openstack/tacker
Code Issues Proposed changes

Add doc of ETSI VNF Deployment as VM with UserData

Browse Source 
Usage guide for ETSI NFV-SOL VNF Deployment as VM with UserData
is added in user docs.

Change-Id: If36586e6ac0cce57fa882d495e99e74ec533ab59
This commit is contained in:
Hiroo Kitamura 2020-12-21 16:24:09 +09:00
parent 4558f9ad37
commit 8d86d50b5a
3 changed files with 348 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

206
doc/source/user/etsi_vnf_deployment_as_vm_with_user_data.rst Normal file
View File

@ -0,0 +1,206 @@
==============================================================
ETSI NFV-SOL VNF Deployment as VM with LCM operation user data
==============================================================
This document describes how to deploy VNF as VM with
LCM operation user data in Tacker using CLI commands.
Please check `LCM-user-data spec`_ and `NFV-SOL014 v2.8.1`_
for more information on LCM operation user data.
.. note::
Most of the content is the same as :doc:`./etsi_vnf_deployment_as_vm_with_tosca`,
but some have 'LCM operation user data' specific descriptions.
The part will be clarified with the notation [This is UserData specific part].
Prerequisites
-------------
The following packages should be installed:
* tacker
* python-tackerclient
A default VIM should be registered according to
:doc:`../cli/cli-legacy-vim`.
The VNF Package(sample_vnf_pkg.zip) used below is prepared
by referring to :doc:`./vnf-package`.
.. note:: [This is UserData specific part]
In order to deploy VNF using LCM operation user data,
it is necessary to include BaseHOT and UserData in the VNF Package,
which are optional in :doc:`./vnf-package`.
VNF deployment procedure as VM
------------------------------
In order to deploy VNF as a VM, it is necessary to execute
the following procedure.
Details of CLI commands are described in
:doc:`../cli/cli-etsi-vnfpkgm` and :doc:`../cli/cli-etsi-vnflcm`.
Many procedures are the same as :doc:`./etsi_vnf_deployment_as_vm_with_tosca`,
so refer to the doc for those procedures.
1. Create VNF Package Info
^^^^^^^^^^^^^^^^^^^^^^^^^^
See procedure in :doc:`./etsi_vnf_deployment_as_vm_with_tosca`.
2. Upload VNF Package
^^^^^^^^^^^^^^^^^^^^^
See procedure in :doc:`./etsi_vnf_deployment_as_vm_with_tosca`.
3. Create VNF Identifier
^^^^^^^^^^^^^^^^^^^^^^^^
See procedure in :doc:`./etsi_vnf_deployment_as_vm_with_tosca`.
4. Instantiate VNF
^^^^^^^^^^^^^^^^^^
Create a sample_param_file.json file with the following format.
This is the file that defines the parameters for an instantiate request.
These parameters will be set in the body of the instantiate request.
Required parameter:
* flavourId
* additionalParams
.. note::
Details of flavourId is described in :doc:`./vnfd-sol001`.
.. note::
[This is UserData specific part]
additionalParams is a parameter that can be described by KeyValuePairs.
By setting the following two parameters in this parameter,
instantiate using LCM operation user data becomes possible.
For file_name.py and class_name, set the file name and class name
described in Prerequisites.
* lcm-operation-user-data: ./UserData/file_name.py
* lcm-operation-user-data-class: class_name
Optional parameters:
* instantiationLevelId
* extVirtualLinks
* extManagedVirtualLinks
* vimConnectionInfo
.. note::
You can skip `vimConnectionInfo` only when you have
the default VIM described in :doc:`../cli/cli-legacy-vim`.
Param file with only required parameters:
.. code-block:: console
{
"flavourId":"simple",
"additionalParams": {
"lcm-operation-user-data": "./UserData/lcm_user_data.py",
"lcm-operation-user-data-class": "SampleUserData"
}
}
Param file with optional parameters:
.. code-block:: console
{
"flavourId": "simple",
"instantiationLevelId": "instantiation_level_1",
"extVirtualLinks": [
{
"id": "net0",
"resourceId": "4bf3e646-7a24-4f04-a985-d8f4bb1203de", #Set the uuid of the network to use
"extCps": [
{
"cpdId": "CP1",
"cpConfig": [
{
"cpProtocolData": [
{
"layerProtocol": "IP_OVER_ETHERNET"
}
]
}
]
}
]
}
],
"vimConnectionInfo": [
{
"id": "e24f9796-a8e9-4cb0-85ce-5920dcddafa1", #Set a random uuid.
"vimId": "8a0fd79d-e224-4c27-85f5-ee79c6e0d870", #Set the uuid of the VIM to use
"vimType": "ETSINFV.OPENSTACK_KEYSTONE.v_2"
}
],
"additionalParams": {
"lcm-operation-user-data": "./UserData/lcm_user_data.py",
"lcm-operation-user-data-class": "SampleUserData"
}
}
Execute the following CLI command to instantiate the VNF instance.
.. code-block:: console
$ openstack vnflcm instantiate VNF_INSTANCE_ID \
./sample_param_file.json
Result:
.. code-block:: console
Instantiate request for VNF Instance 725f625e-f6b7-4bcd-b1b7-7184039fde45 has been accepted.
After that, execute the following CLI command and confirm that
VNF instance instantiation was successful.
* Confirm that the 'Instantiation State' is 'INSTANTIATED'.
.. code-block:: console
$ openstack vnflcm show VNF_INSTANCE_ID \
-c 'Instantiation State'
Result:
.. code-block:: console
+---------------------+--------------+
| Field | Value |
+---------------------+--------------+
| Instantiation State | INSTANTIATED |
+---------------------+--------------+
5. Terminate VNF
^^^^^^^^^^^^^^^^
See procedure in :doc:`./etsi_vnf_deployment_as_vm_with_tosca`.
6. Delete VNF Identifier
^^^^^^^^^^^^^^^^^^^^^^^^
See procedure in :doc:`./etsi_vnf_deployment_as_vm_with_tosca`.
.. _LCM-user-data spec : https://specs.openstack.org/openstack/tacker-specs/specs/ussuri/lcm-operation-with-lcm-operation-user-data.html
.. _NFV-SOL014 v2.8.1 : https://www.etsi.org/deliver/etsi_gs/NFV-SOL/001_099/014/02.08.01_60/gs_NFV-SOL014v020801p.pdf

1
doc/source/user/index.rst
View File

@ -51,6 +51,7 @@ Use cases
dynamic_vnffg_usage_guide.rst
enhanced_placement_awareness_usage_guide.rst
etsi_vnf_deployment_as_vm_with_tosca.rst
etsi_vnf_deployment_as_vm_with_user_data.rst
etsi_containerized_vnf_usage_guide.rst
etsi_vnf_healing.rst
containerized_vnf_usage_guide.rst

142
doc/source/user/vnf-package.rst
View File

@ -31,6 +31,17 @@ The directory structure:
* **Files/images/**
* (optional) **Scripts/**
* (optional) **<manifest file name>.mf**
* (optional) **BaseHOT/**
* (optional) **UserData/**
.. note:: BaseHOT and UserData are optional, but they are required when
running LCM operation user data.
This methodology is under discussion within `NFV-SOL014 v2.8.1`_
and there is no clear directory structure yet.
Please check with :doc:`./etsi_vnf_deployment_as_vm_with_user_data`.
The specification can be modified according to standardization
`NFV-SOL014 v2.8.1`_.
.. code-block::
@ -52,6 +63,15 @@ The directory structure:
!----Scripts (optional)
!---- install.sh
!---- manifest.mf
!----BaseHOT (optional)
!---- df_1
!---- base_hot_df_1.yaml
!---- ..
!---- df_x
!---- base_hot_df_x.yaml
!----UserData (optional)
!---- __init__.py
!---- lcm_user_data.py
TOSCA-Metadata/TOSCA.meta
@ -207,6 +227,123 @@ Example:
Hash: 36f945953929812aca2701b114b068c71bd8c95ceb3609711428c26325649165
BaseHOT/
^^^^^^^^
Base HOT file is a Native cloud orchestration template, HOT in this context,
which is commonly used for LCM operations in different VNFs.
This Base HOT can work on OpenStack API and be filled by Heat input parameters
generated by LCM operation user data.
It is the responsibility of the user to prepare this file, and it is necessary
to make it consistent with VNFD placed under the **Definitions/** directory.
.. note:: Place the directory corresponding to deployment-flavour stored in
the **Definitions/** under the **BaseHOT/** directory, and store the
Base HOT files in it.
In this DQ example, it is assumed that there is a deployment-flavour
from `df_1` to` df_x`.
For more information on deployment-flavour, see
`NFV-SOL001 v2.6.1`_ Annex A.
Example:
.. code-block:: yaml
heat_template_version: 2013-05-23
description: 'Template for test _generate_hot_from_tosca().'
parameters:
nfv:
type: json
resources:
VDU1:
type: OS::Nova::Server
properties:
flavor:
get_resource: VDU1_flavor
name: VDU1
image: { get_param: [ nfv, VDU, VDU1, image ] }
networks:
- port:
get_resource: CP1
CP1:
type: OS::Neutron::Port
properties:
network: { get_param: [ nfv, CP, CP1, network ] }
VDU1_flavor:
type: OS::Nova::Flavor
properties:
ram: { get_param: [ nfv, VDU, VDU1, flavor, ram ] }
vcpus: { get_param: [ nfv, VDU, VDU1, flavor, vcpus ] }
disk: { get_param: [ nfv, VDU, VDU1, flavor, disk ] }
outputs: {}
.. note:: For property (e.g. image in VDU1) whose value is "get_param",
the Heat input parameters generated by script placed under
**UserData/** directory.
UserData/
^^^^^^^^^
LCM operation user data is a script that returns key/value data as
Heat input parameters used for Base HOT.
As Heat input parameter, OpenStack parameters that are not statically defined
in the VNFD(e.g. flavors, images, hardware acceleration, driver-setup, etc.)
can be assigned.
.. note:: It is necessary to generate Heat input parameters for HOT file
This script has the following advantages/disadvantages for VNF
package creators/users.
The advantage is that this script has no operational restrictions,
so it can be freely described by creators and operated by users.
The disadvantage is that creators can write a script that
involves DB operations, which can lead to unexpected behavior for users.
The trade-off between being able to write scripts freely and
limiting operations is an issue for the future.
Example:
.. code-block:: python
import tacker.vnfm.lcm_user_data.utils as UserDataUtil
from tacker.vnfm.lcm_user_data.abstract_user_data import AbstractUserData
class SampleUserData(AbstractUserData):
@staticmethod
def instantiate(base_hot_dict=None,
vnfd_dict=None,
inst_req_info=None,
grant_info=None):
# Create HOT input parameter using util functions.
initial_param_dict = UserDataUtil.create_initial_param_dict(
base_hot_dict)
vdu_flavor_dict = UserDataUtil.create_vdu_flavor_dict(vnfd_dict)
vdu_image_dict = UserDataUtil.create_vdu_image_dict(grant_info)
cpd_vl_dict = UserDataUtil.create_cpd_vl_dict(
base_hot_dict, inst_req_info)
final_param_dict = UserDataUtil.create_final_param_dict(
initial_param_dict, vdu_flavor_dict, vdu_image_dict, cpd_vl_dict)
return final_param_dict
.. note:: It is necessary to generate Heat input parameters for HOT file
placed under **BaseHOT/** directory by this scprit.
CSAR zip without TOSCA-Metadata directory
-----------------------------------------
@ -234,7 +371,7 @@ The file structure:
^^^^^^^^^^^^^^^^^^^^^
This is the top-level VNFD file. It can import additional VNFD files from
**Definitions/** directory.
the **Definitions/** directory.
Definitions/
^^^^^^^^^^^^
@ -259,5 +396,8 @@ YAML file. The contents is exactly same as described in the previous section.
.. _Glance configuration : https://docs.openstack.org/glance/latest/user/signature.html#using-the-signature-verification
.. _ETSI NFV-SOL001 repository : https://forge.etsi.org/rep/nfv/SOL001
.. _IETF RFC 3339 : https://tools.ietf.org/html/rfc3339
.. _NFV-SOL001 v2.6.1 : https://www.etsi.org/deliver/etsi_gs/NFV-SOL/001_099/001/02.06.01_60/gs_NFV-SOL001v020601p.pdf
.. _NFV-SOL004 v2.6.1 : https://www.etsi.org/deliver/etsi_gs/NFV-SOL/001_099/004/02.06.01_60/gs_NFV-SOL004v020601p.pdf
.. _NFV-SOL005 v2.6.1 : https://www.etsi.org/deliver/etsi_gs/NFV-SOL/001_099/005/02.06.01_60/gs_NFV-SOL005v020601p.pdf
.. _NFV-SOL014 v2.8.1 : https://www.etsi.org/deliver/etsi_gs/NFV-SOL/001_099/014/02.08.01_60/gs_NFV-SOL014v020801p.pdf