Mailing-List: contact dev-help@activemq.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@activemq.apache.org
Message-ID: <1863616131.1261369373856.JavaMail.jira@brutus>
Date: Sun, 20 Dec 2009 20:22:53 -0800 (PST)
From: "Sebastien Rodriguez (JIRA)" <jira+amqnet@apache.org>
To: dev@activemq.apache.org
Subject: [jira] Updated: (AMQNET-212) Add an Inactivity Monitor to the
 Transports layer to timeout broken connections
In-Reply-To: <1068276749.1258983712696.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/activemq/browse/AMQNET-212?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sebastien Rodriguez updated AMQNET-212:
---------------------------------------

    Attachment: InactivityMonitor.diff

Initial implementation of the Inactivity monitor, based on the logic introduced in the java an cpp version.
We still need to create the related unit tests. Tested with network of brokers in failover mode over network, unplugging the computer with the broker im connected to works.
This should solve the issue https://issues.apache.org/activemq/browse/AMQNET-211

> Add an Inactivity Monitor to the Transports layer to timeout broken connections
> -------------------------------------------------------------------------------
>
>                 Key: AMQNET-212
>                 URL: https://issues.apache.org/activemq/browse/AMQNET-212
>             Project: ActiveMQ .Net
>          Issue Type: Improvement
>          Components: ActiveMQ
>    Affects Versions: 1.2.0
>            Reporter: Timothy Bish
>            Assignee: Timothy Bish
>             Fix For: 1.2.0
>
>         Attachments: InactivityMonitor.diff
>
>
> In order to more quickly detect broken connections between client and broker the NMS.AcitveMQ client needs to add an InactivityMonitor transport filter that deals with the keep alive messages that allow for detection of broken sockets.  Right now a broken connection such as one that occurs when a cable is pulled isn't detected and the client can sit idle instead of failing over if there's no outbound messages to trigger a failed write.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.