TNOVAConverter: Converting Network Services from T-NOVA: NSD and VNFD files into UNIFY: NFFG

Introduction

TNOVAConverter is a middleware component which is responsible for creating the connection between Marketplace and ESCAPE and enables service initiations from T-NOVA GUI into UNIFY's resource domain.

Installation

Connector module is implemented entirely in Python.

Requirements

• Python 2.7.6+
• NFFG 1.0

Dependencies

To clone the repository use the following command:

git clone git@5gexgit.tmit.bme.hu:unify/tnova_connector.git

For installing the Python dependencies, simply use the pip tool with the requirements.txt file:

sudo apt install python python-pip
sudo -H pip install flask requests networkx

If the submodules are not initialized (NFFG and Virtualizer), use the following command:

git submodule update --init

Configuration

For simplicity every configuration parameter can be set as a global constant in the connector.py module:

# Listening port
LISTENING_PORT = 5000

# Connector configuration parameters
RO_URL = "http://localhost:8008/escape"  # ESCAPE's top level REST-API

USE_VNF_STORE = False  # enable dynamic VNFD acquiring from VNF Store
VNF_STORE_URL = "http://localhost:8080/NFS/vnfds"
CATALOGUE_DIR = "vnf_catalogue"  # read VNFD from dir if VNFStore is disabled

USE_SERVICE_CATALOG = True  # enable dynamic NSD acquiring from service-catalog
SERVICE_CATALOG_URL = "http://localhost:42050/service/catalog"
NSD_DIR = "nsds"  # dir name used for storing received NSD files

SERVICE_NFFG_DIR = "services"  # dir name used for storing converted services

# Monitoring related parameters
MONITORING_URL = None
MONITORING_TIMEOUT = 2  # sec

# Communication related parameters
USE_CALLBACK = False
CALLBACK_URL = "http://localhost:9000/callback"
USE_VIRTUALIZER_FORMAT = False
ENABLE_DIFF = True

# T-NOVA format constants
NS_ID_NAME = "ns_id"
MESSAGE_ID_NAME = "message-id"
CALLBACK_NAME = "call-back"

# Service request related constants
NFFG_SERVICE_RPC = "sg"
NFFG_TOPO_RPC = "topology"
VIRTUALIZER_TOPO_RPC = "get-config"
VIRTUALIZER_SERVICE_RPC = "edit-config"

# Other constants
PWD = os.path.realpath(os.path.dirname(__file__))
LOGGER_NAME = "TNOVAConnector"
HTTP_GLOBAL_TIMEOUT = 10  # sec

Connector tries to acquire the URLs in the following order:

1. Command line argument (-e; -v)
2. Environment variable (use the name of the constants in the connector script: ESCAPE_URL, CALLBACK_URL and VNF_STORE_URL)
3. Default value defined in the top of the script

Usage

$ ./connector.py -h
usage: connector.py [-h] [-d] [-c [URL]] [-m URL] [-r URL] [-p PORT]
                    [-s VNFSTORE] [-S SERVICECATALOG] [-t t] [-v]

TNOVAConnector: Middleware component which make the connection between
Marketplace and RO with automatic request conversion

optional arguments:
  -h, --help            show this help message and exit
  -d, --debug           run in debug mode (can use multiple times for more
                        verbose logging, default logging level: INFO)
  -c [URL], --callback [URL]
                        enable callbacks from the RO with given URL, default:
                        http://localhost:9000/callback
  -m URL, --monitoring URL
                        URL of the monitoring component, default: None
  -r URL, --ro URL      RO's full URL, default: http://localhost:8008/escape
  -p PORT, --port PORT  REST-API port, default: 5000
  -s VNFSTORE, --vnfstore VNFSTORE
                        enable remote VNFStore with given full URL, default:
                        http://localhost:8080/NFS/vnfds
  -S SERVICECATALOG, --servicecatalog SERVICECATALOG
                        enable remote Service Catalog with given full URL,
                        default: http://localhost:42050/service/catalog
  -t t, --timeout t     timeout in sec for HTTP communication, default: 10s
  -v, --virtualizer     enable Virtualizer format, default: False

REST-API

The RESPT-API calls use no prefix in path by default and follow the syntax: http://<ip>:<port|5000>/<operation>

Operation	Params	HTTP verb	Description
/nsd	NSD desc. in JSON	POST	Send an NSD to the connector, convert to NFFG using local VNFDs or a remote VNF Store and store it
/vnfd	VNFD desc. in JSON	POST	Send a VNFD to the connector and store it locally (for backward compatibility and testing purposes)
/service	NSD id in JSON with key: "ns_id"	POST	Initiate a pre-defined NSD with the NSD id by sending the converted NFFG to ESCAPE
/ns-instances	None	GET	List services
/ns-instances/{id}/terminate	None	PUT	Delete a defined service given by {id}

TNOVAConverter as a Docker container

TNOVAConverter can be run in a Docker container. To create the basic image, issue the following command in the project root:

$ sudo docker build --rm --no-cache -t mdo/tnova_connector .

This command creates a minimal image based on the alpine Python image with the name: tnova_connector, installs the required Python dependencies listed in requirement.txt and sets the entry point.

To create and start a persistent container based on the tnova_connector image, use the following commands:

$ sudo docker run --name connector -p 5000:5000 -p 9000:9000 -it mdo/tnova_connector
$ sudo docker start -i connector

To create a one-time container, use the following command:

$ sudo docker run --rm -p 5000:5000 -p 9000:9000 -ti mdo/tnova_connector

Testing

The project contains a docker-compose config file to test the cooperation of tnova_connector and the Resource Orchestrator aka ESCAPE.

The config file requires pre-built Docker images with specific names:

• tnova_connector for the connector (partial NSO functions)
• mdo/ro for ESCAPE (RO)
• dummy for the dummy orchestrator (dataplane emulation)

These images can be built by the following commands:

# Build connector image
tnova_connector$ sudo docker build --rm --no-cache -t mdo/tnova_connector .
# Build ESCAPE image
escape$ sudo docker build --rm --no-cache -t mdo/ro .
# Build Dummy Orchestrator image
dummy$ sudo docker build --no-cache --rm -t dummy .

# Created images
$ sudo docker images
REPOSITORY            TAG                 IMAGE ID            CREATED             SIZE
dummy                 latest              1fddbe0c5647        19 hours ago        88.3 MB
mdo/ro                latest              1148ca557fa5        19 hours ago        854 MB
mdo/tnova_connector   latest              d26bf0c69424        19 hours ago        95.7 MB
python                2.7.13-alpine       b63d02d8829b        4 weeks ago         71.5 MB
python                2.7.13              b4b107fcc777        8 weeks ago         679 MB

To start the test setup use one of the following commands:

$ sudo docker-compose up
# Run in background
$ sudo docker-compose up -d
$ sudo docker-compose logs -f {escape|tnovaconnector|dummy}

To test the latest code the additional "debug" docker-compose file need to be given. This additional config file attach the project folders into the containers. The locations of the project's code are extracted from environment variables. The following variables must be predefined:

export ESCAPEHOME=...
export DUMMYHOME=...
export TNOVACONNECTORHOME=...

To run the setup in "debug" mode use the following command:

$ sudo docker-compose -f docker-compose.yml -f docker-compose.debug.yml up

License

Licensed under the Apache License, Version 2.0; see LICENSE file.

Copyright (C) 2017 by
János Czentye <janos.czentye@tmit.bme.hu>