[
https://issues.apache.org/jira/browse/FLINK-9231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16527672#comment-16527672
]
Andrey Zagrebin edited comment on FLINK-9231 at 6/29/18 2:16 PM:
-----------------------------------------------------------------
[~triones], [[email protected]]
I would suggest to rebase PR on the actual master and update
`RestServerEndpoint` the same way as `WebFrontendBootstrap` in PR.
Still I think, before we change socket options in web servers, it would be
useful to have here in issue description/comments details about the use case
where this is a problem. Is it some kind of frequent restart of web server
locally?
was (Author: azagrebin):
[~triones], [[email protected]]
I would suggest to rebase on actual master and update `RestServerEndpoint` the
same as `WebFrontendBootstrap` in PR.
Still I think, before we change socket options in web servers, it would be
useful to have here in issue description/comments details about the use case
where this is a problem. Is it some kind of frequent restart of web server
locally?
> Enable SO_REUSEADDR on listen sockets
> -------------------------------------
>
> Key: FLINK-9231
> URL: https://issues.apache.org/jira/browse/FLINK-9231
> Project: Flink
> Issue Type: Improvement
> Components: Webfrontend
> Reporter: Ted Yu
> Assignee: Triones Deng
> Priority: Major
> Labels: pull-request-available
>
> This allows sockets to be bound even if there are sockets from a previous
> application that are still pending closure.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
