Boost C++ Libraries: Ticket #5822: asyncconnect generates multiple sockets https://svn.boost.org/trac10/ticket/5822 <p> I already posted this on the mailing list, but since no-one replied and said this is a feature I assume it is a bug. </p> <p> Start the chatserver example. Run <a class="missing wiki">TcpView</a> (from sysinternals). Look at the sockets generated by the example. As expected the server has created 1 socket and its status is "listening". Ok so far. Next start the chat client example. It generates 3 sockets! One is expected: it targets the port the server is listening on. But there are 2 others in "established" state. They target each other. Why are they there and what are they doing? NOTE: ofc the server generated a new "established" socket as well. </p> en-us Boost C++ Libraries /htdocs/site/boost.png https://svn.boost.org/trac10/ticket/5822 Trac 1.4.3 Olaf van der Spek <olafvdspek@…> Sat, 01 Oct 2011 12:36:25 GMT <link>https://svn.boost.org/trac10/ticket/5822#comment:1 </link> <guid isPermaLink="false">https://svn.boost.org/trac10/ticket/5822#comment:1</guid> <description> <p> AFAIK it's for internal usage. I think it's to support cancellation. </p> </description> <category>Ticket</category> </item> <item> <dc:creator>chris_kohlhoff</dc:creator> <pubDate>Tue, 29 May 2012 01:50:48 GMT</pubDate> <title>status changed; resolution set https://svn.boost.org/trac10/ticket/5822#comment:2 https://svn.boost.org/trac10/ticket/5822#comment:2 <ul> <li><strong>status</strong> <span class="trac-field-old">new</span> → <span class="trac-field-new">closed</span> </li> <li><strong>resolution</strong> → <span class="trac-field-new">invalid</span> </li> </ul> <p> These sockets are used by the async connect's select()-based implementation. It's an example of the "self pipe trick". </p> Ticket weiyuemin@… Fri, 14 Jun 2013 05:17:22 GMT <link>https://svn.boost.org/trac10/ticket/5822#comment:3 </link> <guid isPermaLink="false">https://svn.boost.org/trac10/ticket/5822#comment:3</guid> <description> <p> Take the liberty to ask: isn't asio using iocp-based implementation on Windows ? </p> <p> We are enduring problems because of the two unexpected sockets which target each other. Some firewalls / anti-virus softwares also suspected this. </p> </description> <category>Ticket</category> </item> </channel> </rss>