Modeling Concepts
Alexis de Talhouët
Jakob Krieg
Steve Alphonse Siani (Deactivated)
Migrated to ReadTheDocs
Further updates must be done in the corresponding RST file(s) by following the build process for documentation.
https://docs.onap.org/projects/onap-ccsdk-cds/en/latest/modelingconcepts/index.html
2020-08-11, Jakob Krieg
CDS is a framework to automate the resolution of resources for instantiation and any config provisioning operation, such as day0, day1 or day2 configuration.
CDS has a both design time and run time activities; during design time, Designer can define what actions are required for a given service, along with anything comprising the action. The design produce a CBA Package . Its content is driven from a catalog of reusable data dictionary and component, delivering a reusable and simplified self service experience.
CDS modelling is mainly based on TOSCA standard, using JSON as reprensentation.
Most of the TOSCA modeled entity presented in the bellow documentation can be found here.
Controller Blueprint Archive (.cba)
The Controller Blueprint Archive is the overall service design, fully model-driven, intent based package needed for provisioning and configuration management automation.
The CBA is .zip file, comprised of the following folder structure, the files may vary:
├── Definitions │ ├── blueprint.json Overall TOSCA service template (workflow + node_template) │ ├── artifact_types.json (generated by enrichment) │ ├── data_types.json (generated by enrichment) │ ├── policy_types.json (generated by enrichment) │ ├── node_types.json (generated by enrichment) │ ├── relationship_types.json (generated by enrichment) │ ├── resources_definition_types.json (generated by enrichment) | └── *-mapping.json One per Template ├── Environments Contains *.properties files as required by the service ├── Plans Contains Directed Graph ├── Tests Contains uat.yaml file for testing cba actions within a cba package ├── Scripts Contains scripts │ ├── python Python scripts │ └── kotlin Kotlin scripts ├── TOSCA-Metadata │ └── TOSCA.meta Meta-data of overall package └── Templates Contains combination of mapping and template
To process a CBA for any service we need to enrich it first. This will gather all the node-type, data-type, artifact-type, data-dictionary definitions provided in the blueprint.json.
Tosca Meta
Tosca meta file is captures the model entities that compose the cba package name, version, type and searchable tags.
| Attribute | R/C/O | Data Type | Description |
|---|---|---|---|
| TOSCA-Meta-File-Version | Required | String | The attribute that holds TOSCA-Meta-File-Version. Set to 1.0.0 |
| CSAR-Version | Required | String | The attribute that holds CSAR-version. Set to 1.0 |
| Created-By | Required | String | The attribute that holds the entry points |
| Entry-Definitions | Required | String | The attribute that holds the entry points file PATH to the main cba tosca definition file or non tosca script file. |
| Template-Name | Required | String | The attribute that holds the blueprint name |
| Template-Version | Required | String | The attribute that holds the blueprint version X.Y.Z X=Major version Y=Minor Version Z=Revision Version Ex. 1.0.0 |
| Template-Type | Required | String | The attribute that holds the blueprint package types. Valid Options:
If not specified in the tosca.meta file the default is "DEFAULT" |
| Template-Tags | Required | String | The attribute that holds the blueprint package comma delimited list of Searchable attributes. |
Template Type Reference
Default Template Type
KOTLIN_DSL Template Type
GENERIC_SCRIPT Template Type
Dynamic payload
One of the most important API provided by the run time is to execute a CBA Package.
The nature of this API request and response is model driven and dynamic.
Here is how the a generic request and response look like.
{
"commonHeader": {
"originatorId": "",
"requestId": "",
"subRequestId": ""
},
"actionIdentifiers": {
"blueprintName": "",
"blueprintVersion": "",
"actionName": "",
"mode": ""
},
"payload": {
"$actionName-request": {
"$actionName-properties": {
}
}
}
}
{
"commonHeader": {
"originatorId": "",
"requestId": "",
"subRequestId": ""
},
"actionIdentifiers": {
"blueprintName": "",
"blueprintVersion": "",
"actionName": "",
"mode": ""
},
"payload": {
"$actionName-response": {
}
}
}
The actionName, under the actionIdentifiers refers to the name of a Workflow (see Modeling Concepts#workflow)
The content of the payload is what is fully dynamic / model driven.
The first top level element will always be either $actionName-request for a request or $actionName-response for a response.
Then the content within this element is fully based on the workflow inputs and outputs.
During the Enrichment, CDS will aggregate all the resources defined to be resolved as input, within mapping definition files, as data-type, that will then be use as type of an input called $actionName-properties.
Enrichment
The idea is that the CBA is a self-sufficient package, hence requires all the various types definition its using.
Reason for this is the types its using might evolve. In order for the CBA to be bounded to the version it has been using when it has been designed, these types are embedded in the CBA, so if they change, the CBA is not affected.
The enrichment process will complete the package by providing all the definition of types used:
- gather all the node-type used and put them into a
node_types.jsonfile - gather all the data-type used and put them into a
data_types.jsonfile - gather all the artifact-type used and put them into a
artifact_types.jsonfile - gather all the data dictionary definitions used from within the mapping files and put them into a
resources_definition_types.jsonfile
Before uploading a CBA, it must be enriched. If your package is already enrich, you do not need to perform enrichment again.
The enrichment can be run using REST API, and required the .zip file as input. It will return an enriched-cba.zip file.
curl -X POST \
'http://{{ip}}:{{cds-designtime}}/api/v1/blueprint-model/enrich' \
-H 'content-type: multipart/form-data' \
-F file=@cba.zip
The enrichment process will also, for all resources to be resolved as input and default:
- dynamically gather them under a data-type, named
dt-${actionName}-properties - will add it as a input of the workflow, as follow using this name:
${actionName}-properties
Example for workflow named resource-assignment:
"resource-assignment-properties": {
"required": true,
"type": "dt-resource-assignment-properties"
}
External Systems support
Interaction with external systems is made dynamic, removing development cycle to support new endpoint.
In order to define the external system information, TOSCA provides dsl_definitions. Link to TOSCA spec info 1, info 2.
Use cases:
- Resource resolution using Modeling Concepts#REST or Modeling Concepts#SQL external systems
- gRPC is supported for remote execution
- Any REST endpoint can be dynamically injected as part of the scripting framework.
Here are some examples on how to populate the system information within the package:
{
. . .
"dsl_definitions": {
"ipam-1": {
"type": "token-auth",
"url": "http://netbox-nginx.netprog:8080",
"token": "Token 0123456789abcdef0123456789abcdef01234567"
}
}
. . .
}
{
. . .
"dsl_definitions": {
"ipam-1": {
"type": "basic-auth",
"url": "http://localhost:8080",
"username": "bob",
"password": "marley"
}
}
. . .
}
{
. . .
"dsl_definitions": {
"ipam-1": {
"type" : "ssl-basic-auth",
"url" : "http://localhost:32778",
"keyStoreInstance": "JKS or PKCS12",
"sslTrust": "trusture",
"sslTrustPassword": "trustore password",
"sslKey": "keystore",
"sslKeyPassword: "keystore password"
}
}
. . .
}
{
. . .
"dsl_definitions": {
"remote-executor": {
"type": "token-auth",
"host": "cds-command-executor.netprog",
"port": "50051",
"token": "Basic Y2NzZGthcHBzOmNjc2RrYXBwcw=="
}
}
. . .
}
{
. . .
"dsl_definitions": {
"netprog-db": {
"type": "maria-db",
"url": "jdbc:mysql://10.195.196.123:32050/netprog",
"username": "netprog",
"password": "netprog"
}
}
. . .
}
Expression
TOSCA provides for a set of functions to reference elements within the template or to retrieve runtime values.
Below is a list of supported expressions
get_input
The get_input function is used to retrieve the values of properties declared within the inputs section of a TOSCA Service Template.
Within CDS, this is mainly Workflow inputs.
Example:
"resolution-key": {
"get_input": "resolution-key"
}
get_property
The get_property function is used to retrieve property values between modelable entities defined in the same service template.
Example
TBD
get_attribute
The get_attribute function is used to retrieve the values of named attributes declared by the referenced node or relationship template name.
Example:
"get_attribute": [ "resource-assignment", "assignment-params" ]
get_operation_output
The get_operation_output function is used to retrieve the values of variables exposed / exported from an interface operation.
Example
TBD
get_artifact
The get_artifact function is used to retrieve artifact location between modelable entities defined in the same service template.
Example
TBD
Data Dictionary
A data dictionary models the how a specific resource can be resolved.
A resource is a variable/parameter in the context of the service. It can be anything, but it should not be confused with SDC or Openstack resources.
A data dictionary can have multiple sources to handle resolution in different ways.
The main goal of data dictionary is to define re-usable entity that could be shared.
Creation of data dictionaries is a standalone activity, separated from the blueprint design.
As part of modelling a data dictionary entry, the following generic information should be provided:
| Property | Description | Scope |
|---|---|---|
| name | Data dictionary name | Mandatory |
| tags | Information related | Mandatory |
| updated-by | The creator | Mandatory |
| property | Defines type and description, as nested JSON | Mandatory |
| sources | List of resource source instance (see resource source) | Mandatory |
Bellow are properties that all the resource source can have
The modeling does allow for data translation between external capability and CDS for both input and output key mapping.
| Property | Description | Scope |
| input-key-mapping | map of resources required to perform the request/query. The left hand-side is what is used within the query/request, the right hand side refer to a data dictionary instance. | Optional |
| output-key-mapping | name of the resource to be resolved mapped to the value resolved by the request/query. | Optional |
| key-dependencies | list of data dictionary instances to be resolved prior the resolution of this specific resource. during run time execution the key dependencies are recursively sorted and resolved in batch processing using the acyclic graph algorithm. | Optional |
Example:vf-module-model-customization-uuid and vf-module-label are two data dictionaries. A SQL table, VF_MODULE_MODEL, exist to correlate them.
Here is how input-key-mapping, output-key-mapping and key-dependencies can be used:
{
"name" : "vf-module-label",
"tags" : "vf-module-label",
"updated-by" : "adetalhouet",
"property" : {
"description" : "vf-module-label",
"type" : "string"
},
"sources" : {
"primary-db" : {
"type" : "source-primary-db",
"properties" : {
"type" : "SQL",
"query" : "select sdnctl.VF_MODULE_MODEL.vf_module_label as vf_module_label from sdnctl.VF_MODULE_MODEL where sdnctl.VF_MODULE_MODEL.customization_uuid=:customizationid",
"input-key-mapping" : {
"customizationid" : "vf-module-model-customization-uuid"
},
"output-key-mapping" : {
"vf-module-label" : "vf_module_label"
},
"key-dependencies" : [ "vf-module-model-customization-uuid" ]
}
}
}
}
Data type
Represents the schema of a specific type of data.
Supports both primitive and complex data types:
| Primitive | Complex |
|---|---|
|
|
For complex data type, an entry schema is required, defining the type of value contained within the complex type, if list or array.
Users can create as many data type as needed.
Creating Custom Data Types
To create a custom data-type you can use a POST call to CDS endpoint: "<cds-ip>:<cds-port>/api/v1/model-type"
Payload:
{
"model-name": "<model-name>",
"derivedFrom": "tosca.datatypes.Root",
"definitionType": "data_type",
"definition": {
"description": "<description>",
"version": "<version-number: eg 1.0.0>",
"properties": {<add properties of your custom data type in JSON format>},
"derived_from": "tosca.datatypes.Root"
},
"description": "<description",
"version": "<version>",
"tags": "<model-name>,datatypes.Root.data_type",
"creationDate": "<creation timestamp>",
"updatedBy": "<name>"
}
Data type are useful to manipulate data during resource resolution. They can be used to format the JSON output as needed.
List of existing data type: https://github.com/onap/ccsdk-cds/tree/master/components/model-catalog/definition-type/starter-type/data_type
Below is a list of existing data types
datatype-resource-assignment
Used to define entries within Modeling Concepts#artifact-mapping-resource
That datatype represent a resource to be resolved. We also refer this as an instance of a data dictionary as it's directly linked to its definition.
| Property | Description |
|---|---|
| property | Defines how the resource looks like (see Modeling Concepts#datatype-property) |
| input-param | Whether the resource can be provided as input. |
| dictionary-name | Reference to the name of the data dictionary (see Data Dictionary). |
| dictionary-source | Reference the source to use to resolve the resource (see Resource source). |
| dependencies | List of dependencies required to resolve this resource. |
| updated-date | Date when mapping was upload. |
| updated-by | Name of the person that updated the mapping. |
{
"version": "1.0.0",
"description": "This is Resource Assignment Data Type",
"properties": {
"property": {
"required": true,
"type": "datatype-property"
},
"input-param": {
"required": true,
"type": "boolean"
},
"dictionary-name": {
"required": false,
"type": "string"
},
"dictionary-source": {
"required": false,
"type": "string"
},
"dependencies": {
"required": true,
"type": "list",
"entry_schema": {
"type": "string"
}
},
"updated-date": {
"required": false,
"type": "string"
},
"updated-by": {
"required": false,
"type": "string"
}
},
"derived_from": "tosca.datatypes.Root"
}
datatype-property
Used to defined the property entry of a resource assignment.
| Property | Description |
|---|---|
| type | Whether it's a primitive type, or a defined data-type |
| description | Description of for the property |
| required | Whether it's required or not |
| default | If there is a default value to provide |
| entry_schema | If the type is a complex one, such as list, define what is the type of element within the list. |
{
"version": "1.0.0",
"description": "This is Entry point Input Data Type, which is dynamic datatype, The parameter names will be populated during the Design time for each inputs",
"properties": {
"type": {
"required": true,
"type": "string"
},
"description": {
"required": false,
"type": "string"
},
"required": {
"required": false,
"type": "boolean"
},
"default": {
"required": false,
"type": "string"
},
"entry_schema": {
"required": false,
"type": "string"
}
},
"derived_from": "tosca.datatypes.Root"
}
Artifact Type
Represents the type of a artifact, used to identify the implementation of the functionality supporting this type of artifact.
This node was created, derived from tosca.artifacts.Root to be the root TOSCA node for all artifact.
{
"description": "TOSCA base type for implementation artifacts",
"version": "1.0.0",
"derived_from": "tosca.artifacts.Root"
}
Bellow is a list of supported artifact types
artifact-template-velocity
Represents an Apache Velocity template.
Apache Velocity allow to insert logic (if / else / loops / etc) when processing the output of a template/text.
File must have .vtl extension.
The template can represent anything, such as device config, payload to interact with 3rd party systems, resource-accumulator template, etc...
Often a template will be parameterized, and each parameter must be defined within an mapping file.
Here is the TOSCA artifact type:
{
"description": " Velocity Template used for Configuration",
"version": "1.0.0",
"file_ext": [
"vtl"
],
"derived_from": "tosca.artifacts.Implementation"
}
artifact-template-jinja
Represents an Jinja template.
Jinja template allow to insert logic (if / else / loops / etc) when processing the output of a template/text.
File must have .jinja extension.
The template can represent anything, such as device config, payload to interact with 3rd party systems, resource-accumulator template, etc...
Often a template will be parameterized, and each parameter must be defined within an mapping file.
Here is the TOSCA artifact type:
{
"description": " Jinja Template used for Configuration",
"version": "1.0.0",
"file_ext": [
"jinja"
],
"derived_from": "tosca.artifacts.Implementation"
}
artifact-mapping-resource
This type is meant to represent mapping files defining the contract of each resource to be resolved.
Each parameter in a template must have a corresponding mapping definition, modeled using Modeling Concepts#datatype-resource-assignment.
Hence the mapping file is meant to be a list of entries defined using Modeling Concepts#datatype-resource-assignment.
File must have .json extension.
Here is the TOSCA artifact type:
{
"description": "Resource Mapping File used along with Configuration template",
"version": "1.0.0",
"file_ext": [
"json"
],
"derived_from": "tosca.artifacts.Implementation"
}
The mapping file basically contains a reference to the data dictionary to use to resolve a particular resource.
The data dictionary defines the HOW and the mapping defines the WHAT.
Relation between data dictionary, mapping and template.
Below are two examples using color coding to help understand the relationships.
In orange is the information regarding the template. As mentioned before, template is part of the blueprint itself, and for the blueprint to know what template to use, the name has to match.
In green is the relationship between the value resolved within the template, and how it's mapped coming from the blueprint.
In blue is the relationship between a resource mapping to a data dictionary.
In red is the relationship between the resource name to be resolved and the HEAT environment variables.
The key takeaway here is that whatever the value is for each color, it has to match all across. This means both right and left hand side are equivalent; it's all on the designer to express the modeling for the service. That said, best practice is example 1.
artifact-directed-graph
Represents a directed graph.
This is to represent a workflow.
File must have .xml extension.
Here is the list of executors currently supported (see here for explanation and full potential list: Service Logic Interpreter Nodes)
- execute
- block
- return
- break
- exit
Here is the TOSCA artifact type:
{
"description": "Directed Graph File",
"version": "1.0.0",
"file_ext": [
"json",
"xml"
],
"derived_from": "tosca.artifacts.Implementation"
}
Node type
In CDS, we have mainly two distinct types: components and source. We have some other type as well, listed in the other section.
Component
Used to represent a functionality along with its contract, such as inputs, ouputs, and attributes
Here is the root component TOSCA node type from which other node type will derive:
{
"description": "This is default Component Node",
"version": "1.0.0",
"derived_from": "tosca.nodes.Root"
}
Bellow is a list of supported components
component-resource-resolution
Used to perform resolution of resources.
Requires as many as Modeling Concepts#artifact-mapping-resource AND Modeling Concepts#artifact-template-velocity or artifact-template-jinja as needed.
Ouput result
Will put the resolution result as an attribute in the workflow context called assignment-params.
Using the Modeling Concepts#get_attribute expression, this attribute can be retrieve to be provided as workflow output (see Workflow).
Specify which template to resolve
Currently, resolution is bounded to a template. To specify which template to use, you need to fill in the `artifact-prefix-names` field.
See Template to understand what the artifact prefix name is.
Storing the result
To store each resource being resolved, along with their status, and the resolved template, `store-result` should be set to `true`.
Also, when storing the data, it must be in the context of either a `resource-id` and resource-type`, or based on a given `resolution-key`
The concept of resource-id / resource-type, or resolution-key, is to uniquely identify a specific resolution that has been performed for a given action. Hence the resolution-key has to be unique for a given blueprint name, blueprint version, action name.
Through the combination of the fields mentioned previously, one could retrieved what has been resolved. This is useful to manage the life-cycle of the resolved resource, the life-cycle of the template, along with sharing with external systems the outcome of a given resolution.
The resource-id / resource-type combo is more geared to uniquely identify a resource in AAI, or external system. For example, for a given AAI resource, say a PNF, you can trigger a given CDS action, and then you will be able to manage all the resolved resources bound to this PNF. Even we could have a history of what has been assigned, unassigned for this given AAI resource.
Important not to confuse and AAI resource (e.g. a topology element, or service related element) with the resources resolved by CDS, which can be seen as parameters required to derived a network configuration.
Run the resolution multiple time
If you need to run the same resolution component multiple times, use the field `occurence`. This will add the notion of occurrence to the resolution, and if storing the results, resources and templates, they will be accessible for each occurrence.
Occurrence is a number between 1 and N; when retrieving information for a given occurrence, the first iteration starts at 1.
This feature is usefull when you need to apply the same configuration accross network elements.
Here is the definition:
{
"description": "This is Resource Assignment Component API",
"version": "1.0.0",
"attributes": {
"assignment-params": {
"required": true,
"type": "string"
}
},
"capabilities": {
"component-node": {
"type": "tosca.capabilities.Node"
}
},
"interfaces": {
"ResourceResolutionComponent": {
"operations": {
"process": {
"inputs": {
"resolution-key": {
"description": "Key for service instance related correlation.",
"required": false,
"type": "string"
},
"occurrence": {
"description": "Number of time to perform the resolution.",
"required": false,
"default": 1,
"type": "integer"
},
"store-result": {
"description": "Whether or not to store the output.",
"required": false,
"type": "boolean"
},
"resource-type": {
"description": "Request type.",
"required": false,
"type": "string"
},
"artifact-prefix-names": {
"required": true,
"description": "Template , Resource Assignment Artifact Prefix names",
"type": "list",
"entry_schema": {
"type": "string"
}
},
"request-id": {
"description": "Request Id, Unique Id for the request.",
"required": true,
"type": "string"
},
"resource-id": {
"description": "Resource Id.",
"required": false,
"type": "string"
},
"action-name": {
"description": "Action Name of the process",
"required": false,
"type": "string"
},
"dynamic-properties": {
"description": "Dynamic Json Content or DSL Json reference.",
"required": false,
"type": "json"
}
},
"outputs": {
"resource-assignment-params": {
"required": true,
"type": "string"
},
"status": {
"required": true,
"type": "string"
}
}
}
}
}
},
"derived_from": "tosca.nodes.Component"
}
component-script-executor
Used to execute a script to perform NETCONF, RESTCONF, SSH commands from within the runtime container of CDS.
Two type of scripts are supported:
- Kotlin: offer a way more integrated scripting framework, along with a way faster processing capability. See more about Kotlin script: https://github.com/Kotlin/KEEP/blob/master/proposals/scripting-support.md
- Python: uses Jython which is bound to Python 2.7, end of life Januray 2020. See more about Jython: https://www.jython.org/
The `script-class-reference` field need to reference
- for kotlin: the package name up to the class. e.g. com.example.Bob
- for python: it has to be the path from the Scripts folder, e.g. Scripts/python/Bob.py
Here is the definition
{
"description": "This is Netconf Transaction Configuration Component API",
"version": "1.0.0",
"interfaces": {
"ComponentScriptExecutor": {
"operations": {
"process": {
"inputs": {
"script-type": {
"description": "Script type, kotlin type is supported",
"required": true,
"type": "string",
"default": "internal",
"constraints": [
{
"valid_values": [
"kotlin",
"jython",
"internal"
]
}
]
},
"script-class-reference": {
"description": "Kotlin Script class name with full package or jython script name.",
"required": true,
"type": "string"
},
"dynamic-properties": {
"description": "Dynamic Json Content or DSL Json reference.",
"required": false,
"type": "json"
}
},
"outputs": {
"response-data": {
"description": "Execution Response Data in JSON format.",
"required": false,
"type": "string"
},
"status": {
"description": "Status of the Component Execution ( success or failure )",
"required": true,
"type": "string"
}
}
}
}
}
},
"derived_from": "tosca.nodes.Component"
}
component-remote-script-executor
Used to execute a python script in a dedicated micro-service, providing a Python 3.6 environment.
Ouput result
prepare-environment-logs: will contain the logs for all the pip install of ansible_galaxy setup
execute-command-logs: will contain the execution logs of the script, that were printed into stdout
Using the Modeling Concepts#get_attribute expression, this attribute can be retrieve to be provided as workflow output (see Workflow).
Params
The `command` field need to reference the path from the Scripts folder of the scripts to execute, e.g. Scripts/python/Bob.py
The `packages` field allow to provide a list of PIP package to install in the target environment, or a requirements.txt file. Also, it supports Ansible role.
If requirements.txt is specified, then it should be provided as part of the Environment folder of the CBA.
"packages": [
{
"type": "pip",
"package": [
"requirements.txt"
]
},
{
"type": "ansible_galaxy",
"package": [
"juniper.junos"
]
}
]
The `argument-properties` allows to specified input argument to the script to execute. They should be expressed in a DSL, and they will be ordered as specified.
"ansible-argument-properties": {
"arg0": "-i",
"arg1": "Scripts/ansible/inventory.yaml",
"arg2": "--extra-vars",
"arg3": {
"get_attribute": [
"resolve-ansible-vars",
"",
"assignment-params",
"ansible-vars"
]
}
}
}
The `dynamic-properties` can be anything that needs to be passed to the script that couldn't be passed as an argument, such as JSON object, etc... If used, they will be passed in as the last argument of the Python script.
Here is the definition
{
"description": "This is Remote Python Execution Component.",
"version": "1.0.0",
"attributes": {
"prepare-environment-logs": {
"required": false,
"type": "string"
},
"execute-command-logs": {
"required": false,
"type": "list",
"entry_schema": {
"type": "string"
}
},
"response-data": {
"required": false,
"type": "json"
}
},
"capabilities": {
"component-node": {
"type": "tosca.capabilities.Node"
}
},
"interfaces": {
"ComponentRemotePythonExecutor": {
"operations": {
"process": {
"inputs": {
"endpoint-selector": {
"description": "Remote Container or Server selector name.",
"required": false,
"type": "string",
"default": "remote-python"
},
"dynamic-properties": {
"description": "Dynamic Json Content or DSL Json reference.",
"required": false,
"type": "json"
},
"argument-properties": {
"description": "Argument Json Content or DSL Json reference.",
"required": false,
"type": "json"
},
"command": {
"description": "Command to execute.",
"required": true,
"type": "string"
},
"packages": {
"description": "Packages to install based on type.",
"required": false,
"type" : "list",
"entry_schema" : {
"type" : "dt-system-packages"
}
}
}
}
}
}
},
"derived_from": "tosca.nodes.Component"
}
component-remote-ansible-executor
Used to execute an ansible playbook hosted in AWX/Anisble Tower.
Ouput result
ansible-command-status: status of the command
ansible-command-logs: will contain the execution logs of the playbook
Using the Modeling Concepts#get_attribute expression, this attribute can be retrieve to be provided as workflow output (see Workflow).
Param
TBD
Here is the definition
{
"description": "This is Remote Ansible Playbook (AWX) Execution Component.",
"version": "1.0.0",
"attributes": {
"ansible-command-status": {
"required": true,
"type": "string"
},
"ansible-command-logs": {
"required": true,
"type": "string"
}
},
"capabilities": {
"component-node": {
"type": "tosca.capabilities.Node"
}
},
"interfaces": {
"ComponentRemoteAnsibleExecutor": {
"operations": {
"process": {
"inputs": {
"job-template-name": {
"description": "Primary key or name of the job template to launch new job.",
"required": true,
"type": "string"
},
"limit": {
"description": "Specify host limit for job template to run.",
"required": false,
"type": "string"
},
"inventory": {
"description": "Specify inventory for job template to run.",
"required": false,
"type": "string"
},
"extra-vars" : {
"required" : false,
"type" : "json",
"description": "json formatted text that contains extra variables to pass on."
},
"tags": {
"description": "Specify tagged actions in the playbook to run.",
"required": false,
"type": "string"
},
"skip-tags": {
"description": "Specify tagged actions in the playbook to omit.",
"required": false,
"type": "string"
},
"endpoint-selector": {
"description": "Remote AWX Server selector name.",
"required": true,
"type": "string"
}
}
}
}
}
},
"derived_from": "tosca.nodes.Component"
}
Source
Used to represent a type of source to resolve a resource, along with the expected properties
Defines the contract to resolve a resource.
Here is the root component TOSCA node type from which other node type will derive:
{
"description": "TOSCA base type for Resource Sources",
"version": "1.0.0",
"derived_from": "tosca.nodes.Root"
}
Bellow is a list of supported sources
Input
Expects the value to be provided as input to the request.
Here is the definition:
{
"description": "This is Input Resource Source Node Type",
"version": "1.0.0",
"properties": {},
"derived_from": "tosca.nodes.ResourceSource"
}
Default
Expects the value to be defaulted in the model itself.
Here is the definition:
{
"description": "This is Default Resource Source Node Type",
"version": "1.0.0",
"properties": {},
"derived_from": "tosca.nodes.ResourceSource"
}
REST
Expects the URI along with the VERB and the payload, if needed.
CDS is currently deployed along the side of SDNC, hence the default rest connection provided by the framework is to SDNC MDSAL.
| Property | Description | Scope |
|---|---|---|
| type | Expected output value, only JSON supported for now | Optional |
| verb | HTTP verb for the request - default value is GET | Optional |
| payload | Payload to sent | Optional |
| endpoint-selector | Specific REST system to interact with to (see Dynamic Endpoint) | Optional |
url-path | URI | Mandatory |
| path | JSON path to the value to fetch from the response | Mandatory |
expression-type | Path expression type - default value is JSON_PATH | Optional |
Here is the definition:
{
"description": "This is Rest Resource Source Node Type",
"version": "1.0.0",
"properties": {
"type": {
"required": false,
"type": "string",
"default": "JSON",
"constraints": [
{
"valid_values": [
"JSON"
]
}
]
},
"verb": {
"required": false,
"type": "string",
"default": "GET",
"constraints": [
{
"valid_values": [
"GET", "POST", "DELETE", "PUT"
]
}
]
},
"payload": {
"required": false,
"type": "string",
"default": ""
},
"endpoint-selector": {
"required": false,
"type": "string"
},
"url-path": {
"required": true,
"type": "string"
},
"path": {
"required": true,
"type": "string"
},
"expression-type": {
"required": false,
"type": "string",
"default": "JSON_PATH",
"constraints": [
{
"valid_values": [
"JSON_PATH",
"JSON_POINTER"
]
}
]
},
"input-key-mapping": {
"required": false,
"type": "map",
"entry_schema": {
"type": "string"
}
},
"output-key-mapping": {
"required": false,
"type": "map",
"entry_schema": {
"type": "string"
}
},
"key-dependencies": {
"required": true,
"type": "list",
"entry_schema": {
"type": "string"
}
}
},
"derived_from": "tosca.nodes.ResourceSource"
}
SQL
Expects the SQL query to be modeled; that SQL query can be parameterized, and the parameters be other resources resolved through other means. If that's the case, this data dictionary definition will have to define key-dependencies along with input-key-mapping.
CDS is currently deployed along the side of SDNC, hence the primary database connection provided by the framework is to SDNC database.
| Property | Description | Scope |
|---|---|---|
| type | Database type, only SQL supported for now | Mandatory |
| endpoint-selector | Specific Database system to interact with to (see Dynamic Endpoint) | Optional |
| query | Statement to execute | Mandatory |
Here is the definition:
{
"description": "This is Database Resource Source Node Type",
"version": "1.0.0",
"properties": {
"type": {
"required": true,
"type": "string",
"constraints": [
{
"valid_values": [
"SQL"
]
}
]
},
"endpoint-selector": {
"required": false,
"type": "string"
},
"query": {
"required": true,
"type": "string"
},
"input-key-mapping": {
"required": false,
"type": "map",
"entry_schema": {
"type": "string"
}
},
"output-key-mapping": {
"required": false,
"type": "map",
"entry_schema": {
"type": "string"
}
},
"key-dependencies": {
"required": true,
"type": "list",
"entry_schema": {
"type": "string"
}
}
},
"derived_from": "tosca.nodes.ResourceSource"
}
Capability
Expects a script to be provided.
| Property | Description | Scope |
|---|---|---|
| script-type | The type of the script - default value is Koltin | Optional |
script-class-reference | The name of the class to use to create an instance of the script | Mandatory |
Here is the definition:
{
"description": "This is Component Resource Source Node Type",
"version": "1.0.0",
"properties": {
"script-type": {
"required": true,
"type": "string",
"default": "kotlin",
"constraints": [
{
"valid_values": [
"internal",
"kotlin",
"jython"
]
}
]
},
"script-class-reference": {
"description": "Capability reference name for internal and kotlin, for jython script file path",
"required": true,
"type": "string"
},
"key-dependencies": {
"description": "Resource Resolution dependency dictionary names.",
"required": true,
"type": "list",
"entry_schema": {
"type": "string"
}
}
},
"derived_from": "tosca.nodes.ResourceSource"
}
Other
dg-generic
Identifies a Directed Graph used as imperative workflow.
| Property | Description | Scope |
|---|---|---|
| dependency-node-templates | The node template the workflow depends on | Required |
Here is the definition:
{
"description": "This is Generic Directed Graph Type",
"version": "1.0.0",
"properties": {
"content": {
"required": true,
"type": "string"
},
"dependency-node-templates": {
"required": true,
"description": "Dependent Step Components NodeTemplate name.",
"type": "list",
"entry_schema": {
"type": "string"
}
}
},
"derived_from": "tosca.nodes.DG"
}
A node_template of this type always provide one artifact, of type artifact-directed-graph, which will be located under the Plans/ folder within the CBA.
"config-deploy-process" : {
"type" : "dg-generic",
"properties" : {
"content" : {
"get_artifact" : [ "SELF", "dg-config-deploy-process" ]
},
"dependency-node-templates" : [ "nf-account-collection", "execute" ]
},
"artifacts" : {
"dg-config-deploy-process" : {
"type" : "artifact-directed-graph",
"file" : "Plans/CONFIG_ConfigDeploy.xml"
}
}
}
In the DG bellow, the execute node refers to the node_template.
<service-logic
xmlns='http://www.onap.org/sdnc/svclogic'
xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance'
xsi:schemaLocation='http://www.onap.org/sdnc/svclogic ./svclogic.xsd' module='CONFIG' version='1.0.0'>
<method rpc='ConfigDeploy' mode='sync'>
<block atomic="true">
<execute plugin="nf-account-collection" method="process">
<outcome value='failure'>
<return status="failure">
</return>
</outcome>
<outcome value='success'>
<execute plugin="execute" method="process">
<outcome value='failure'>
<return status="failure">
</return>
</outcome>
<outcome value='success'>
<return status='success'>
</return>
</outcome>
</execute>
</outcome>
</execute>
</block>
</method>
</service-logic>
tosca.nodes.VNF
Identifies a VNF, can be used to correlate any type of VNF related information.
{
"description": "This is VNF Node Type",
"version": "1.0.0",
"derived_from": "tosca.nodes.Root"
}
vnf-netconf-device
Represents the VNF information to establish a NETCONF communication.
{
"description": "This is VNF Device with Netconf Capability",
"version": "1.0.0",
"capabilities": {
"netconf": {
"type": "tosca.capabilities.Netconf",
"properties": {
"login-key": {
"required": true,
"type": "string",
"default": "sdnc"
},
"login-account": {
"required": true,
"type": "string",
"default": "sdnc-tacacs"
},
"source": {
"required": false,
"type": "string",
"default": "npm"
},
"target-ip-address": {
"required": true,
"type": "string"
},
"port-number": {
"required": true,
"type": "integer",
"default": 830
},
"connection-time-out": {
"required": false,
"type": "integer",
"default": 30
}
}
}
},
"derived_from": "tosca.nodes.Vnf"
}
Workflow
Workflow Scope within CDS Framework
The workflow is within the scope of the micro provisioning and configuration management in controller domain and does NOT account for the MACRO service orchestration workflow which is covered by the SO Project.
A workflow defines an overall action to be taken on the service, hence is an entry-point for the run-time execution of the CBA package.
A workflow also defines inputs and outputs that will defined the payload contract of the request and response (see Dynamic API)
A workflow can be composed of one or multiple sub-actions to execute.
A CBA package can have as many workflows as needed.
Single action
The workflow is directly backed by a Component
In the example bellow, the target of the workflow's steps resource-assignment is `resource-assignment` which actually is the name of the node_template defined after, of type component-resource-resolution.
Link to the example.
. . .
"topology_template": {
"workflows": {
"resource-assignment": {
"steps": {
"resource-assignment": {
"description": "Resource Assign Workflow",
"target": "resource-assignment"
]
}
},
"inputs": {
"resource-assignment-properties": {
"description": "Dynamic PropertyDefinition for workflow(resource-assignment).",
"required": true,
"type": "dt-resource-assignment-properties"
}
},
"outputs": {
"meshed-template": {
"type": "json",
"value": {
"get_attribute": [
"resource-assignment",
"assignment-params"
]
}
}
}
},
"node_templates": {
"resource-assignment": {
"type": "component-resource-resolution",
"interfaces": {
"ResourceResolutionComponent": {
"operations": {
"process": {
"inputs": {
"artifact-prefix-names": [
"vf-module-1"
]
}
}
}
}
},
"artifacts": {
"vf-module-1-template": {
"type": "artifact-template-velocity",
"file": "Templates/vf-module-1-template.vtl"
},
"vf-module-1-mapping": {
"type": "artifact-mapping-resource",
"file": "Templates/vf-module-1-mapping.json"
}
}
}
}
. . .
Multiple sub-actions
The workflow is backed by a Directed Graph engine, dg-generic, and is an imperative workflow.
A DG used as workflow for CDS is composed of multiple execute nodes; each individual execute node refers to an modelled Component instance.
In the example above, you can see the target of the workflow's steps execute-script is `execute-remote-ansible-process`, which is a node_template of type dg_generic
. . .
"topology_template": {
"workflows": {
"execute-remote-ansible": {
"steps": {
"execute-script": {
"description": "Execute Remote Ansible Script",
"target": "execute-remote-ansible-process"
]
}
},
"inputs": {
"ip": {
"required": false,
"type": "string"
},
"username": {
"required": false,
"type": "string"
},
"password": {
"required": false,
"type": "string"
},
"execute-remote-ansible-properties": {
"description": "Dynamic PropertyDefinition for workflow(execute-remote-ansible).",
"required": true,
"type": "dt-execute-remote-ansible-properties"
}
},
"outputs": {
"ansible-variable-resolution": {
"type": "json",
"value": {
"get_attribute": [
"resolve-ansible-vars",
"assignment-params"
]
}
},
"prepare-environment-logs": {
"type": "string",
"value": {
"get_attribute": [
"execute-remote-ansible",
"prepare-environment-logs"
]
}
},
"execute-command-logs": {
"type": "string",
"value": {
"get_attribute": [
"execute-remote-ansible",
"execute-command-logs"
]
}
}
}
}
},
"node_templates": {
"execute-remote-ansible-process": {
"type": "dg-generic",
"properties": {
"content": {
"get_artifact": [
"SELF",
"dg-execute-remote-ansible-process"
]
},
"dependency-node-templates": [
"resolve-ansible-vars",
"execute-remote-ansible"
]
},
"artifacts": {
"dg-execute-remote-ansible-process": {
"type": "artifact-directed-graph",
"file": "Plans/CONFIG_ExecAnsiblePlaybook.xml"
}
}
}
Properties of a workflow
| Property | Description | |||||||
|---|---|---|---|---|---|---|---|---|
| workflow-name | Defines the name of the action that can be triggered by external system | |||||||
| inputs | They are two types of inputs, the dynamic ones, and the static one. static Specified at workflow level
These will end up under dynamic Represent the resources defined as Modeling Concepts#input within mapping definition files. The enrichment process will (see Modeling Concepts#Enrichment)
Example for workflow named resource-assignment: dynamic input "resource-assignment-properties": {
"required": true,
"type": "dt-resource-assignment-properties"
}
| |||||||
| outputs | Defines the outputs of the execution; there can be as many output as needed. Depending on the
| |||||||
| steps | Defines the actual step to execute as part of the workflow
|
Example:
{
"workflow": {
"resource-assignment": { <- workflow-name
"inputs": {
"vnf-id": { <- static inputs
"required": true,
"type": "string"
},
"resource-assignment-properties": { <- dynamic inputs
"required": true,
"type": "dt-resource-assignment-properties"
}
},
"steps": {
"call-resource-assignment": { <- step-name
"description": "Resource Assignment Workflow",
"target": "resource-assignment-process" <- node_template targeted by the step
}
},
"outputs": {
"template-properties": { <- output
"type": "json", <- complex type
"value": {
"get_attribute": [ <- uses expression to retrieve attribute from context
"resource-assignment",
"assignment-params"
]
}
}
}
}
}
}
Template
A template is an artifact, and uses Modeling Concepts#artifact-mapping-resource and artifact-template-velocity.
A template is parameterized and each parameter must be defined in a corresponding mapping file.
In order to know which mapping correlates to which template, the file name must start with an artifact-prefix, serving as identifier to the overall template + mapping.
The requirement is as follows:
${artifact-prefix}-template
${artifact-prefix}-mappingScripts
Library
NetconfClient
In order to facilitate NETCONF interaction within scripts, a python NetconfClient binded to our Kotlin implementation is made available. This NetconfClient can be used when using the component-netconf-executor.
The client can be find here: https://github.com/onap/ccsdk-cds/blob/master/components/scripts/python/ccsdk_netconf/netconfclient.py
ResolutionHelper
When executing a component executor script, designer might want to perform resource resolution along with template meshing directly from the script itself.
The helper can be find here: https://github.com/onap/ccsdk-cds/blob/master/components/scripts/python/ccsdk_netconf/common.py
Southbound Interfaces
CDS comes with native python 3.6 support and Ansible AWX (Ansible Tower): idea is Network Ops are familiar with Python and/or Ansible, and our goal is not to dictate the SBI to use for their operations. Ansible and Python provide already many, and well adopted, SBI libraries, hence they could be utilized as needed.
CDS also provide native support for the following libraries:
- NetConf
- REST
- CLI
- SSH
- gRPC (hence gNMI / gNOI should be supported)
CDS also has extensible REST support, meaning any RESTful interface used for network interaction can be used, such as external VNFM or EMS.
Tests
The tests folder contains the uat.yaml file for execution the cba actions for sunny day and rainy day scenario using mock data. The process to generate the uat file is documented TBD. The file can be dragged and drop to the Tests folder after the test for all actions are executed.
NOTE: You need to activate the "uat" Spring Boot profile in order to enable the spy/verify endpoints. They are disabled by default because the mocks created at runtime can potentially cause collateral problems in production. You can either pass an option to JVM (-Dspring.profiles.active=uat) or set and export an environment variable (export spring_profiles_active=uat).
A quick outline of the UAT generation process follows:
- Create a minimum
uat.yamlcontaining only the NB requests to be sent to the BlueprintsProcessor (BPP) service; - Submit the blueprint CBA and this draft
uat.yamlto BPP in a single HTTP POST call:$ curl -u ccsdkapps:ccsdkapps -F cba=@<path to your CBA file> -F uat=@<path to the draft uat.yaml> http://localhost:8080/api/v1/uat/spy - If your environment is properly setup, at the end this service will generate the complete
uat.yaml; - Revise the generate file, eventually removing superfluous message fields;
- Include this file in your CBA under
Tests/uat.yaml; - Submit the candidate CBA + UAT to be validated by BPP, that now will create runtime mocks to simulate all SB collaborators, by running:
$ curl -u ccsdkapps:ccsdkapps -F cba=@<path to your CBA file> http://localhost:8080/api/v1/uat/verify - Once validated, your CBA enhanced with its corresponding UAT is eligible to be integrated into the CDS project, under the folder
components/model-catalog/blueprint-model/uat-blueprints.
Reference link for sample generated uat.yaml file for pnf plug & play use case: uat.yaml file.
As UAT is part of unit testing, it runs in jenkins job ccsdk-cds-master-verify-java whenever a new commit/patch pushed on gerrit in ccsdk/cds repo.