I’ve been working on (albeit slowly) getting the keystone implementation of 
dogpile.cache into oslo.cache. It’s been slow due to other demands, but I’m 
hoping to get back to it in the near future here so we can make moves like this 
more easily.
—
Morgan Fainberg
Principal Software Engineer
Core Developer, Keystone
[email protected]


On March 31, 2014 at 10:11:21, Dolph Mathews ([email protected]) wrote:

dogpile.cache would be substantially lighter on the client-side as it only has 
a hard dependency on dogpile.core. It supports plenty of backends beyond 
memcached and we already use it in keystone quite heavily.

  http://dogpilecache.readthedocs.org/en/latest/


On Mon, Mar 31, 2014 at 11:35 AM, Doug Hellmann <[email protected]> 
wrote:



On Mon, Mar 31, 2014 at 12:18 PM, Kurt Griffiths <[email protected]> 
wrote:
Hi folks, has there been any discussion on using oslo.cache within the 
auth_token middleware to allow for using other cache backends besides 
memcached? I didn’t find a Keystone blueprint for it, and was considering 
registering one for Juno if the team thinks this feature makes sense. I’d be 
happy to put some time into the implementation.

That does make sense. We need to look at the dependency graph between the 
keystoneclient and oslo.cache, though. It appears the current version of 
oslo.cache is going to bring in quite a few oslo libraries that we would not 
want keystoneclient to depend on [1]. Moving the middleware to a separate 
library would solve that.

[1] https://wiki.openstack.org/wiki/Oslo/Dependencies

Doug


_______________________________________________
OpenStack-dev mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev


_______________________________________________  
OpenStack-dev mailing list  
[email protected]  
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev  
_______________________________________________
OpenStack-dev mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Reply via email to