Skip to content

Home

Jump to bottom
Ravindhar Uppada edited this page Sep 3, 2020 · 19 revisions

Welcome to the HPE OneView SDK for Python (oneview-python) wiki!

Introduction

HPE OneView makes it simple to deploy and manage today’s complex hybrid cloud infrastructure. HPE OneView can help you transform your data center to software-defined, and it supports HPE’s broad portfolio of servers, storage, and networking solutions, ensuring the simple and automated management of your hybrid infrastructure. Software-defined intelligence enables a template-driven approach for deploying, provisioning, updating, and integrating compute, storage, and networking infrastructure.

The HPE OneView Python library provides a pure Python interface to the HPE OneView REST APIs. It depends on the Python-Future library to provide Python 2/3 compatibility.

Installation

HPE OneView SDK for Python can be installed automatically through any of the installation methods described below.

1. From Source

Either:

$ git clone https://github.com/HewlettPackard/oneview-python.git
$ cd oneview-python
$ python setup.py install --user  # to install in the user directory (~/.local)
$ sudo python setup.py install    # to install globally

Or if using PIP:

$ git clone https://github.com/HewlettPackard/oneview-python.git
$ cd oneview-python
$ pip install .

2. From Pypi

$ git clone https://github.com/HewlettPackard/oneview-python.git
$ cd oneview-python
$ pip install hpeOneView

3. From Docker Image / Container

Running Examples with local docker container

If you'd rather run the examples in a Docker container, you can use the Dockerfile at the top level of this repo. All you need is Docker and git (optional).

  1. Clone this repo and cd into it:

    $ git clone https://github.com/HewlettPackard/oneview-python.git
$ cd oneview-python

    Note: You can navigate to the repo url and download the repo as a zip file if you don't want to use git.

  2. Build the docker image: $ docker build -t oneview-python .

    Note: If you're behind a proxy, please edit the Dockerfile before building, uncommenting/adding the necessary ENV directives for your environment.

  3. Now you can run any of the example in this directory:

    # Run the container, passing in your credentials to OneView and specifying which example recipe to run.
# -v : The volume on which repo code is mounted
# Replace "connection_templates" with the name of the example you'd like to run
# Replace "pwd" with the path of the example file you'd like to run.

$ docker run -it --rm \
  -v $(pwd)/:/root/oneview/
  python examples/connection_templates.py

That's it! If you'd like to modify a example, simply modify the example file (on the host), then re-run the image.

Running Examples with published docker image

We also provide a lightweight and easy way to test and run oneview-python. The hewlettpackardenterprise/hpe-oneview-sdk-for-python:<tag> docker image contains an installation of oneview-python installation you can use by just pulling down the Docker Image:

The Docker Store image tag consist of two sections: <sdk_version-OV_version>

# Download and store a local copy of hpe-oneview-sdk-for-python and
# use it as a Docker image.
$ docker pull hewlettpackardenterprise/hpe-oneview-sdk-for-python:v5.2.0-OV5.2

# Run docker commands and this will in turn create
# a sh session where you can create files, issue commands and execute the tests
$ docker run -it hewlettpackardenterprise/hpe-oneview-sdk-for-python:v5.2.0-OV5.2 /bin/sh

All above installation methods work if you are using virtualenv, which you should be!

API Implementation

A status of the HPE OneView REST interfaces that have been implemented in this Python library can be found in the Wiki section.

SDK Documentation

The latest version of the OneView Python SDK documentation can be found in the SDK Documentation section.

Configuration

1. JSON:

Connection properties for accessing the OneView appliance can be set in a JSON file.

Before running the samples or your own scripts, you must create the JSON file. An example can be found at: OneView configuration sample.

Note: If you have an active and valid login session and want to use it, define the sessionID in the Credentials. When sessionID is defined, you can remove username and password from your JSON (they will be disregarded anyway).

Once you have created the JSON file, you can initialize the OneViewClient:

oneview_client = OneViewClient.from_json_file('/path/config.json')

🔒 Tip: Check the file permissions because the password is stored in clear-text.

2. Environment Variables:

Configuration can also be defined through environment variables:

# Required
export ONEVIEWSDK_IP='172.16.102.82'

export ONEVIEWSDK_USERNAME='Administrator'
export ONEVIEWSDK_PASSWORD='secret123'
# Or sessionID
export ONEVIEWSDK_SESSIONID='123'

# Optional
export ONEVIEWSDK_API_VERSION='800'
export ONEVIEWSDK_AUTH_LOGIN_DOMAIN='authdomain'
export ONEVIEWSDK_SSL_CERTIFICATE='<path_to_cert.crt_file>'
export ONEVIEWSDK_PROXY='<proxy_host>:<proxy_port>'
export ONEVIEWSDK_CONNECTION_TIMEOUT='<connection time-out in seconds>'

Once you have defined the environment variables, you can initialize the OneViewClient using the following code snippet:

oneview_client = OneViewClient.from_environment_variables()

🔒 Tip: Make sure no unauthorized person has access to the environment variables, since the password is stored in clear-text.

Note: If you have an active and valid login session and want to use it, define the ONEVIEWSDK_SESSIONID. When a sessionID is defined, it will be used for authentication (username and password will be ignored in this case).

3. Dictionary:

You can also set the configuration using a dictionary. As described above, for authentication you can use username/password:

config = {
    "ip": "172.16.102.82",
    "credentials": {
        "userName": "Administrator",
        "password": "secret123"
    }
}

or if you have an active and valid login session and want to use it, define the sessionID in the Credentials:

config = {
    "ip": "172.16.102.82",
    "credentials": {
        "sessionID": "123"
    }
}

oneview_client = OneViewClient(config)

🔒 Tip: Check the file permissions because the password is stored in clear-text.

4. Logging - How to configure logging?:

This module uses Python’s Standard Library logging module. An example of how to configure logging is provided on /examples/logger.py.

More information about the logging configuration can be found in the Python Documentation.

5. SSL Server Certificate:

To enable the SDK to establish a SSL connection to the HPE OneView server, it is necessary to generate a CA Cert file containing the server credentials.

  1. Fetch the HPE OneView Appliance CA certificate.
$ openssl s_client -showcerts -host <host> -port 443
  1. Copy the server certificate wrapped with a header line and a footer line into a <file_name>.crt file.
-----BEGIN CERTIFICATE-----
... (HPE OneView Appliance certificate in base64 PEM encoding) ...
-----END CERTIFICATE-----

When using HPE Image Streamer, the server certificate for the HPE Image Streamer should also be added to the certificates file. Example:

-----BEGIN CERTIFICATE-----
... (HPE OneView Appliance certificate in base64 PEM encoding) ...
-----END CERTIFICATE-----
-----BEGIN CERTIFICATE-----
... (HPE Image Streamer Appliance certificate in base64 PEM encoding) ...
-----END CERTIFICATE-----
  1. Declare the CA Certificate location when creating a config dictionary.
config = {
    "ip": "172.16.102.82",
    "credentials": {
        "userName": "Administrator",
        "password": "secret123"
    },
    "ssl_certificate": "/home/oneview-python/my_ov_certificate.crt"
}

6. Proxy Setting:

If your environment requires a proxy, define the proxy properties in the JSON file using the following syntax:

  "proxy": "<proxy_host>:<proxy_port>"

OR export the ONEVIEWSDK_PROXY environment variable:

export ONEVIEWSDK_PROXY='<proxy_host>:<proxy_port>'

7. Setting your OneView version

The OneView Python SDK now supports the API endpoints for HPE OneView 5.00, 5.20, 5.30, 5.40

The current default HPE OneView version used by the Python SDK will pick the OneView appliance API version. To use a different API, you must set the API version on the OneViewClient configuration, either using the JSON configuration:

"api_version": 1600

OR using the Environment variable:

export ONEVIEWSDK_API_VERSION='1600'

If this property is not specified, it will fall back to the default value as per OneView appliance.

The API list is as follows:

  • HPE OneView 5.00 API version: 1200
  • HPE OneView 5.20 API version: 1600
  • HPE OneView 5.30 API version: 1800
  • HPE OneView 5.40 API version: 2000

8. HPE Synergy Image Streamer

The OneView Python SDK also supports the API endpoints for HPE Synergy Image Streamer. To configure the SDK for Image Streamer, you must set the Image Streamer IP on the OneViewClient configuration, either using the JSON configuration:

"image_streamer_ip": "100.100.100.100"

OR using the Environment variable:

export ONEVIEWSDK_IMAGE_STREAMER_IP='100.100.100.100'

To create the ImageStreamerClient, you must call the create_image_streamer_client method from the pre-existent OneViewClient instance. Through the created ImageStreamerClient, you are able to access the API Clients of the Image Streamer resources:

image_streamer_client = oneview_client.create_image_streamer_client()

build_plans = image_streamer_client.build_plans.get_all()

You can find more usage examples in the folder /examples/image_streamer

9. OneView Connection Timeout

By default the system timeout is used when connecting to OneView. If you want to change this, then the timeout can be set by either:

  1. Setting the time-out in the JSON configuration file using the following syntax:
"timeout": <timeout in seconds>

or

  1. Setting the appropriate environment variable:
export ONEVIEWSDK_CONNECTION_TIMEOUT='<connection time-out in seconds>'

10. Exception handling

All exceptions raised by the OneView Python SDK inherit from HPEOneViewException.

HPEOneViewException has the following properties:

  • msg - a string containing the error message sent by the OneView REST API;
  • oneview_response - contains the entire JSON data dictionary with error details that are returned by the OneView Python SDK. It can be: None.
Exception Handling example:
try:
    fc_network_client = oneview_client.fc_networks
    fc_network = fc_network_client.get_by_name(name)
except HPEOneViewException as e:
    print(e.msg)
    if e.oneview_response:
    	pprint(e.oneview_response)

For compatibility purposes, the Exception args property is defined with the error arguments. For example:

except Exception as e:
	print(arg[0]) # e.msg equivalent
    print(arg[1]) # e.oneview_reponse equivalent

Contributing and feature requests

Contributing: You know the drill. Fork it, branch it, change it, commit it, and pull-request it. We are passionate about improving this project, and glad to accept help to make it better.

NOTE: We reserve the right to reject changes that we feel do not fit the scope of this project. For feature additions, please open an issue to discuss your ideas before doing the work.

Feature Requests: If you have needs not being met by the current implementation, please let us know (via a new issue). This feedback is crucial for us to deliver a useful product. Do not assume we have already thought of everything, because we assure you that is not the case.

Naming Convention for OneView Resources

The following summarizes code structure and naming conventions for the OneView resources.

  • Packages: The package is named according to the HPE OneView API Reference group, with all characters in lowercase, replacing spaces with underscores.
  • Modules: The module is named according to the HPE OneView API Reference endpoint title, with all characters in lowercase, replacing spaces with underscores. For example: In the documentation we have FC Networks, so the module name will be fc_networks.
  • Classes: We are using camel case to define the class name, for example: FcNetworks.
  • OneViewClient properties: In the oneview_client, the property name follows exactly the module name, for example: fc_networks.
  • Examples: The example is named with the same name of the resource module: fc_networks.
  • Tests: The unit test folders follow the same structure of the resources. The name of the test modules should start with "test," for example: test_fc_networks.

Testing

When contributing code to this project, we require tests to accompany the code being delivered. That ensures a higher standing of quality, and also helps to avoid minor mistakes and future regressions.

When writing the unit tests, the standard approach we follow is to use the python library unittest.mock to patch all calls that would be made to a OneView appliance and return mocked values.

We have packaged everything required to verify if the code is passing the tests in a tox file. The tox call runs all unit tests against Python 2 and 3, runs a flake8 validation, and generates the test coverage report.

To run it, use the following command:

$ tox

You can also check out examples of tests for different resources in the tests folder.

License

This project is licensed under the Apache license. Please see LICENSE for more information.

Version and Changes

To view history and notes for this version, view the Changelog

Wiki Table of Contents

Clone this wiki locally