Mailing-List: contact issues-help@mesos.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@mesos.apache.org
Date: Tue, 29 Nov 2016 21:53:58 +0000 (UTC)
From: "Benjamin Hindman (JIRA)" <jira@apache.org>
To: issues@mesos.apache.org
Message-ID: <JIRA.12980831.1466425737000.394305.1480456438360@Atlassian.JIRA>
In-Reply-To: <JIRA.12980831.1466425737000@Atlassian.JIRA>
References: <JIRA.12980831.1466425737000@Atlassian.JIRA> <JIRA.12980831.1466425737752@arcas>
Subject: [jira] [Commented] (MESOS-5658) Socket::shutdown() is inconsistent
 with BSD sockets API
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Tue, 29 Nov 2016 21:54:00 -0000


    [ https://issues.apache.org/jira/browse/MESOS-5658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15706651#comment-15706651 ] 

Benjamin Hindman commented on MESOS-5658:
-----------------------------------------

Partially completed in:

{code}
commit 4417a4e8917e17e70942a77bd796857978778888
Author: Benjamin Hindman <benjamin.hindman@gmail.com>
Date:   Sun Nov 27 12:10:14 2016 -0800

    Added support for specifying how a socket should be shutdown.
    
    Review: https://reviews.apache.org/r/54113

commit b3013ffb0711ad70f91b50e90ea17674137c9c15
Author: Greg Mann <greg@mesosphere.io>
Date:   Tue Nov 29 12:18:14 2016 -0800

    Added POSIX socket shutdown types to Windows header.
    
    This patch adds two missing BSD socket shutdown types
    to a Windows header which maps these POSIX constants
    to their Windows counterparts.
    
    Review: https://reviews.apache.org/r/53990/
{code}

Still need to remove the default.

> Socket::shutdown() is inconsistent with BSD sockets API
> -------------------------------------------------------
>
>                 Key: MESOS-5658
>                 URL: https://issues.apache.org/jira/browse/MESOS-5658
>             Project: Mesos
>          Issue Type: Improvement
>          Components: libprocess
>            Reporter: Neil Conway
>            Priority: Minor
>
> In libprocess, the {{Socket::shutdown()}} member function is inconsistent with the {{shutdown(2)}} function from the Berkeley sockets API: the former doesn't take any parameters (and implicitly *only* shuts down the receiver-side of the socket), whereas the latter takes a parameter that controls which side(s) are shutdown.
> IMO we should either make these behave the same or document why they are different.


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)