I'm trying to use django-cache-machine to cache queries within my application, but I want to use Redis as a backend. The docs don't really explain how to do this, yet the repository is filled with Redis references, so I'm pretty sure it's possible. I want to make sure I do it right though, so I'm wondering if anyone has any experience with configuring this and maybe more importantly, knows if there are any caveats?
Add a comment
|
2 Answers
In your settings set:
CACHE_MACHINE_USE_REDIS = True
REDIS_BACKEND = redis://127.0.0.1:6379?socket_timeout=0.1
https://github.com/jbalogh/django-cache-machine/blob/master/caching/invalidation.py#L187 https://github.com/jbalogh/django-cache-machine/blob/master/caching/invalidation.py#L213
1 Comment
tobias.mcnulty
Please note, when setting CACHE_MACHINE_USE_REDIS = True, django-cache-machine doesn't actually use Redis to cache your model objects, so you still need memcached configured. It only uses Redis to store and manipulate flush lists.
I have a little experience in my project, a report system that generate tables from about 50 million records.
The database is Mysql and I could show my settings and models FYI.
settings:
# cache machine
CACHES = {
'default': {
'BACKEND': 'caching.backends.memcached.MemcachedCache',
'LOCATION': [
'127.0.0.1:11211',
],
'PREFIX': 'report:',
},
}
CACHE_COUNT_TIMEOUT = 60 * 24 # one day
CACHE_EMPTY_QUERYSETS = True
models:
class App(**CachingMixin**, models.Model):
**objects = CachingManager()**
name = models.CharField(max_length=64,
default='')
Note that cache-machine works fine for query_set.filter and count, not good for query_set.annotate or aggregate. Of course do not forget launch your memcache client first.
And when running you can see cache-machine logs in your django*.log to tell you hit or miss cache.
