heat-translator/doc/source/usage.rst
Nguyen Hai 060361fb11 Switch from oslosphinx to openstackdocstheme
- Change to openstackdocstheme which is a theme and extension
support for Sphinx documentation that is published to
docs.openstack.org and developer.openstack.org.
- This patch also updates README.rst

Change-Id: I5922eb9b0f5835be65a5ba3eead78bdf92458452
2018-05-29 19:43:49 +09:00

8.3 KiB

Usage

Use Heat-Translator with OpenStackClient (OSC)

Assuming that OpenStackClient (OSC) is available in your environment, you can easily install Heat-Translator to use with OSC by following three steps:

git clone https://github.com/openstack/heat-translator
cd heat-translator
sudo python setup.py install

Alternatively, you can install a particular release of Heat-Translator as available at https://pypi.org/project/heat-translator.

Once installation is complete, Heat-Translator is ready to use. The only required argument is --template-file. By default, the --template-type is set to tosca which is the only supported template type at present. Currently you can use Heat-Translator in following three ways.

Translate and get output on command line. For example: :

openstack translate template --template-file /home/openstack/heat-translator/translator/tests/data/tosca_helloworld.yaml --template-type tosca

Translate and save output of translated file to a desired destination. For example: :

openstack translate template --template-file /home/openstack/heat-translator/translator/tests/data/tosca_helloworld.yaml --template-type tosca --output-file /tmp/hot_hello_world.yaml

Do not translate but only validate template file. For example: :

openstack translate template --template-file /home/openstack/heat-translator/translator/tests/data/tosca_helloworld.yaml --template-type tosca --validate-only=true

You can learn more about available options by running following help command:

openstack help translate template

Use Heat-Translator on its own

Heat-Translator can be used without any specific OpenStack environment set up as below:

git clone https://github.com/openstack/heat-translator
python heat_translator.py --template-file==<path to the YAML template> --template-type=<type of template e.g. tosca> --parameters="purpose=test"

The heat_translator.py test program is at the root level of the project. The program has currently tested with TOSCA templates. The only required argument is --template-file. By default, the --template-type is set to tosca which is the only supported template type at present. The value to the --template-file is a path to the file that needs to be translated. The file, flat YAML template or CSAR, can be specified as a local file in your system or via URL.

For example, a TOSCA hello world template can be translated by running the following command from the project location:

python heat_translator.py --template-file=translator/tests/data/tosca_helloworld.yaml

This should produce a translated Heat Orchestration Template on the command line. The translated content can be saved to a desired file by setting --output-file=<path>. For example: :

python heat_translator.py --template-file=translator/tests/data/tosca_helloworld.yaml --template-type=tosca --output-file=/tmp/hot_helloworld.yaml

An optional argument can be provided to handle user inputs parameters. Also, a template file can only be validated instead of translation by using --validate-only=true optional argument. The command below shows an example usage:

python heat_translator.py --template-file=<path to the YAML template> --template-type=<type of template e.g. tosca> --validate-only=true

Alternatively, you can install a particular release of Heat-Translator as available at https://pypi.org/project/heat-translator. In this case, you can simply run translation via CLI entry point:

heat-translator --template-file=translator/tests/data/tosca_helloworld.yaml --template-type=tosca

Things To Consider

  • When use Heat-Translator in an OpenStack environment, please ensure that you have one or more preferred flavors and images available in your OpenStack environment. To find an appropriate flavor and image, that meets constraints defined in the TOSCA template for the HOST and OS capabilities of TOSCA Compute node, the Heat-Translator project first runs a query against Nova flavors and Glance images. During the query call, it uses the metadata of flavors and images. If call to Nova or Glance can not be made or no flavor or image is found, the Heat-Translator project will set flavor and image from a pre-defined set of values (as listed in /home/openstack/heat-translator/translator/hot/tosca/tosca_compute.py) with the best possible match to the constraints defined in the TOSCA template.

  • The key_name property of Nova server is irrelevant to the TOSCA specification and can not be used in TOSCA template. In order to use it in the translated templates, the user must provide it via parameters, and the heat-translator will set it to all resources of OS::Nova::Server type.

  • Since properties of TOSCA Compute OS and HOST capabilities are optional, the user should make sure that either they set these properties correctly in the TOSCA template or provide them via CLI parameters in order to find best match of flavor and image.

  • The flavor and image properties of OS::Nova::Server resource is irrelevant to the TOSCA specification and can not be used in the TOSCA template as such. Heat-Translator sets these properties in the translated template based on constraints defined per TOSCA Compute OS and HOST capabilities. However, user may required to use these properties in template in certain circumstances, so in that case, TOSCA Compute can be extended with these properties and later used in the node template. For a good example, refer to the translator/tests/data/test_tosca_flavor_and_image.yaml test template.

  • The Heat-Translator can be used to automatically deploy translated TOSCA template given that your environment has python-heatclient and python-keystoneclient. This can be achieved by providing --deploy argument to the Heat-Translator. You can provide desired stack name by providing it as --stack-name <name> argument. If you do not provide --stack-name, an unique name will be created and used. Below is an example command to deploy translated template with a desired stack name:

    heat-translator --template-file translator/tests/data/tosca_helloworld.yaml --stack-name mystack --deploy
  • The Heat-Translator supports translation of TOSCA templates to Heat Senlin resources (e.g. OS::Senlin::Cluster) but that requires to use a specific TOSCA node type called tosca.policies.Scaling.Cluster. The tosca.policies.Scaling.Cluster is a custom type that derives from tosca.policies.Scaling. For example usage, refer to the tosca_cluster_autoscaling.yaml and hot_cluster_autoscaling.yaml provided under the translator/tests/data/autoscaling and translator/tests/data/hot_output/autoscaling directories respectively in the heat-translator project (https://github.com/openstack/heat-translator). When you use tosca.policies.Scaling normative node type, the Heat-Translator will translate it to OS::Heat::AutoScalingGroup Heat resource. Related example templates, tosca_autoscaling.yaml and hot_autoscaling.yaml can be found for reference purposes under the same directory structure mentioned above.

  • With the version 0.7.0 of Heat-Translator, output of multiple template files (for example, nested templates in autoscaling) can be accessed via newly introduced API called translate_to_yaml_files_dict(<output_filename>) where <output_filename> is the name of file where you want to store parent HOT template. The return value of this API call will be a dictionary in HOT YAML with one or multiple file names as keys and translated content as values. In order to use this on the command line, simply invoke Heat-Translator with --output-file argument. Here, the parent template will be stored in the value specified to the --output-file. Whereas, child templates, if any, will be saved at the same location of the parent template.

    Below is an example of how to call the API in your code, where translator is an instance of Heat-Translator:

    yaml_files = translator.translate_to_yaml_files_dict(filename)

    Below is an example of how to use this on the command line:

    heat-translator --template-file translator/tests/data/autoscaling/tosca_autoscaling.yaml --output-file /tmp/hot.yaml