Workflow and Execution Parameters

Workflows can have parameters. Workflow parameters are declared in the blueprint, and each parameter can be declared as either mandatory or optional with a default value. To learn more about parameter declaration please refer to Creating your own workflow.

Viewing a workflow's parameters can be done in the CLI using the following command:

cfy workflows get my_workflow -d my_deployment

This command shows information on the my_workflow workflow of the my_deployment deployment, including the workflow’s mandatory parameters as well as the optional parameters and their default values.

Example: Retrieving a workflow’s parameters

$ cfy workflows get -d my_deployment my_workflow
Getting workflow 'my_workflow' of deployment 'my_deployment' [manager=]

| blueprint_id | deployment_id |     name    | created_at |
| my_blueprint | my_deployment | my_workflow |    None    |

Workflow Parameters:
    Mandatory Parameters:
        mandatory_parameter (this parameter is mandatory)
    Optional Parameters:
        optional_parameter:     optional_parameter_default_value        (this parameter is optional)
        nested_parameter:       {'key2': 'value2', 'key1': 'value1'}    (this parameter is also optional)
The workflow has a single mandatory parameter named mandatory_parameter*, and two optional parameters, one named* optional_parameter *which has a default value of* optional_parameter_default_value*, and another named* nested_parameter *which has a complex default value.*

When executing a workflow, it's required to specify values for all mandatory parameters, and it's possible to override the default values for any of the optional parameters. Parameters are passed in the CLI with the `-p` flag, and in YAML format. (Could be either a path to a YAML file or inline YAML [JSON is a subset of YAML, so inlining could also be in JSON format]).

Example: Executing a workflow with parameters

$ cfy executions start -d my_deployment my_workflow -p my_parameters.yaml
Executing workflow 'my_workflow' on deployment 'my_deployment' at management server [timeout=900 seconds]
2014-12-04T10:02:47 CFY <my_deployment> Starting 'my_workflow' workflow execution
2014-12-04T10:02:47 CFY <my_deployment> 'my_workflow' workflow execution succeeded
Finished executing workflow 'my_workflow' on deployment'my_deployment'
* Run 'cfy events list --include-logs --execution-id 7cfd8b9c-dcd6-41bc-bc88-6aa0b00ffa62' for retrieving the execution's events/logs
mandatory_parameter: mandatory_parameter_value
  key1: overridden_value
Executing the workflow and passing the value mandatory_parameter_value for the mandatory_parameter parameter, and overriding the value of the nested_parameter parameter with a new complex value (though it could have been overridden with a non-complex value as well).

Execution parameters are the actual parameters the execution was run with. To view those in the CLI, use the following command:

cfy executions get my_execution

Example: Retrieving an execution’s parameters

$ cfy executions get -e 7cfd8b9c-dcd6-41bc-bc88-6aa0b00ffa62
Getting execution: '7cfd8b9c-dcd6-41bc-bc88-6aa0b00ffa62' [manager=]

|                  id                  | workflow_id |   status   |         created_at         | error |
| 7cfd8b9c-dcd6-41bc-bc88-6aa0b00ffa62 | my_workflow | terminated | 2014-12-04 10:02:22.728372 |       |

Execution Parameters:
    nested_parameter:       {'key1': 'overridden_value'}
    optional_parameter:     optional_parameter_default_value
    mandatory_parameter:    mandatory_parameter_value

The workflow was executed with three parameters with the presented values. It can be seen that the optional parameter parameter was assigned with its default value, while the nested_parameter parameter’s value was overridden with the new complex value.

Workflow parameters might be defined as intrinsic functions. Given a following blueprint bp:

tosca_definitions_version: cloudify_dsl_1_5

    type: blueprint_id
    type: blueprint_id

    mapping: scripts/
        type: blueprint_id
        default: {get_input: first_blueprint_id}

… deployed with the inputs (provided blueprints b1 and b2 exist and are available to the user):

$ cfy deployments create -b bp -i first_blueprint_id=b1 -i second_blueprint_id=b2 d1

… one could run the wf workflow either with the default blueprint_id (effectively b1):

$ cfy executions start wf -d d1

… or with the default value overwritten by the parameter declared in the params.yaml file (effectively b2):

blueprint_id: {get_input: second_blueprint_id}
$ cfy executions start wf -d d1 -p params.yaml

It is also possible to pass custom parameters that weren’t declared for the workflow in the blueprint. By default, providing such parameters will raise an error, to help avoid mistakes - but if the need for such parameters arises, they can be allowed on a per-execution basis by enabling the allow-custom-parameters flag. For a syntax reference, see the CFY CLI commands reference.