd.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
/usr/bin/pkginfo, /usr/sbin/pkgadd, /usr/sbin/pkgrmosfamily == solaris["osfamily", "solaris"] == Package management using sunfreeware.com’s pkg-get command on Solaris.
At this point, support is exactly the same as blastwave support and
has not actually been tested.
pkg-getosfamily == solarisinstallable, uninstallable, upgradeableSupport via tdnf.
This provider supports the install_options attribute, which allows command-line flags to be passed to tdnf.
These options should be spcified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}), or an
array where each element is either a string or a hash.
tdnf, rpm["operatingsystem", "PhotonOS"] == Support for Red Hat’s proprietary up2date package update
mechanism.
/usr/sbin/up2date-noxosfamily == redhat["osfamily", "redhat"] == ["lsbdistrelease", "[\"2.1\", \"3\", \"4\"]"]Support via urpmi.
urpmi, urpmq, rpm, urpme["operatingsystem", "[:mandriva, :mandrake]"] == Windows package management.
This provider supports either MSI or self-extracting executable installers.
This provider requires a source attribute when installing the package.
It accepts paths to local files, mapped drives, or UNC paths.
This provider supports the install_options and uninstall_options
attributes, which allow command-line flags to be passed to the installer.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
If the executable requires special arguments to perform a silent install or
uninstall, then the appropriate arguments should be specified using the
install_options or uninstall_options attributes, respectively. Puppet
will automatically quote any option that contains spaces.
operatingsystem == windows["operatingsystem", "windows"] == Support via yum.
Using this provider’s uninstallable feature will not remove dependent packages. To
remove dependent packages with this provider use the purgeable feature, but note this
feature is destructive and should be used with the utmost care.
This provider supports the install_options attribute, which allows command-line flags to be passed to yum.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
yum, rpm["osfamily", "redhat"] == Support for SuSE zypper package manager. Found in SLES10sp2+ and SLES11.
This provider supports the install_options attribute, which allows command-line flags to be passed to zypper.
These options should be specified as a string (e.g. ‘–flag’), a hash (e.g. {‘–flag’ => ‘value’}),
or an array where each element is either a string or a hash.
/usr/bin/zypperoperatingsystem == [:suse, :sles, :sled, :opensuse]["operatingsystem", "[:suse, :sles, :sled, :opensuse]"] == Available features:
holdable — The provider is capable of placing packages on hold such that they are not automatically upgraded as a result of other package dependencies unless explicit action is taken by a user or another package. Held is considered a superset of installed.install_options — The provider accepts options to be passed to the installer command.installable — The provider can install packages.package_settings — The provider accepts package_settings to be ensured for the given package. The meaning and format of these settings is provider-specific.purgeable — The provider can purge packages. This generally means that all traces of the package are removed, including existing configuration files. This feature is thus destructive and should be used with the utmost care.reinstallable — The provider can reinstall packages.uninstall_options — The provider accepts options to be passed to the uninstaller command.uninstallable — The provider can uninstall packages.upgradeable — The provider can upgrade to the latest version of a package. This feature is used by specifying latest as the desired value for the package.versionable — The provider is capable of interrogating the package database for installed version(s), and can select which out of a set of available versions of a package to install if asked.virtual_packages — The provider accepts virtual package names for install and uninstall.Provider support:
| Provider | holdable | install options | installable | package settings | purgeable | reinstallable | uninstall options | uninstallable | upgradeable | versionable | virtual packages | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| aix | X | X | X | |||||||||||
| appdmg | ||||||||||||||
| apple | ||||||||||||||
| apt | X | X | ||||||||||||
| aptitude | X | X | X | |||||||||||
| aptrpm | X | X | X | |||||||||||
| blastwave | X | |||||||||||||
| dnf | X | X | X | X | X | |||||||||
| dpkg | X | X | X | |||||||||||
| fink | X | X | X | X | ||||||||||
| freebsd | X | X | ||||||||||||
| gem | X | X | X | |||||||||||
| hpux | ||||||||||||||
| macports | X | X | X | X | ||||||||||
| nim | X | X | ||||||||||||
| openbsd | X | X | X | X | X | X | ||||||||
| opkg | X | |||||||||||||
| pacman | X | X | X | X | ||||||||||
| pip | X | X | X | X | X | |||||||||
| pip3 | X | X | X | X | X | |||||||||
| pkg | X | X | X | |||||||||||
| pkgdmg | ||||||||||||||
| pkgin | X | X | X | X | ||||||||||
| pkgng | X | X | ||||||||||||
| pkgutil | X | |||||||||||||
| portage | X | X | X | X | X | X | ||||||||
| ports | X | X | ||||||||||||
| portupgrade | X | |||||||||||||
| puppet_gem | X | X | X | |||||||||||
| rpm | X | X | X | X | ||||||||||
| rug | X | X | X | |||||||||||
| sun | X | X | ||||||||||||
| sunfreeware | X | |||||||||||||
| tdnf | X | X | X | X | X | |||||||||
| up2date | X | |||||||||||||
| urpmi | X | X | X | |||||||||||
| windows | X | X | X | X | X | |||||||||
| yum | X | X | X | X | X | |||||||||
| zypper | X | X | X | X |
This is a metatype that can manage other resource types. Any
metaparams specified here will be passed on to any generated resources,
so you can purge unmanaged resources but set noop to true so the
purging is only logged and does not actually happen.
resources { 'resource title':
name => # (namevar) The name of the type to be...
purge => # Whether to purge unmanaged resources. When set...
unless_system_user => # This keeps system users from being purged. By...
unless_uid => # This keeps specific uids or ranges of uids from...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The name of the type to be managed.
(↑ Back to resources attributes)
Whether to purge unmanaged resources. When set to true, this will
delete any resource that is not specified in your configuration and is not
autorequired by any managed resources. Note: The ssh_authorized_key
resource type can’t be purged this way; instead, see the purge_ssh_keys
attribute of the user type.
Default: false
Allowed values:
truefalseyesno(↑ Back to resources attributes)
This keeps system users from being purged. By default, it does not purge users whose UIDs are less than the minimum UID for the system (typically 500 or 1000), but you can specify a different UID as the inclusive limit.
Allowed values:
truefalse/^\d+$/(↑ Back to resources attributes)
This keeps specific uids or ranges of uids from being purged when purge is true. Accepts integers, integer strings, and arrays of integers or integer strings. To specify a range of uids, consider using the range() function from stdlib.
(↑ Back to resources attributes)
Currently, schedules can only be used to stop a resource from being applied; they cannot cause a resource to be applied when it otherwise wouldn’t be, and they cannot accurately specify a time when a resource should run.
Every time Puppet applies its configuration, it will apply the set of resources whose schedule does not eliminate them from running right then, but there is currently no system in place to guarantee that a given resource runs at a given time. If you specify a very restrictive schedule and Puppet happens to run at a time within that schedule, then the resources will get applied; otherwise, that work may never get done.
Thus, it is advisable to use wider scheduling (for example, over a couple of hours) combined with periods and repetitions. For instance, if you wanted to restrict certain resources to only running once, between the hours of two and 4 AM, then you would use this schedule:
schedule { 'maint':
range => '2 - 4',
period => daily,
repeat => 1,
}
With this schedule, the first time that Puppet runs between 2 and 4 AM, all resources with this schedule will get applied, but they won’t get applied again between 2 and 4 because they will have already run once that day, and they won’t get applied outside that schedule because they will be outside the scheduled range.
Puppet automatically creates a schedule for each of the valid periods
with the same name as that period (such as hourly and daily).
Additionally, a schedule named puppet is created and used as the
default, with the following attributes:
schedule { 'puppet':
period => hourly,
repeat => 2,
}
This will cause resources to be applied every 30 minutes by default.
schedule { 'resource title':
name => # (namevar) The name of the schedule. This name is used...
period => # The period of repetition for resources on this...
periodmatch => # Whether periods should be matched by a numeric...
range => # The earliest and latest that a resource can be...
repeat => # How often a given resource may be applied in...
weekday => # The days of the week in which the schedule...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The name of the schedule. This name is used when assigning the schedule
to a resource with the schedule metaparameter:
schedule { 'everyday':
period => daily,
range => '2 - 4',
}
exec { '/usr/bin/apt-get update':
schedule => 'everyday',
}
(↑ Back to schedule attributes)
The period of repetition for resources on this schedule. The default is for resources to get applied every time Puppet runs.
Note that the period defines how often a given resource will get
applied but not when; if you would like to restrict the hours
that a given resource can be applied (for instance, only at night
during a maintenance window), then use the range attribute.
If the provided periods are not sufficient, you can provide a value to the repeat attribute, which will cause Puppet to schedule the affected resources evenly in the period the specified number of times. Take this schedule:
schedule { 'veryoften':
period => hourly,
repeat => 6,
}
This can cause Puppet to apply that resource up to every 10 minutes.
At the moment, Puppet cannot guarantee that level of repetition; that is, the resource can applied up to every 10 minutes, but internal factors might prevent it from actually running that often (for instance, if a Puppet run is still in progress when the next run is scheduled to start, that next run will be suppressed).
See the periodmatch attribute for tuning whether to match
times by their distance apart or by their specific value.
Allowed values:
hourlydailyweeklymonthlyneverTip: You can use
period => never,to prevent a resource from being applied in the givenrange. This is useful if you need to create a blackout window to perform sensitive operations without interruption.
(↑ Back to schedule attributes)
Whether periods should be matched by a numeric value (for instance, whether two times are in the same hour) or by their chronological distance apart (whether two times are 60 minutes apart).
Default: distance
Allowed values:
numberdistance(↑ Back to schedule attributes)
The earliest and latest that a resource can be applied. This is always a hyphen-separated range within a 24 hour period, and hours must be specified in numbers between 0 and 23, inclusive. Minutes and seconds can optionally be provided, using the normal colon as a separator. For instance:
schedule { 'maintenance':
range => '1:30 - 4:30',
}
This is mostly useful for restricting certain resources to being
applied in maintenance windows or during off-peak hours. Multiple
ranges can be applied in array context. As a convenience when specifying
ranges, you can cross midnight (for example, range => "22:00 - 04:00").
(↑ Back to schedule attributes)
How often a given resource may be applied in this schedule’s period.
Must be an integer.
Default: 1
(↑ Back to schedule attributes)
The days of the week in which the schedule should be valid. You may specify the full day name ‘Tuesday’, the three character abbreviation ‘Tue’, or a number (as a string or as an integer) corresponding to the day of the week where 0 is Sunday, 1 is Monday, and so on. Multiple days can be specified as an array. If not specified, the day of the week will not be considered in the schedule.
If you are also using a range match that spans across midnight then this parameter will match the day that it was at the start of the range, not necessarily the day that it is when it matches. For example, consider this schedule:
schedule { 'maintenance_window':
range => '22:00 - 04:00',
weekday => 'Saturday',
}
This will match at 11 PM on Saturday and 2 AM on Sunday, but not at 2 AM on Saturday.
(↑ Back to schedule attributes)
Manage running services. Service support unfortunately varies widely by platform — some platforms have very little if any concept of a running service, and some have a very codified and powerful concept. Puppet’s service support is usually capable of doing the right thing, but the more information you can provide, the better behaviour you will get.
Puppet 2.7 and newer expect init scripts to have a working status command.
If this isn’t the case for any of your services’ init scripts, you will
need to set hasstatus to false and possibly specify a custom status
command in the status attribute. As a last resort, Puppet will attempt to
search the process table by calling whatever command is listed in the ps
fact. The default search pattern is the name of the service, but you can
specify it with the pattern attribute.
Refresh: service resources can respond to refresh events (via
notify, subscribe, or the ~> arrow). If a service receives an
event from another resource, Puppet will restart the service it manages.
The actual command used to restart the service depends on the platform and
can be configured:
hasrestart to true, Puppet will use the init script’s restart command.restart attribute.service { 'resource title':
name => # (namevar) The name of the service to run. This name is...
ensure => # Whether a service should be running. Allowed...
binary => # The path to the daemon. This is only used for...
control => # The control variable used to manage services...
enable => # Whether a service should be enabled to start at...
flags => # Specify a string of flags to pass to the startup
hasrestart => # Specify that an init script has a `restart...
hasstatus => # Declare whether the service's init script has a...
manifest => # Specify a command to config a service, or a path
path => # The search path for finding init scripts....
pattern => # The pattern to search for in the process table...
provider => # The specific backend to use for this `service...
restart => # Specify a *restart* command manually. If left...
start => # Specify a *start* command manually. Most...
status => # Specify a *status* command manually. This...
stop => # Specify a *stop* command...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The name of the service to run.
This name is used to find the service; on platforms where services have short system names and long display names, this should be the short name. (To take an example from Windows, you would use “wuauserv” rather than “Automatic Updates.”)
(↑ Back to service attributes)
(Property: This attribute represents concrete state on the target system.)
Whether a service should be running.
Allowed values:
stoppedrunningfalsetrue(↑ Back to service attributes)
The path to the daemon. This is only used for
systems that do not support init scripts. This binary will be
used to start the service if no start parameter is
provided.
(↑ Back to service attributes)
The control variable used to manage services (originally for HP-UX).
Defaults to the upcased service name plus START replacing dots with
underscores, for those providers that support the controllable feature.
(↑ Back to service attributes)
(Property: This attribute represents concrete state on the target system.)
Whether a service should be enabled to start at boot. This property behaves quite differently depending on the platform; wherever possible, it relies on local tools to enable or disable a given service.
Allowed values:
truefalsemanualmaskRequires features enableable.
(↑ Back to service attributes)
(Property: This attribute represents concrete state on the target system.)
Specify a string of flags to pass to the startup script.
Requires features flaggable.
(↑ Back to service attributes)
Specify that an init script has a restart command. If this is
false and you do not specify a command in the restart attribute,
the init script’s stop and start commands will be used.
Defaults to false.
Allowed values:
truefalse(↑ Back to service attributes)
Declare whether the service’s init script has a functional status command. This attribute’s default value changed in Puppet 2.7.0.
The init script’s status command must return 0 if the service is running and a nonzero value otherwise. Ideally, these exit codes should conform to the LSB’s specification for init script status actions, but Puppet only considers the difference between 0 and nonzero to be relevant.
If a service’s init script does not support any kind of status command,
you should set hasstatus to false and either provide a specific
command using the status attribute or expect that Puppet will look for
the service name in the process table. Be aware that ‘virtual’ init
scripts (like ‘network’ under Red Hat systems) will respond poorly to
refresh events from other resources if you override the default behavior
without providing a status command.
Default: true
Allowed values:
truefalse(↑ Back to service attributes)
Specify a command to config a service, or a path to a manifest to do so.
(↑ Back to service attributes)
The search path for finding init scripts. Multiple values should be separated by colons or provided as an array.
(↑ Back to service attributes)
The pattern to search for in the process table. This is used for stopping services on platforms that do not support init scripts, and is also used for determining service status on those service whose init scripts do not include a status command.
Defaults to the name of the service. The pattern can be a simple string or any legal Ruby pattern, including regular expressions (which should be quoted without enclosing slashes).
(↑ Back to service attributes)
The specific backend to use for this service
resource. You will seldom need to specify this — Puppet will usually
discover the appropriate provider for your platform.
Available providers are:
basebsddaemontoolsdebianfreebsdgentooinitlaunchdopenbsdopenrcopenwrtrcngredhatrunitservicesmfsrcsystemdupstartwindows(↑ Back to service attributes)
Specify a restart command manually. If left unspecified, the service will be stopped and then started.
(↑ Back to service attributes)
Specify a start command manually. Most service subsystems
support a start command, so this will not need to be
specified.
(↑ Back to service attributes)
Specify a status command manually. This command must return 0 if the service is running and a nonzero value otherwise. Ideally, these exit codes should conform to the LSB’s specification for init script status actions, but Puppet only considers the difference between 0 and nonzero to be relevant.
If left unspecified, the status of the service will be determined automatically, usually by looking for the service in the process table.
(↑ Back to service attributes)
Specify a stop command manually.
(↑ Back to service attributes)
The simplest form of Unix service support.
You have to specify enough about your service for this to work; the
minimum you can specify is a binary for starting the process, and this
same binary will be searched for in the process table to stop the
service. As with init-style services, it is preferable to specify start,
stop, and status commands.
killrefreshableGeneric BSD form of init-style service management with rc.d.
Uses rc.conf.d for service enabling and disabling.
operatingsystem == [:freebsd, :dragonfly]enableable, refreshable.Daemontools service management.
This provider manages daemons supervised by D.J. Bernstein daemontools. When detecting the service directory it will check, in order of preference:
/service/etc/service/var/lib/svscanThe daemon directory should be in one of the following locations:
/var/lib/service/etc…or this can be overridden in the resource’s attributes:
service { 'myservice':
provider => 'daemontools',
path => '/path/to/daemons',
}
This provider supports out of the box:
If a service has ensure => "running", it will link /path/to/daemon to
/path/to/service, which will automatically enable the service.
If a service has ensure => "stopped", it will only shut down the service, not
remove the /path/to/service link.
/usr/bin/svc, /usr/bin/svstatenableable, refreshableDebian’s form of init-style management.
The only differences from init are support for enabling and disabling
services via update-rc.d and the ability to determine enabled status via
invoke-rc.d.
/usr/sbin/update-rc.d, /usr/sbin/invoke-rc.d, /usr/sbin/service["operatingsystem", "cumuluslinux"] == ["operatingsystemmajrelease", "['1','2']"], ["operatingsystem", "debian"] == ["operatingsystemmajrelease", "['5','6','7']"], ["operatingsystem", "devuan"] == Provider for FreeBSD and DragonFly BSD. Uses the rcvar argument of init scripts and parses/edits rc files.
operatingsystem == [:freebsd, :dragonfly]["operatingsystem", "[:freebsd, :dragonfly]"] == Gentoo’s form of init-style service management.
Uses rc-update for service enabling and disabling.
/sbin/rc-updateoperatingsystem == gentooStandard init-style service management.
true == begin
os = Facter.value(:operatingsystem).downcase
family = Facter.value(:osfamily).downcase
!(os == 'debian' || os == 'ubuntu' || family == 'redhat')
endThis provider manages jobs with launchd, which is the default service
framework for Mac OS X (and may be available for use on other platforms).
For more information, see the launchd man page:
This provider reads plists out of the following directories:
/System/Library/LaunchDaemons/System/Library/LaunchAgents/Library/LaunchDaemons/Library/LaunchAgents…and builds up a list of services based upon each plist’s “Label” entry.
This provider supports:
Here is how the Puppet states correspond to launchd states:
Note that this allows you to do something launchctl can’t do, which is to
be in a state of “stopped/enabled” or “running/disabled”.
Note that this provider does not support overriding ‘restart’
/bin/launchctloperatingsystem == darwin, feature == cfpropertylist["operatingsystem", "darwin"] == Provider for OpenBSD’s rc.d daemon control scripts
/usr/sbin/rcctloperatingsystem == openbsd["operatingsystem", "openbsd"] == Support for Gentoo’s OpenRC initskripts
Uses rc-update, rc-status and rc-service to manage services.
/sbin/rc-service, /sbin/rc-update["operatingsystem", "gentoo"] == , ["operatingsystem", "funtoo"] == Support for OpenWrt flavored init scripts.
Uses /etc/init.d/service_name enable, disable, and enabled.
operatingsystem == openwrt["operatingsystem", "openwrt"] == RCng service management with rc.d
operatingsystem == [:netbsd, :cargos]["operatingsystem", "[:netbsd, :cargos]"] == Red Hat’s (and probably many others’) form of init-style service
management. Uses chkconfig for service enabling and disabling.
/sbin/chkconfig, /sbin/service["osfamily", "redhat"] == , ["osfamily", "suse"] == ["operatingsystemmajrelease", "[\"10\", \"11\"]"]Runit service management.
This provider manages daemons running supervised by Runit. When detecting the service directory it will check, in order of preference:
/service/etc/service/var/serviceThe daemon directory should be in one of the following locations:
/etc/sv/var/lib/serviceor this can be overridden in the service resource parameters:
service { 'myservice':
provider => 'runit',
path => '/path/to/daemons',
}
This provider supports out of the box:
status
/usr/bin/svThe simplest form of service support.
Support for Sun’s new Service Management Framework.
Starting a service is effectively equivalent to enabling it, so there is only support for starting and stopping services, which also enables and disables them, respectively.
By specifying manifest => "/path/to/service.xml", the SMF manifest will
be imported if it does not exist.
/usr/sbin/svcadm, /usr/bin/svcs, /usr/sbin/svccfgosfamily == solaris["osfamily", "solaris"] == Support for AIX’s System Resource controller.
Services are started/stopped based on the stopsrc and startsrc
commands, and some services can be refreshed with refresh command.
Enabling and disabling services is not supported, as it requires
modifications to /etc/inittab. Starting and stopping groups of subsystems
is not yet supported.
/usr/bin/lssrc, /usr/bin/refresh, /usr/bin/startsrc, /usr/bin/stopsrc, /usr/sbin/chitab, /usr/sbin/lsitab, /usr/sbin/mkitab, /usr/sbin/rmitaboperatingsystem == aix["operatingsystem", "aix"] == Manages systemd services using systemctl.
Because systemd defaults to assuming the .service unit type, the suffix
may be omitted. Other unit types (such as .path) may be managed by
providing the proper suffix.
systemctl["osfamily", "[:archlinux]"] == , ["osfamily", "redhat"] == ["operatingsystemmajrelease", "[\"7\", \"8\"]"], ["osfamily", "redhat"] == ["operatingsystem", "fedora"], ["osfamily", "suse"] == , ["osfamily", "coreos"] == , ["operatingsystem", "amazon"] == ["operatingsystemmajrelease", "[\"2\"]"], ["operatingsystem", "debian"] == ["operatingsystemmajrelease", "[\"8\", \"stretch/sid\", \"9\", \"buster/sid\"]"], ["operatingsystem", "ubuntu"] == , ["operatingsystem", "cumuluslinux"] == ["operatingsystemmajrelease", "[\"3\"]"]Ubuntu service management with upstart.
This provider manages upstart jobs on Ubuntu. For upstart documentation,
see http://upstart.ubuntu.com/.
/sbin/start, /sbin/stop, /sbin/restart, /sbin/status, /sbin/initctl, true == begin
initctl(‘version’, ‘–quiet’)
true
rescue
false
end`["operatingsystem", "ubuntu"] == ["operatingsystemmajrelease", "[\"10.04\", \"12.04\", \"14.04\", \"14.10\"]"]Support for Windows Service Control Manager (SCM). This provider can start, stop, enable, and disable services, and the SCM provides working status methods for all services.
Control of service groups (dependencies) is not yet supported, nor is running services as a specific user.
net.exeoperatingsystem == windows["operatingsystem", "windows"] == Available features:
controllable — The provider uses a control variable.enableable — The provider can enable and disable the serviceflaggable — The provider can pass flags to the service.maskable — The provider can ‘mask’ the service.refreshable — The provider can restart the service.Provider support:
| Provider | controllable | enableable | flaggable | maskable | refreshable |
|---|---|---|---|---|---|
| base | |||||
| bsd | |||||
| daemontools | |||||
| debian | |||||
| freebsd | |||||
| gentoo | X | X | |||
| init | X | ||||
| launchd | X | X | |||
| openbsd | X | X | X | ||
| openrc | X | X | |||
| openwrt | X | X | |||
| rcng | X | X | |||
| redhat | X | X | |||
| runit | X | X | |||
| service | X | ||||
| smf | X | X | |||
| src | X | X | |||
| systemd | X | X | X | ||
| upstart | X | X | |||
| windows | X | X |
A resource type for creating new run stages. Once a stage is available,
classes can be assigned to it by declaring them with the resource-like syntax
and using
the stage metaparameter.
Note that new stages are not useful unless you also declare their order
in relation to the default main stage.
A complete run stage example:
stage { 'pre':
before => Stage['main'],
}
class { 'apt-updates':
stage => 'pre',
}
Individual resources cannot be assigned to run stages; you can only set stages for classes.
stage { 'resource title':
name => # (namevar) The name of the stage. Use this as the value for
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The name of the stage. Use this as the value for the stage metaparameter
when assigning classes to this stage.
Remove unwanted files based on specific criteria. Multiple criteria are OR’d together, so a file that is too large but is not old enough will still get tidied.
If you don’t specify either age or size, then all files will
be removed.
This resource type works by generating a file resource for every file that should be deleted and then letting that resource perform the actual deletion.
tidy { 'resource title':
path => # (namevar) The path to the file or directory to manage....
age => # Tidy files whose age is equal to or greater than
backup => # Whether tidied files should be backed up. Any...
matches => # One or more (shell type) file glob patterns...
recurse => # If target is a directory, recursively descend...
rmdirs => # Tidy directories in addition to files; that is...
size => # Tidy files whose size is equal to or greater...
type => # Set the mechanism for determining age. Default:
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The path to the file or directory to manage. Must be fully qualified.
Tidy files whose age is equal to or greater than the specified time. You can choose seconds, minutes, hours, days, or weeks by specifying the first letter of any of those words (for example, ‘1w’ represents one week).
Specifying 0 will remove all files.
Whether tidied files should be backed up. Any values are passed
directly to the file resources used for actual file deletion, so consult
the file type’s backup documentation to determine valid values.
One or more (shell type) file glob patterns, which restrict the list of files to be tidied to those whose basenames match at least one of the patterns specified. Multiple patterns can be specified using an array.
Example:
tidy { '/tmp':
age => '1w',
recurse => 1,
matches => [ '[0-9]pub*.tmp', '*.temp', 'tmpfile?' ],
}
This removes files from /tmp if they are one week old or older,
are not in a subdirectory and match one of the shell globs given.
Note that the patterns are matched against the basename of each file – that is, your glob patterns should not have any ‘/’ characters in them, since you are only specifying against the last bit of the file.
Finally, note that you must now specify a non-zero/non-false value for recurse if matches is used, as matches only apply to files found by recursion (there’s no reason to use static patterns match against a statically determined path). Requiring explicit recursion clears up a common source of confusion.
If target is a directory, recursively descend into the directory looking for files to tidy.
Allowed values:
truefalseinf/^[0-9]+$/Tidy directories in addition to files; that is, remove directories whose age is older than the specified criteria. This will only remove empty directories, so all contained files must also be tidied before a directory gets removed.
Allowed values:
truefalseyesnoTidy files whose size is equal to or greater than the specified size. Unqualified values are in kilobytes, but b, k, m, g, and t can be appended to specify bytes, kilobytes, megabytes, gigabytes, and terabytes, respectively. Only the first character is significant, so the full word can also be used.
Set the mechanism for determining age.
Default: atime
Allowed values:
atimemtimectimeManage users. This type is mostly built to manage system users, so it is lacking some features useful for managing normal users.
This resource type uses the prescribed native tools for creating
groups and generally uses POSIX APIs for retrieving information
about them. It does not directly modify /etc/passwd or anything.
Autorequires: If Puppet is managing the user’s primary group (as
provided in the gid attribute) or any group listed in the groups
attribute then the user resource will autorequire that group. If Puppet
is managing any role accounts corresponding to the user’s roles, the
user resource will autorequire those role accounts.
user { 'resource title':
name => # (namevar) The user name. While naming limitations vary by...
ensure => # The basic state that the object should be in....
allowdupe => # Whether to allow duplicate UIDs. Defaults to...
attribute_membership => # Whether specified attribute value pairs should...
attributes => # Specify AIX attributes for the user in an array...
auth_membership => # Whether specified auths should be considered the
auths => # The auths the user has. Multiple auths should...
comment => # A description of the user. Generally the user's
expiry => # The expiry date for this user. Provide as either
forcelocal => # Forces the management of local accounts when...
gid => # The user's primary group. Can be specified...
groups => # The groups to which the user belongs. The...
home => # The home directory of the user. The directory...
ia_load_module => # The name of the I&A module to use to manage this
iterations => # This is the number of iterations of a chained...
key_membership => # Whether specified key/value pairs should be...
keys => # Specify user attributes in an array of key ...
loginclass => # The name of login class to which the user...
managehome => # Whether to manage the home directory when Puppet
membership => # If `minimum` is specified, Puppet will ensure...
password => # The user's password, in whatever encrypted...
password_max_age => # The maximum number of days a password may be...
password_min_age => # The minimum number of days a password must be...
password_warn_days => # The number of days before a password is going to
profile_membership => # Whether specified roles should be treated as the
profiles => # The profiles the user has. Multiple profiles...
project => # The name of the project associated with a user.
provider => # The specific backend to use for this `user...
purge_ssh_keys => # Whether to purge authorized SSH keys for this...
role_membership => # Whether specified roles should be considered the
roles => # The roles the user has. Multiple roles should...
salt => # This is the 32-byte salt used to generate the...
shell => # The user's login shell. The shell must exist...
system => # Whether the user is a system user, according to...
uid => # The user ID; must be specified numerically. If...
# ...plus any applicable metaparameters.
}(Namevar: If omitted, this attribute’s value defaults to the resource’s title.)
The user name. While naming limitations vary by operating system, it is advisable to restrict names to the lowest common denominator, which is a maximum of 8 characters beginning with a letter.
Note that Puppet considers user names to be case-sensitive, regardless of the platform’s own rules; be sure to always use the same case when referring to a given user.
(Property: This attribute represents concrete state on the target system.)
The basic state that the object should be in.
Allowed values:
presentabsentroleWhether to allow duplicate UIDs.
Default: false
Allowed values:
truefalseyesnoWhether specified attribute value pairs should be treated as the
complete list (inclusive) or the minimum list (minimum) of
attribute/value pairs for the user.
Default: minimum
Allowed values:
inclusiveminimum(Property: This attribute represents concrete state on the target system.)
Specify AIX attributes for the user in an array of attribute = value pairs.
Requires features manages_aix_lam.
Whether specified auths should be considered the complete list
(inclusive) or the minimum list (minimum) of auths the user
has.
Default: minimum
Allowed values:
inclusiveminimum(Property: This attribute represents concrete state on the target system.)
The auths the user has. Multiple auths should be specified as an array.
Requires features manages_solaris_rbac.
(Property: This attribute represents concrete state on the target system.)
A description of the user. Generally the user’s full name.
(Property: This attribute represents concrete state on the target system.)
The expiry date for this user. Provide as either the special
value absent to ensure that the account never expires, or as
a zero-padded YYYY-MM-DD format – for example, 2010-02-19.
Allowed values:
absent/^\d{4}-\d{2}-\d{2}$/Requires features manages_expiry.
Forces the management of local accounts when accounts are also being managed by some other NSS
Default: false
Allowed values:
truefalseyesnoRequires features libuser.
(Property: This attribute represents concrete state on the target system.)
The user’s primary group. Can be specified numerically or by name.
This attribute is not supported on Windows systems; use the groups
attribute instead. (On Windows, designating a primary group is only
meaningful for domain accounts, which Puppet does not currently manage.)
(Property: This attribute represents concrete state on the target system.)
The groups to which the user belongs. The primary group should not be listed, and groups should be identified by name rather than by GID. Multiple groups should be specified as an array.
(Property: This attribute represents concrete state on the target system.)
The home directory of the user. The directory must be created separately and is not currently checked for existence.
The name of the I&A module to use to manage this user.
Requires features manages_aix_lam.
(Property: This attribute represents concrete state on the target system.)
This is the number of iterations of a chained computation of the PBKDF2 password hash. This parameter is used in OS X, and is required for managing passwords on OS X 10.8 and newer.
Requires features manages_password_salt.
Whether specified key/value pairs should be considered the
complete list (inclusive) or the minimum list (minimum) of
the user’s attributes.
Default: minimum
Allowed values:
inclusiveminimum(Property: This attribute represents concrete state on the target system.)
Specify user attributes in an array of key = value pairs.
Requires features manages_solaris_rbac.
(Property: This attribute represents concrete state on the target system.)
The name of login class to which the user belongs.
Requires features manages_loginclass.
Whether to manage the home directory when Puppet creates or removes the user. This creates the home directory if Puppet also creates the user account, and deletes the home directory if Puppet also removes the user account.
This parameter has no effect unless Puppet is also creating or removing the user in the
resource at the same time. For instance, Puppet creates a home directory for a managed
user if ensure => present and the user does not exist at the time of the Puppet run.
If the home directory is then deleted manually, Puppet will not recreate it on the next
run.
Default: false
Allowed values:
truefalseyesnoIf minimum is specified, Puppet will ensure that the user is a
member of all specified groups, but will not remove any other groups
that the user is a part of.
If inclusive is specified, Puppet will ensure that the user is a
member of only specified groups.
Default: minimum
Allowed values:
inclusiveminimum(Property: This attribute represents concrete state on the target system.)
The user’s password, in whatever encrypted format the local system requires. Consult your operating system’s documentation for acceptable password encryption formats and requirements.
sha1 function to generate a salted SHA1
hash from a password.str2saltedsha512 function
which can generate password hashes for these operating systems.salt and iterations attributes need to be specified as
well as the password.Enclose any value that includes a dollar sign ($) in single quotes (‘) to avoid accidental variable interpolation.
To redact passwords from reports to PuppetDB, use the Sensitive data type. For
example, this resource protects the password:
user { 'foo':
ensure => present,
password => Sensitive("my secret password")
}
This results in the password being redacted from the report, as in the
previous_value, desired_value, and message fields below.
events:
- !ruby/object:Puppet::Transaction::Event
audited: false
property: password
previous_value: "[redacted]"
desired_value: "[redacted]"
historical_value:
message: changed [redacted] to [redacted]
name: :password_changed
status: success
time: 2017-05-17 16:06:02.934398293 -07:00
redacted: true
corrective_change: false
corrective_change: false
Requires features manages_passwords.
(Property: This attribute represents concrete state on the target system.)
The maximum number of days a password may be used before it must be changed.
Requires features manages_password_age.
(Property: This attribute represents concrete state on the target system.)
The minimum number of days a password must be used before it may be changed.
Requires features manages_password_age.
(Property: This attribute represents concrete state on the target system.)
The number of days before a password is going to expire (see the maximum password age) during which the user should be warned.
Requires features manages_password_age.
Whether specified roles should be treated as the complete list
(inclusive) or the minimum list (minimum) of roles
of which the user is a member.
Default: minimum
Allowed values:
inclusiveminimum(Property: This attribute represents concrete state on the target system.)
The profiles the user has. Multiple profiles should be specified as an array.
Requires features manages_solaris_rbac.
(Property: This attribute represents concrete state on the target system.)
The name of the project associated with a user.
Requires features manages_solaris_rbac.
The specific backend to use for this user
resource. You will seldom need to specify this — Puppet will usually
discover the appropriate provider for your platform.
Available providers are:
Whether to purge authorized SSH keys for this user if they are not managed
with the ssh_authorized_key resource type. This parameter is a noop if the
ssh_authorized_key type is not available.
Allowed values are:
false (default) — don’t purge SSH keys for this user.true — look for keys in the .ssh/authorized_keys file in the user’s
home directory. Purge any keys that aren’t managed as ssh_authorized_key
resources.ssh_authorized_key resources. If any of
these paths starts with ~ or %h, that token will be replaced with
the user’s home directory.Default: false
Allowed values:
truefalseWhether specified roles should be considered the complete list
(inclusive) or the minimum list (minimum) of roles the user
has.
Default: minimum
Allowed values:
inclusiveminimum(Property: This attribute represents concrete state on the target system.)
The roles the user has. Multiple roles should be specified as an array.
Requires features manages_solaris_rbac.
(Property: This attribute represents concrete state on the target system.)
This is the 32-byte salt used to generate the PBKDF2 password used in OS X. This field is required for managing passwords on OS X >= 10.8.
Requires features manages_password_salt.
(Property: This attribute represents concrete state on the target system.)
The user’s login shell. The shell must exist and be executable.
This attribute cannot be managed on Windows systems.
Requires features manages_shell.
Whether the user is a system user, according to the OS’s criteria; on most platforms, a UID less than or equal to 500 indicates a system user. This parameter is only used when the resource is created and will not affect the UID when the user is present.
Default: false
Allowed values:
truefalseyesno(Property: This attribute represents concrete state on the target system.)
The user ID; must be specified numerically. If no user ID is specified when creating a new user, then one will be chosen automatically. This will likely result in the same user having different UIDs on different systems, which is not recommended. This is especially noteworthy when managing the same user on both Darwin and other platforms, since Puppet does UID generation on Darwin, but the underlying tools do so on other platforms.
On Windows, this property is read-only and will return the user’s security identifier (SID).
User management for AIX.
/usr/sbin/lsuser, /usr/bin/mkuser, /usr/sbin/rmuser, /usr/bin/chuser, /usr/sbin/lsgroup, /bin/chpasswdoperatingsystem == aixoperatingsystem == aix.manages_aix_lam, manages_expiry, manages_homedir, manages_password_age, manages_passwords, manages_shell.User management on OS X.
/usr/bin/uuidgen, /usr/bin/dsimport, /usr/bin/dscl, /usr/bin/dscacheutiloperatingsystem == darwin, feature == cfpropertylistoperatingsystem == darwin.manages_password_salt, manages_passwords, manages_shell.User management for HP-UX. This provider uses the undocumented -F
switch to HP-UX’s special usermod binary to work around the fact that
its standard usermod cannot make changes while the user is logged in.
New functionality provides for changing trusted computing passwords and
resetting password expirations under trusted computing.
/usr/sam/lbin/usermod.sam, /usr/sam/lbin/userdel.sam, /usr/sam/lbin/useradd.samoperatingsystem == hp-uxoperatingsystem == hp-ux.allows_duplicates, manages_homedir, manages_passwords.User management via LDAP.
This provider requires that you have valid values for all of the
LDAP-related settings in puppet.conf, including ldapbase. You will
almost definitely need settings for ldapuser and ldappassword in order
for your clients to write to LDAP.
Note that this provider will automatically generate a UID for you if you do not specify one, but it is a potentially expensive operation, as it iterates across all existing users to pick the appropriate next one.
feature == ldap, false == (Puppet[:ldapuser] == "")manages_passwords, manages_shell.User management via useradd and its ilk for OpenBSD. Note that you
will need to install Ruby’s shadow password library (package known as
ruby-shadow) if you wish to manage user passwords.
useradd, userdel, usermod, passwdoperatingsystem == openbsdoperatingsystem == openbsd.manages_expiry, manages_homedir, manages_shell, system_users.User management via pw on FreeBSD and DragonFly BSD.
pwoperatingsystem == [:freebsd, :dragonfly]operatingsystem == freebsd, dragonfly.allows_duplicates, manages_expiry, manages_homedir, manages_passwords, manages_shell.User and role management on Solaris, via useradd and roleadd.
useradd, userdel, usermod, passwd, roleadd, roledel, rolemodosfamily == solaris.allows_duplicates, manages_homedir, manages_password_age, manages_passwords, manages_shell, manages_solaris_rbac.User management via useradd and its ilk. Note that you will need to
install Ruby’s shadow password library (often known as ruby-libshadow)
if you wish to manage user passwords.
useradd, userdel, usermod, chageallows_duplicates, manages_expiry, manages_homedir, manages_shell, system_users.Local user management for Windows.
operatingsystem == windowsoperatingsystem == windows.manages_homedir, manages_passwords.Available features:
allows_duplicates — The provider supports duplicate users with the same UID.libuser — Allows local users to be managed on systems that also use some other remote NSS method of managing accounts.manages_aix_lam — The provider can manage AIX Loadable Authentication Module (LAM) system.manages_expiry — The provider can manage the expiry date for a user.manages_homedir — The provider can create and remove home directories.manages_loginclass — The provider can manage the login class for a user.manages_password_age — The provider can set age requirements and restrictions for passwords.manages_password_salt — The provider can set a password salt. This is for providers that implement PBKDF2 passwords with salt properties.manages_passwords — The provider can modify user passwords, by accepting a password hash.manages_shell — The provider allows for setting shell and validates if possiblemanages_solaris_rbac — The provider can manage roles and normal userssystem_users — The provider allows you to create system users with lower UIDs.Provider support:
| Provider | allows duplicates | libuser | manages aix lam | manages expiry | manages homedir | manages loginclass | manages password age | manages password salt | manages passwords | manages shell | manages solaris rbac | system users |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| aix | X | X | X | X | X | X | ||||||
| directoryservice | X | X | X | |||||||||
| hpuxuseradd | X | X | X | |||||||||
| ldap | X | X | ||||||||||
| openbsd | X | X | X | X | X | X | ||||||
| pw | X | X | X | X | X | |||||||
| user_role_add | X | X | X | X | X | X | ||||||
| useradd | X | X | X | X | X | X | X | X | ||||
| windows_adsi | X | X |