Mailing-List: contact dev-help@activemq.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@activemq.apache.org
Date: Wed, 5 Jun 2013 14:48:24 +0000 (UTC)
From: "Karen Morrissey (JIRA)" <jira@apache.org>
To: dev@activemq.apache.org
Message-ID: <JIRA.12650892.1370357845833.71797.1370443704729@arcas>
In-Reply-To: <JIRA.12650892.1370357845833@arcas>
References: <JIRA.12650892.1370357845833@arcas>
Subject: [jira] [Commented] (AMQ-4569) Stomp inactivity timeout takes
 significantly longer than specified
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/AMQ-4569?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13675980#comment-13675980 ] 

Karen Morrissey commented on AMQ-4569:
--------------------------------------

Is 3 times the heartbeat period the upper bound for when an inactive client connection will be cleared by the broker? If not, what is the upper bound?

If there was a statement I could measure my observations against, such as that connection clearing would happen no sooner than 1 heartbeat and no later than X heartbeats (or one heartbeat + X seconds/minutes/hours), we could probably settle this hear and prevent this issue from popping up again.
                
> Stomp inactivity timeout takes significantly longer than specified
> ------------------------------------------------------------------
>
>                 Key: AMQ-4569
>                 URL: https://issues.apache.org/jira/browse/AMQ-4569
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: stomp
>    Affects Versions: 5.8.0
>         Environment: Windows 7, JRE 7.
>            Reporter: Karen Morrissey
>            Priority: Minor
>              Labels: heartbeat, inactivity, stomp
>         Attachments: activemq.log, activemq.log, activemq.log
>
>
> When I specify a timeout using a connector like the following to get a 5-minute idle timeout
> {code:xml}
> <transportConnector name="stomp" uri="stomp://0.0.0.0:61613?transport.defaultHeartBeat=300000,0"/>
> {code}
> It takes 15 minutes for it to time out. I tried one previously at 900000 (15 minutes), and it had not timed out after 35 minutes, when I had to leave.
> So, it appears the timeout takes 3 times longer than specified. I find it suspicious that there is a call somewhere I saw in the code that divides the timeout period by 3.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira