[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Incorporating Remoting http transport into Messaging
by ron_sigal
"tim_fox" wrote : ... when you close a consumer, closing() first gets called on the ServerConsumerEndpoint, this causes stop() to be called, which waits for any deliveries of messages to the client which might be in progress to complete first.
|
| Once that is done, the actual client consumer is closed and removed from the callback manager.
Looking a the code again, I noticed something I missed the first time around. When MessageCallbackHandler gets a message, it does one of two things:
1. If a thread is blocked in receive(), it wakes up the thread and returns.
2. If a MessageListener is registered, it queues up a Runnable to call onMessage().
Now, 1 is irrelevant if the MessageConsumer has called close(), so I thought the point was to make sure that a MessageConsumer didn't close while a listener was in onMessage(). But it looks like that could still happen. Is that ok? Just curious.
View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3974707#3974707
Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3974707
19 years, 6 months
[Design of JBoss ESB] - Re: PropertyManager
by mark.little@jboss.com
Here's an updated config file:
| <?xml version="1.0" encoding="ISO-8859-1"?>
| <!--
| JBoss, Home of Professional Open Source
| Copyright 2006, JBoss Inc., and others contributors as indicated
| by the @authors tag. All rights reserved.
| See the copyright.txt in the distribution for a
| full listing of individual contributors.
| This copyrighted material is made available to anyone wishing to use,
| modify, copy, or redistribute it subject to the terms and conditions
| of the GNU Lesser General Public License, v. 2.1.
| This program is distributed in the hope that it will be useful, but WITHOUT A
| WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
| PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
| You should have received a copy of the GNU Lesser General Public License,
| v.2.1 along with this distribution; if not, write to the Free Software
| Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
| MA 02110-1301, USA.
|
| (C) 2005-2006,
| @author JBoss Inc.
| -->
| <!-- $Id: default-jbossesb-properties.xml $ -->
| <!--
| These options are described in the JBossESB manual.
| Defaults are provided here for convenience only.
|
| Please read through this file prior to using the system, and consider
| updating the specified entries.
| -->
| <esb
| xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
| xsi:noNamespaceSchemaLocation="jbossesb-1_0.xsd">
| <properties name="core">
| <property name="org.jboss.soa.esb.jndi.server.type" value="jboss"/>
| <property name="org.jboss.soa.esb.jndi.server.url" value="localhost"/>
| <!-- property name="org.jboss.soa.esb.paramsRepository.class" value="" -->
| <!-- property name="org.jboss.soa.esb.encryption.factory.class" value="" -->
| <!-- property name="org.jboss.soa.esb.objStore.configFile" value="" -->
| </properties>
| <properties name="transports" depends="core">
| <property name="org.jboss.soa.esb.mail.smtp.host" value="localhost"/>
| <property name="org.jboss.soa.esb.mail.smtp.user" value=""/>
| <property name="org.jboss.soa.esb.mail.smtp.password" value=""/>
| <property name="org.jboss.soa.esb.mail.smtp.port" value="25"/>
| <!-- property name="org.jboss.soa.esb.mail.smtp.from" value="" -->
| <!-- property name="org.jboss.soa.esb.mail.smtp.auth" value="" -->
| </properties>
| <properties name="connection">
| <!-- property name="isolation" value="" -->
| <!-- property name="connection-url" value="" -->
| <property name="min-pool-size" value="5"/>
| <property name="max-pool=size" value="10"/>
| <property name="blocking-timeout-millis" value="5000"/>
| <!-- property name="connection-pool-class" value="" -->
| <!-- property name="user-name" value="" -->
| <!-- property name="password" value="" -->
| <!-- property name"password-decrypter" value="" -->
| <property name="abandoned-connection-timeout" value="10000"/>
| <property name="abandoned-connection-time-interval" value="30000"/>
| </properties>
| </esb>
|
I've broken the properties into different modules, with associated dependencies. The way the PropertyManager works means that it builds up a list of dependant modules in a directed acyclic graph. This means that there is a root, which can see every property below it, but nodes below cannot see properties above them: there is a level of isolation.
When creating PropertyManagers, you need to specify where in the graph it is to "tie in". This obviously has an affect on what that instance can then see or do: if it goes in at the wrong level, it won't be able to see properties it may need (unless they were defined to be "system" properties, and in which case they are globally accessible.)
This means that we now have isolated PropertyManagers for each module. It helps the management side of things a lot, but does mean that we have to consider where in the graph a property needs to be, or which PropertyManager we want to use to try to access it.
What we have at the moment in terms of properties is pretty basic, so this 3 module approach seems about right to me. I'll check the code in now (along with some tests!) and we can discuss the structuring tomorrow. It's fairly trivial to change.
View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3974702#3974702
Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3974702
19 years, 6 months
[Design of JCA on JBoss] - JBAS-1429 ReAuthentication Support
by weston.price@jboss.com
I have been working on this, and am at the point where I need to start testing...that's where the problem comes in :-)
Hypersonic doesn't support changing the user identity of a connection and since this is our default, using HyperSonic for this isn't really adequate. While I can mock something up to test the mechanisms which the underlying identity changes, we need some specific JCBC vendor testsuite to really test the conditions against valid RDMBS installations.
This would actually be useful for a few things
ValidConnectionChecker
ExceptionSorter
StaleConnectionChecker (for 4.0.6)
Further, we really need something like this for our JMS integration strategies with varying providers (notably WebSphereMQ).
Have we ever done, thought about, anything like this beyond the installations in the labs for specific support issues?
View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3974686#3974686
Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3974686
19 years, 6 months