Mailing-List: contact dev-help@directory.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Apache Directory Developers List" <dev@directory.apache.org>
Message-ID: <1302052122.1133166940310.JavaMail.jira@ajax.apache.org>
Date: Mon, 28 Nov 2005 09:35:40 +0100 (CET)
From: "Trustin Lee (JIRA)" <jira@apache.org>
To: dev@directory.apache.org
Subject: [jira] Resolved: (DIRMINA-129) "Pipe broken" IOExceptions countered
 on non-active stream connections
In-Reply-To: <2095603591.1132907276156.JavaMail.jira@ajax.apache.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit

     [ http://issues.apache.org/jira/browse/DIRMINA-129?page=all ]
     
Trustin Lee resolved DIRMINA-129:
---------------------------------

    Resolution: Fixed

I removed the use of PipedInput(Output)Stream in StreamIoHandler.  Now, a brand-new and high-performing implementation replaces it.  I tested a little bit with HTTP server example, and it seems to work great.  Please confirm my fix and close this issue.

Thanks,
Trustin

> "Pipe broken" IOExceptions countered on non-active stream connections
> ---------------------------------------------------------------------
>
>          Key: DIRMINA-129
>          URL: http://issues.apache.org/jira/browse/DIRMINA-129
>      Project: Directory MINA
>         Type: Bug
>     Versions: 0.8.1
>  Environment: All
>     Reporter: dave irving
>     Assignee: Trustin Lee
>     Priority: Minor
>      Fix For: 0.8.2

>
> The StreamIOHandler implementation uses PipedInput/Output streams for providing read data.
> PipedInputStream is quite strict about the relationship between the producer and consumer thread.
> When blocking on a read, PipedInputStream checks every second to see if the last producer thread to write to it is still alive. If it detects that the producer thread is no longer alive, an IO Exception is thrown indicating "Pipe Broken".
> This causes a problem in Mina becuase worker threads in the ThreadPoolFilter have a "keep alive time" (default is one second). So, if a user is blocked on a stream IO read (really a PipedInputStream), and the last ThreadPoolFilter thread which produced data for the stream idles out, the user encounters a "Pipe Broken" IOException (not related to session idle).
> Users can currently work around the problem by increasing the keep-alive time on the ThreadPoolFilter.
> My gut feeling is that ditching PipedInputStream / PipedOutputStream in favour of something more suited to Minas needs would be a good solution to the problem.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira