[ 
https://issues.apache.org/jira/browse/IGNITE-10651?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Vyacheslav Koptilin updated IGNITE-10651:
-----------------------------------------
    Description: 
Under some circumstances, 
{{GridCachePartitionExchangeManager.affinityReadyFuture}} returns 
{{GridFinishedFuture}} or 
{{GridCachePartitionExchangeManager.AffinityReadyFuture}}
 Obviously, the returned future cannot be used in order to validate a cache. It 
seems that this behavior can be improved if {{affinityReadyFuture()}} returns 
an instance of {{GridDhtTopologyFuture,}} and therefore, it will simplify the 
logic related to cache validation.

Perhaps, {{ExchangeFutureSet}} can be changed to {{ExchangeFutureMap}} that 
should contain a mapping of topology version to the corresponding 
{{GridDhtTopologyFuture}}.

  was:
Under some circumstances, 
{{GridCachePartitionExchangeManager.affinityReadyFuture()}}returns 
{{GridFinishedFuture}} or 
{{GridCachePartitionExchangeManager.AffinityReadyFuture}}
Obviously, the returned future cannot be used in order to validate a cache. It 
seems that this behavior can be improved if {{affinityReadyFuture()}} returns 
an instance of {{GridDhtTopologyFuture,}} and therefore, it will simplify the 
logic related to cache validation.

Perhaps, {{ExchangeFutureSet}} can be changed to {{ExchangeFutureMap}} that 
should contain a mapping of topology version to the corresponding 
{{GridDhtTopologyFuture}}.


> GridCachePartitionExchangeManager.affinityReadyFuture() should return 
> GridDhtTopologyFuture instead of IgniteInternalFuture
> ---------------------------------------------------------------------------------------------------------------------------
>
>                 Key: IGNITE-10651
>                 URL: https://issues.apache.org/jira/browse/IGNITE-10651
>             Project: Ignite
>          Issue Type: Improvement
>    Affects Versions: 2.7
>            Reporter: Vyacheslav Koptilin
>            Priority: Minor
>
> Under some circumstances, 
> {{GridCachePartitionExchangeManager.affinityReadyFuture}} returns 
> {{GridFinishedFuture}} or 
> {{GridCachePartitionExchangeManager.AffinityReadyFuture}}
>  Obviously, the returned future cannot be used in order to validate a cache. 
> It seems that this behavior can be improved if {{affinityReadyFuture()}} 
> returns an instance of {{GridDhtTopologyFuture,}} and therefore, it will 
> simplify the logic related to cache validation.
> Perhaps, {{ExchangeFutureSet}} can be changed to {{ExchangeFutureMap}} that 
> should contain a mapping of topology version to the corresponding 
> {{GridDhtTopologyFuture}}.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to