[infinispan-dev] Infinispan 4.1 | Eviction policies for Memcached protocol

Vladimir Blagojevic vblagoje at redhat.com
Thu Sep 23 14:50:43 EDT 2010


On 2010-09-23, at 12:56 PM, Galder Zamarreño wrote:

> Hi Owen,
> 
> On Sep 23, 2010, at 6:34 PM, Owen Tran wrote:
> 
>> Hi all,
>> 
>> Thank you all for the quick responses. I was expecting to be able to use both as well, Sanne.
>> 
>> I wrote a junit test setting expiration -1 and doesn't matter. Once you put them in a cache with a time, it make those entries mortal.
>> 
>> Do you still want me to create a bug in Jira or is this working as design? I'd be happy to update the documentation as well, since I spent a lot of time combing through the docs. 
> 
> Yes please, create a JIRA and add any test cases you have. For the docu, sync up with Vladimir since he was going to do some modifications there as well.

Yes, the current design is different than 4.0. Treating all cache elements (be it mortal or immortal) as eviction candidates is the only thing that makes sense. Assign the created JIRA to me. Fix needs to be on all branches, except 4.0.

Regards,
Vladimir 


More information about the infinispan-dev mailing list