What's the issue here ?
You return true from RspFilter.needMoreResponses(), but you don't *have*
more responses available ?
Galder Zamarreno wrote:
On 07/07/2009 02:03 PM, Manik Surtani wrote:
>
> On 7 Jul 2009, at 08:03, Galder Zamarreno wrote:
>
>> Hi all,
>>
>> Why does ClusteredGetResponseValidityFilter maintain a list of
>> pendingResponders? Wouldn't it be more efficient if
>> needMoreResponses() returned as soon as one positive response has been
>> received regardless of who's pending to return anything? I'm assuming
>> here that SuccessfulResponse means that the clustered get returned
>> what we're trying to get.
>
> IIRC this was an issue where if we hadn't seen any valid responses,
> JGroups would timeout. So needMoreReponses() needed to return true at
> some point, and in this case that point is if we don't have a valid
> response AND there are more pending responders.
>
> This may not be the case anymore with JGroups 2.8 though.
Hmmmmmm, not sure I understand. Vladimir/Bela, can you clarify this?
As a FYI: This came from the stoppgaes I see in DIST mode where some
clustered get responses are somehow not getting through (see
"Sometimes TCP responses not getting through on localhost" titled
email sent to this dev list). While looking at that
ClusteredGetResponseValidityFilter was waiting for one node to respond
even if it had already received a successful response to the clustered
get.
>
>> I implemented a very similar thing for HAJNDI in such way that as soon
>> as a look up had succeed, we wouldn't wait for anyone else.
>>
>> Cheers,
>> --
>> Galder ZamarreƱo
>> Sr. Software Engineer
>> Infinispan, JBoss Cache
>> _______________________________________________
>> infinispan-dev mailing list
>> infinispan-dev(a)lists.jboss.org
>>
https://lists.jboss.org/mailman/listinfo/infinispan-dev
>
> --
> Manik Surtani
> manik(a)jboss.org
> Lead, Infinispan
> Lead, JBoss Cache
>
http://www.infinispan.org
>
http://www.jbosscache.org
>
>
>
>