[
https://issues.apache.org/jira/browse/HTTPCORE-48?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12509310
]
Andrea Selva commented on HTTPCORE-48:
--------------------------------------
Hi, i've some questions about this issue. The handshaking phase is encapsulated
in the bind method inside SSLIOSession class, which is then called by the
connected method in SSLServerIOEventDispatch and SSLClientIOEventDispatch. I
think that when the call to bind return, the caller expect that the IOSession
has done all the bind stuff including ssl handshaking, because this call is
made only once in the the event dispathers specified, i can't understand how an
executor could help to parallelize the job.I'm loosing some details perhaps.
Andrea
> Make SSL IOSession decorator to use an Executor interface to execute all
> potentially blocking handshake tasks
> -------------------------------------------------------------------------------------------------------------
>
> Key: HTTPCORE-48
> URL: https://issues.apache.org/jira/browse/HTTPCORE-48
> Project: HttpComponents Core
> Issue Type: Improvement
> Components: HttpCore NIO
> Reporter: Oleg Kalnichevski
> Fix For: 4.0-beta1
>
>
> Presently the SSL IOSession decorator executes all potentially blocking
> handshake tasks on the I/O thread. Use an Executor interface from
> java.util.concurrent to make possible the execution of handshake tasks using
> worker threads, thus making the I/O thread available for processing I/O
> events even if some SSL connections are blocked pending completion of a
> handshake task.
> Oleg
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
