MemCachier Django usage example
Python HTML JavaScript
Switch branches/tags
Nothing to show
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
memcachier_algebra
.gitignore Tweak readme and dev mode. Sep 17, 2013
LICENSE
Procfile Fix procfile for heroku deployment Mar 2, 2016
Procfile.dev
README.md Markdown fixes to readme Nov 28, 2016
app.json More updating to latest version Nov 28, 2016
dev.env Tweak readme and dev mode. Sep 17, 2013
manage.py
pylibmc_test.py Fix test script Nov 28, 2016
requirements.txt Updrade to django 1.8.16 Nov 28, 2016
runtime.txt

README.md

MemCachier Django Example App

This is an example Django app that uses MemCachier to cache algebraic computations. This example is written with Django 1.8.16.

You can view a working version of this app here that uses MemCachier on Heroku. Running this app on your local machine in development will work as well, although then you won't be using MemCachier -- you'll be using a local dummy cache. MemCachier is currently only available with various cloud providers.

Setting up MemCachier to work in Django is very easy. You need to make changes to requirements.txt, settings.py, and any app code that you want cached. These changes are covered in detail below.

Deploy to Heroku

You can deploy this app yourself to Heroku to play with.

Deploy

Building

It is best to use the python virtualenv tool to build locally:

$ virtualenv-2.7 venv
$ source venv/bin/activate
$ pip install -r requirements.txt
$ DEVELOPMENT=1 python manage.py runserver

Then visit http://localhost:8000 to view the app. Alternatively you can use foreman and gunicorn to run the server locally (after copying dev.env to .env):

$ foreman start

Deploy to Heroku

Run the following commands to deploy the app to Heroku:

$ git clone https://github.com/memcachier/examples-django.git
$ cd examples-django
$ heroku create
$ heroku addons:add memcachier:dev
$ git push heroku master:master
$ heroku open

requirements.txt

MemCachier has been tested with the pylibmc memcache client, but the default client doesn't support SASL authentication. Run the following commands to install the necessary pips:

$ sudo brew install libmemcached
$ pip install django-pylibmc pylibmc

Don't forget to update your requirements.txt file with these new pips. requirements.txt should have the following two lines:

django-pylibmc==0.6.1
pylibmc==1.5.1

Configuring MemCachier (settings.py)

To configure Django to use pylibmc with SASL authentication. You'll also need to setup your environment, because pylibmc expects different environment variables than MemCachier provides. Somewhere in your settings.py file you should have the following lines:

os.environ['MEMCACHE_SERVERS'] = os.environ.get('MEMCACHIER_SERVERS', '').replace(',', ';')
os.environ['MEMCACHE_USERNAME'] = os.environ.get('MEMCACHIER_USERNAME', '')
os.environ['MEMCACHE_PASSWORD'] = os.environ.get('MEMCACHIER_PASSWORD', '')

CACHES = {
    'default': {
        # Use pylibmc
        'BACKEND': 'django_pylibmc.memcached.PyLibMCCache',

        # Use binary memcache protocol (needed for authentication)
        'BINARY': True,

        # TIMEOUT is not the connection timeout! It's the default expiration
        # timeout that should be applied to keys! Setting it to `None`
        # disables expiration.
        'TIMEOUT': None,
        'OPTIONS': {
            # Enable faster IO
            'tcp_nodelay': True,

            # Keep connection alive
            'tcp_keepalive': True,

            # Timeout settings
            'connect_timeout': 2000, # ms
            'send_timeout': 750 * 1000, # us
            'receive_timeout': 750 * 1000, # us
            '_poll_timeout': 2000, # ms

            # Better failover
            'ketama': True,
            'remove_failed': 1,
            'retry_timeout': 2,
            'dead_timeout': 30,
        }
    }
}

Persistent Connections

By default, Django doesn't use persistent connections with memcached. This is a huge performance problem, especially when using SASL authentication as the connection setup is even more expensive than normal.

You can fix this by putting the following code in your wsgi.py file:

# Fix django closing connection to MemCachier after every request (#11331)
from django.core.cache.backends.memcached import BaseMemcachedCache
BaseMemcachedCache.close = lambda self, **kwargs: None

There is a bug file against Django for this issue (#11331).

Application Code

In your application, use django.core.cache methods to access MemCachier. A description of the low-level caching API can be found here. All the built-in Django caching tools will work, too.

Take a look at memcachier_algebra/views.py in this repository for an example.

Get involved!

We are happy to receive bug reports, fixes, documentation enhancements, and other improvements.

Please report bugs via the github issue tracker.

Master git repository:

  • git clone git://github.com/memcachier/examples-django.git

Licensing

This library is BSD-licensed.