A Python interface to MUNGE
pymunge is a Python wrapper for the C API of MUNGE, called libmunge. pymunge provides functions and classes to create and validate credentials with MUNGE, and to use and interact with MUNGE contexts.
MUNGE (MUNGE Uid ‘N’ Gid Emporium, https://dun.github.io/munge/) is an authentication service for creating and validating credentials designed to be highly scalable for use in an HPC cluster environment.
Official pymunge repository: https://github.com/nomadictype/pymunge
PyPI project page: https://pypi.python.org/pypi/pymunge
API reference: https://pymunge.readthedocs.io/en/latest/
Python 3.4 or later (or Python 2.7 with the ‘enum34’ package).
MUNGE 0.5.x or later.
A munged daemon must be running on the same machine in order for pymunge to be able to create and validate credentials.
Make sure that all the above requirements are satisfied. Afterwards, there are several possible ways to proceed:
To install pymunge from PyPI, run the following command (preferably in a virtualenv):
python3 -m pip install pymunge
Alternatively, your OS distribution may include pymunge as a package, with a name such as pymunge, python3-pymunge, or python-pymunge.
pymunge can also be used directly without installation. Just ensure that Python can find the pymunge package (for example by appending the parent directory of the pymunge package to the PYTHONPATH environment variable).
Getting started / Tutorial
This quick tutorial describes how to use the pymunge API. If you want, you can follow along in an interactive Python 3 session; simply copy all the code preceded by >>>.
First of all, import the package:
>>> import pymunge
The simplest way to encode (= create) and decode (= validate) credentials is to use the pymunge.encode() and pymunge.decode() functions. For example:
>>> cred = pymunge.encode(b"some payload") >>> cred b'MUNGE:AwQDAA...'
The credential cred can now be sent to some other process to decode it (via a socket or some other IPC mechanism) – this is the responsibility of the program which uses pymunge, pymunge does not provide any functions to do this! For testing purposes, you can also pipe the credential into the unmunge program by hand. To keep this tutorial simple, let us decode the credential directly in the same process:
>>> payload, uid, gid, ctx = pymunge.decode(cred) >>> payload b'some payload'
pymunge.decode() returns 4 values: the payload that was encapsulated within the credential, the UID/GID of the process that created the credential, and a MUNGE context. This context can be examined to obtain additional information about the credential:
>>> ctx.cipher_type <CipherType.AES128: 4> >>> ctx.encode_time 1514469923 >>> ctx.ttl 300 >>> ctx.uid_restriction -1
(Also try running help(ctx) to see a list of all attributes a context can have.)
It is possible to encode and decode within existing MUNGE contexts. This is useful to customize the options used to encode a credential:
>>> with pymunge.MungeContext() as ctx: >>> ctx.uid_restriction = 0 # allow only root to decode the credential >>> cred = ctx.encode(b"some other payload")
Similarly, MungeContext.decode() can be used to decode within an existing context.
This concludes the basic tutorial. A collection of similar examples is provided in the file pymunge_example.py distributed with pymunge.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Hashes for pymunge-0.1.3-py2.py3-none-any.whl