[
https://issues.apache.org/jira/browse/FLINK-4580?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15509607#comment-15509607
]
ASF GitHub Bot commented on FLINK-4580:
---------------------------------------
Github user StephanEwen commented on the issue:
https://github.com/apache/flink/pull/2473
The idea is nice, but it comes at the cost of yet another message roundtrip
at startup.
Since we should anyways forward the target endpoint invocation failures to
the source gateway, I think that should improve the failure handling in a
similar way.
> Check that the RpcEndpoint supports the specified RpcGateway
> ------------------------------------------------------------
>
> Key: FLINK-4580
> URL: https://issues.apache.org/jira/browse/FLINK-4580
> Project: Flink
> Issue Type: Sub-task
> Components: Distributed Coordination
> Affects Versions: 1.2.0
> Reporter: Till Rohrmann
> Assignee: Till Rohrmann
> Priority: Minor
>
> When calling {{RpcService.connect}} the user specifies the type of the
> {{RpcGateway}}. At the moment, it is not checked whether the {{RpcEndpoint}}
> actually supports the specified {{RpcGateway}}.
> I think it would be good to add a runtime check that the corresponding
> {{RpcEndpoint}} supports the specified {{RpcGateway}}. If not, then we can
> let the connect method fail fast.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
