cdist/docs/src/cdist-type.rst

328 lines
9.8 KiB
ReStructuredText

cdist type
==========
Description
-----------
Types are the main component of cdist and define functionality. If you
use cdist, you'll write a type for every functionality you would like
to use.
Synopsis
--------
.. code-block:: sh
__TYPE ID --parameter value [--parameter value ...]
__TYPE --parameter value [--parameter value ...] (for singletons)
How to use a type
-----------------
You can use types from the initial manifest or the type manifest like a
normal shell command:
.. code-block:: sh
# Creates empty file /etc/cdist-configured
__file /etc/cdist-configured --type file
# Ensure tree is installed
__package tree --state installed
A list of supported types can be found in the `cdist reference <cdist-reference.html>`_ manpage.
Singleton types
---------------
If a type is flagged as a singleton, it may be used only
once per host. This is useful for types which can be used only once on a
system. Singleton types do not take an object name as argument.
Example:
.. code-block:: sh
# __issue type manages /etc/issue
__issue
# Probably your own type - singletons may use parameters
__myfancysingleton --colour green
Config types
------------
By default types are used with config command. These are types that are not
flagged by any known command flag. If a type is marked then it will be skipped
with config command.
Install types
-------------
If a type is flagged with 'install' flag then it is used only with install command.
With other commands, i.e. config, these types are skipped if used.
How to write a new type
-----------------------
A type consists of
- parameter (optional)
- manifest (optional)
- singleton (optional)
- explorer (optional)
- gencode (optional)
Types are stored below cdist/conf/type/. Their name should always be prefixed with
two underscores (__) to prevent collisions with other executables in $PATH.
To implement a new type, create the directory **cdist/conf/type/__NAME**.
Defining parameters
-------------------
Every type consists of required, optional and boolean parameters, which must
each be declared in a newline separated file in **parameter/required**,
**parameter/required_multiple**, **parameter/optional**,
**parameter/optional_multiple** and **parameter/boolean**.
Parameters which are allowed multiple times should be listed in
required_multiple or optional_multiple respectively. All other parameters
follow the standard unix behaviour "the last given wins".
If either is missing, the type will have no required, no optional, no boolean
or no parameters at all.
Default values for optional parameters can be predefined in
**parameter/default/<name>**.
Example:
.. code-block:: sh
echo servername >> cdist/conf/type/__nginx_vhost/parameter/required
echo logdirectory >> cdist/conf/type/__nginx_vhost/parameter/optional
echo loglevel >> cdist/conf/type/__nginx_vhost/parameter/optional
mkdir cdist/conf/type/__nginx_vhost/parameter/default
echo warning > cdist/conf/type/__nginx_vhost/parameter/default/loglevel
echo server_alias >> cdist/conf/type/__nginx_vhost/parameter/optional_multiple
echo use_ssl >> cdist/conf/type/__nginx_vhost/parameter/boolean
Using parameters
----------------
The parameters given to a type can be accessed and used in all type scripts
(e.g manifest, gencode, explorer). Note that boolean parameters are
represented by file existence. File exists -> True,
file does not exist -> False
Example: (e.g. in cdist/conf/type/__nginx_vhost/manifest)
.. code-block:: sh
# required parameter
servername="$(cat "$__object/parameter/servername")"
# optional parameter
if [ -f "$__object/parameter/logdirectory" ]; then
logdirectory="$(cat "$__object/parameter/logdirectory")"
fi
# optional parameter with predefined default
loglevel="$(cat "$__object/parameter/loglevel")"
# boolean parameter
if [ -f "$__object/parameter/use_ssl" ]; then
# file exists -> True
# do some fancy ssl stuff
fi
# parameter with multiple values
if [ -f "$__object/parameter/server_alias" ]; then
for alias in $(cat "$__object/parameter/server_alias"); do
echo $alias > /some/where/useful
done
fi
Input from stdin
----------------
Every type can access what has been written on stdin when it has been called.
The result is saved into the **stdin** file in the object directory.
Example use of a type: (e.g. in cdist/conf/type/__archlinux_hostname)
.. code-block:: sh
__file /etc/rc.conf --source - << eof
...
HOSTNAME="$__target_host"
...
eof
If you have not seen this syntax (<< eof) before, it may help you to read
about "here documents".
In the __file type, stdin is used as source for the file, if - is used for source:
.. code-block:: sh
if [ -f "$__object/parameter/source" ]; then
source="$(cat "$__object/parameter/source")"
if [ "$source" = "-" ]; then
source="$__object/stdin"
fi
....
Writing the manifest
--------------------
In the manifest of a type you can use other types, so your type extends
their functionality. A good example is the __package type, which in
a shortened version looks like this:
.. code-block:: sh
os="$(cat "$__global/explorer/os")"
case "$os" in
archlinux) type="pacman" ;;
debian|ubuntu) type="apt" ;;
gentoo) type="emerge" ;;
*)
echo "Don't know how to manage packages on: $os" >&2
exit 1
;;
esac
__package_$type "$@"
As you can see, the type can reference different environment variables,
which are documented in `cdist reference <cdist-reference.html>`_.
Always ensure the manifest is executable, otherwise cdist will not be able
to execute it. For more information about manifests see `cdist manifest <cdist-manifest.html>`_.
Singleton - one instance only
-----------------------------
If you want to ensure that a type can only be used once per target, you can
mark it as a singleton: Just create the (empty) file "singleton" in your type
directory:
.. code-block:: sh
touch cdist/conf/type/__NAME/singleton
This will also change the way your type must be called:
.. code-block:: sh
__YOURTYPE --parameter value
As you can see, the object ID is omitted, because it does not make any sense,
if your type can be used only once.
Install - type with install command
-----------------------------------
If you want a type to be used with install command, you must mark it as
install: create the (empty) file "install" in your type directory:
.. code-block:: sh
touch cdist/conf/type/__install_NAME/install
With other commands, i.e. config, it will be skipped if used.
The type explorers
------------------
If a type needs to explore specific details, it can provide type specific
explorers, which will be executed on the target for every created object.
The explorers are stored under the "explorer" directory below the type.
It could for instance contain code to check the md5sum of a file on the
client, like this (shortened version from the type __file):
.. code-block:: sh
if [ -f "$__object/parameter/destination" ]; then
destination="$(cat "$__object/parameter/destination")"
else
destination="/$__object_id"
fi
if [ -e "$destination" ]; then
md5sum < "$destination"
fi
Writing the gencode script
--------------------------
There are two gencode scripts: **gencode-local** and **gencode-remote**.
The output of gencode-local is executed locally, whereas
the output of gencode-remote is executed on the target.
The gencode scripts can make use of the parameters, the global explorers
and the type specific explorers.
If the gencode scripts encounters an error, it should print diagnostic
messages to stderr and exit non-zero. If you need to debug the gencode
script, you can write to stderr:
.. code-block:: sh
# Debug output to stderr
echo "My fancy debug line" >&2
# Output to be saved by cdist for execution on the target
echo "touch /etc/cdist-configured"
Notice: if you use __remote_copy or __remote_exec directly in your scripts
then for IPv6 address with __remote_copy execution you should enclose IPv6
address in square brackets. The same applies to __remote_exec if it behaves
the same as ssh for some options where colon is a delimiter, as for -L ssh
option (see :strong:`ssh`\ (1) and :strong:`scp`\ (1)).
Variable access from the generated scripts
------------------------------------------
In the generated scripts, you have access to the following cdist variables
- __object
- __object_id
but only for read operations, means there is no back copy of this
files after the script execution.
So when you generate a script with the following content, it will work:
.. code-block:: sh
if [ -f "$__object/parameter/name" ]; then
name="$(cat "$__object/parameter/name")"
else
name="$__object_id"
fi
Hints for typewriters
----------------------
It must be assumed that the target is pretty dumb and thus does not have high
level tools like ruby installed. If a type requires specific tools to be present
on the target, there must be another type that provides this tool and the first
type should create an object of the specific type.
If your type wants to save temporary data, that may be used by other types
later on (for instance \__file), you can save them in the subdirectory
"files" below $__object (but you must create it yourself).
cdist will not touch this directory.
If your type contains static files, it's also recommended to place them in
a folder named "files" within the type (again, because cdist guarantees to
never ever touch this folder).
How to include a type into upstream cdist
-----------------------------------------
If you think your type may be useful for others, ensure it works with the
current master branch of cdist and have a look at `cdist hacking <cdist-hacker.html>`_ on
how to submit it.