[openstack-dev] improving the keystone middleware cache

Michael Basnight mbasnight at gmail.com
Thu Aug 9 21:01:32 UTC 2012


On Aug 9, 2012, at 3:27 PM, Jay Pipes wrote:

> On 08/08/2012 05:08 PM, Michael Basnight wrote:
>> Ive looked @ the caching mechanism for tokens in the keystone
>> middleware. It appears that the only caching mechanism is memcache if
>> you specify a memcache_servers. I dont see anyone has tackled
>> https://blueprints.launchpad.net/keystone/+spec/unifed-caching-system-for-middleware
>> . Id like to take that task on, given that we can make sure we can use
>> more than just memcache as the cache backend.
> 
> ++ Also, check out the memcache and memory "fake memcache" stuff in
> Nova. It might be useful. Consider putting the effort into
> openstack-common if you think the end result will be commonly useful to
> more than just Keystone.

Jay, fair enough. ive peeped the memcache stuff in nova, and see that its used for caching metadata and a few other things. i agree this should be a common thing at this point. ill look into getting cache code into openstack common and replace the keystone auth middleware & the nova memcache stuff with it, especially if no one has objections. BP to follow for openstack common stuffs.

> 
>> IE, configuration based.
>> Id like to start with beaker (or any other python in memory cache if
>> one is better), and memcache as the two first shot config based
>> caches. Thoughts?
> 
> As mentioned above, check out the simple python memory cache in Nova.
> Beaker would be fine too.
> 
> After that, adding support for Riak, Redis, etc, would be cool.

done. https://github.com/didip/beaker_extensions :D

> 
> Best,
> -jay
> 
> _______________________________________________
> OpenStack-dev mailing list
> OpenStack-dev at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev




More information about the OpenStack-dev mailing list