[
https://issues.apache.org/jira/browse/IGNITE-1032?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14638447#comment-14638447
]
Denis Magda commented on IGNITE-1032:
-------------------------------------
The solution is one that is suggested by Yakov: trySendMessageDirectly is
called from threads pool avoiding blocking ring message worker thread.
Reviewed by Yakov, merged to the main development branch.
> discovery:
> discovery.tcp.ServerImpl.RingMessageWorker#processStatusCheckMessage should
> not block
> ------------------------------------------------------------------------------------------------
>
> Key: IGNITE-1032
> URL: https://issues.apache.org/jira/browse/IGNITE-1032
> Project: Ignite
> Issue Type: Improvement
> Reporter: Yakov Zhdanov
> Priority: Blocker
> Fix For: sprint-7
>
>
> It may be blocked on
> {{trySendMessageDirectly(msg.creatorNode(), msg);}}
> which should be moved to another thread or reapproached to NIO fashion.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
