Mailing-List: contact users-help@activemq.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@activemq.apache.org
Received-SPF: pass (athena.apache.org: domain of s.d.c.martes@gmail.com
 designates 74.125.82.171 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        b=Zrd1tNdbHLGRYYwyRxkCFTAyK9lcJurDhV6ERoMrpMvVKXU+OCxp9xc7I1qdVGRZ84
         M5grimHPKdfwrgFnywOSR8QRNGwEvqQTu6dH6dc9SU9nwizscDAnYlq8AOTElxFpbU16
         curWreq1SywzHVJkI6EldbcxE5I36KPbOhLgo=
MIME-Version: 1.0
In-Reply-To: <AANLkTikyZdQ63TFBdPvQgFhmmdFWYWa6zfWW1XjK1igR@mail.gmail.com>
References: <AANLkTikWoikX_UgR0MYTs4DeDAahC0q_drvYO4WkYWU-@mail.gmail.com>
	<AANLkTikyZdQ63TFBdPvQgFhmmdFWYWa6zfWW1XjK1igR@mail.gmail.com>
Date: Tue, 24 Aug 2010 10:01:18 -0400
Message-ID: <AANLkTim+akVdN+zOUMmenspaYdW7NXCpe28_01RauGMQ@mail.gmail.com>
Subject: Re: Multiple sockets connection TIME_WAIT
From: Shurbann Martes <s.d.c.martes@gmail.com>
To: users@activemq.apache.org
Content-Type: multipart/alternative; boundary=0016367fad71fe00ef048e9232e7

--0016367fad71fe00ef048e9232e7
Content-Type: text/plain; charset=UTF-8

You're right. I was using for the jmsTransactionManager
the org.apache.activemq.ActiveMQConnectionFactory instead of
the org.apache.activemq.pool.PooledConnectionFactory.

I will test this later. Thank you for your quick response.

Regards,
Shurbann Martes

On Tue, Aug 24, 2010 at 9:42 AM, Gary Tully <gary.tully@gmail.com> wrote:

> You need to look into the cache level, think the spring default is to
> cache nothing if there is a transaction manager present as there is in
> your case.
>
> You may want to try cache_connection at a minimum.
>
> There are OS TCP level config options to configure the TIME_WAIT
> window but it may be better to tackle the excessive connection
> creation a the jms application level.
>
> On 24 August 2010 11:56, Shurbann Martes <s.d.c.martes@gmail.com> wrote:
> > Hi ActiveMQ,
> >
> > Thank you for taking time to answer my question.
> >
> > I have the following issue. I'm using ActiveMQ in combination with
> Spring:
> >
> > <bean
> > class="org.springframework.jms.listener.DefaultMessageListenerContainer">
> > <property name="connectionFactory" ref="pooledFactory" />
> > <property name="destination" ref="requestQueue" />
> > <property name="messageListener" ref="productRequestProcessorMdp" />
> > <property name="concurrentConsumers" value="1" />
> > <property name="maxConcurrentConsumers" value="5" />
> > <property name="transactionManager" ref="jmsTransactionManager" />
> > </bean>
> >
> > It's using the TCP Transport: tcp://localhost:61616
> >
> > This was working fine, until an increase in usage of the service lately:
> >
> > I noticed lately that the amount of socket connection connecting on the
> port
> > 61616 is huge (> 780) of which only 15 is ESTABLISHED and the other are
> > TIME_WAIT.
> >
> > Lately my container/machine starts giving too many socket connections
> during
> > peek hours. (I've taken care of ulimit -n on OS level). Now I'm starting
> to
> > look at the connections of this machine and more then 80% is on the
> account
> > ActiveMQ.
> >
> > My questions are
> > - Anyone have/had this problem?
> > - Is there a way of putting a max connection to the broker?
> > - How can I make the TIME_WAIT interval smaller?
> > - Any other suggestion of decreasing the amount of connections to the
> > broker?
> >
> > Thank you in advance.
> >
> > Regards,
> > SM
> >
>
>
>
> --
> http://blog.garytully.com
>
> Open Source Integration
> http://fusesource.com
>

--0016367fad71fe00ef048e9232e7--