Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Cache Puppet data in memcached for speed reasons
Branch: master

Memcache key creation was not including request key

Signed-off-by: Luke Kanies <>
latest commit 59be6f71cd
@lak authored


Puppet Memcached

Prototype integration with memcached. This is purely for demonstration purposes and should not be used in production, I bet.

This caches all file metadata in memcached, and returns that metadata rather than recalculating it each time.

The data expires after an hour or so by default, but the keys in the data use the Puppet config_version; if you change the config version, then you'll begin using a new set of cached data, and the old data will expire eventually.

In production, you should probably plan on having a config_version script, most likely producing your current git version or something similar.

This actually replaces your default 'rest' terminus for file_metadata, so to use it, just stick it somewhere your server can read it, and put it in front of your Puppet install in your RUBYLIB.

Something went wrong with that request. Please try again.