Resent-Date: Fri, 1 Dec 1995 12:05:02 -0500
From: Mike Shaver <shaver@neon.ingenia.com>
Message-Id: <199512011738.MAA02792@neon.ingenia.com>
Subject: Re: sun.net.* classes broken in Linux port of JDK?
To: ak@hplb.hpl.hp.com (Anders Kristensen)
Date: Fri, 1 Dec 1995 12:38:36 -0500 (EST)
Cc: java-linux@java.blackdown.org
In-Reply-To: <199512011026.AA070223568@rabbit.hpl.hp.com> from "Anders Kristensen" at Dec 1, 95 10:26:07 am
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length:       1304
Resent-Message-Id: <"A30N1.0.Fb1.ELplm"@liquor>
Resent-From: java-linux@java.blackdown.org
Precedence: list
Resent-Sender: java-linux-request@java.blackdown.org

Thus spake Anders Kristensen:
> > To test it, save this to the file "demo.java" and do
> > "javac demo.java". Then do "java DatagramServer &" and
> > "java DatagramClient localhost port#" where port# should
> > be the port number DatagramServer announced on startup.
> > 
> > When I run this, I get no errors, but the server never
> > reports having received a datagram and thus never exits.
> 
> Someone else asked a similar question on the pthreads mailing list
> recently, and the answer was that Linux doesn't have working Unix domain
> sockets (yes it socks, I know ;-).  You can use inet domain sockets or
> sys V IPC.

It sounds like the guy _is_ using AF_INET sockets, from the mention of
"localhost", anyway.

And I think the latest kernel releases have near-complete support for
readv/writev, which would be what was missing from AF_UNIX support.
It's quite possible, though, that the libc you're using does not.

Mike

-- 
#> Mike Shaver (shaver@ingenia.com) Ingenia Communications Corporation <#
#>        Technical specialist -- Head geek -- System exorcist         <#
#>                                                                     <#
#>   "Have you considered a life?  I hear they're quite affordable     <#
#>          these days." --- shields@tembel.org                        <#