September 2006 - jboss-dev-forums

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Graphic Installer for Messaging

by alex.pinkin＠jboss.com

btw this task is not as simple as adding a Messaging/JBossMQ radio button to the installer. We need to have full support for mutually exlcusive packs. izpack version that we are using doesn't support this. It's supposed to be addressed in the upcoming version of the izpack due in a couple of months. It's possible to patch the version of izpack that we are using of course. I'll see if we can bump up the priority of this task, but so far we have a few large tasks on the installer which have higher priority than this. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3969011#3969011 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3969011

18 years, 2 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: JBMESSAGING-410 - Use of JmsXA in non transactional envi

by ovidiu.feodorov＠jboss.com

Alright, after 30+ messages, the conclusion is: we'll do it. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3969010#3969010 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3969010

18 years, 2 months

1
0
0 / 0

[Design of JBoss/Tomcat Integration] - Re: Modifying session's jvmRoute when cookies are disabled

by bstansberry＠jboss.com

Sorry, in the above example context.xml, it should say 'cookies="false"'. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968998#3968998 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968998

18 years, 2 months

1
0
0 / 0

[Design of JBoss/Tomcat Integration] - Re: Modifying session's jvmRoute when cookies are disabled

by bstansberry＠jboss.com

http://labs.jboss.com/portal/jbossas/download You don't want 4.0.4.CR2, you want 4.0.4.GA, which is the final release. I do think you're grasping at straws though, as I can't see how my change would affect the behavior of response.encodeURL(). :( Ahh, something coming back from buried memory... If you really want to prevent the use of cookies (as opposed to supporting users with them turned off), you want to make sure you have a context.xml file in WEB-INF that looks like this: [url]   org.jboss.web.tomcat.security.RunAsListener [/url] The relevant part is 'cookies="false"'; the rest are just the defaults from jbossweb-tomcat55.sar/context.xml. IIRC, if 'cookies="true"' and the browser has presented a session cookie, encodeURL() knows this and skips the unnecessary session id encoding. Perhaps this is your problem. You could also edit the jbossweb-tomcat55.sar/context.xml file itself and turn cookies off by default for the entire server. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968996#3968996 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968996

18 years, 2 months

1
0
0 / 0

[Design of JBoss/Tomcat Integration] - Re: Modifying session's jvmRoute when cookies are disabled

by timward＠ddisystems.com

Where do I go to download 4.0.4CR2 release? I am having trouble with URL rewriting (it is a requirement to turn cookies off in our application). Passing the outgoing URL through response.encodeURL() does not seem to be working (It leaves the URL unchanged). I know I am grasping at straws, but I thought the fix that Brian made for JBAS-3015 might make a difference. Thanks. Tim View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968993#3968993 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968993

18 years, 2 months

1
0
0 / 0

[Design of JBoss jBPM] - Re: tx configs and command interface for web app

by tom.baeyens＠jboss.com

the problem is that with a command interface, i don't see a way to prevent many small transactions during the rendering phase. doing it with commands, each backing bean that is accessed during rendering of a page would result in a separate transaction. i don't see a way around that with jsf and ejb cmt. unless we use JTA directly (through the UserTransaction) we can't make sure that the rendering is done in one transaction. but then all the commands are executed in the same transaction. which conflicts with the idea of the each command being executed in one transaction... using the UserTransaction, we can get the exact same behaviour in enterprise as in standard edition. namely one transaction for the user command (if there is one) and one transactino for the rendering phase. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968986#3968986 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968986

18 years, 2 months

1
0
0 / 0

[Design of JCA on JBoss] - Re: Quartz, timer unification, message inflow RAR

by tom.baeyens＠jboss.com

here's the main reasons why i opted to revise our own implementation over quartz: http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3968946#3968946 note that our own implementation is only targetted at the standard java platform. we have a pluggable interface for async messages and timers. for the enterprise platform, we will be supplying a JMS and EJB Timer based bridges as first priorities. maybe later if it turns out to be an advantage, we could also build a bridge to the jboss internal timer SPI as it is in the making if i understand correctly. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968947#3968947 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968947

18 years, 2 months

1
0
0 / 0

[Design of JBoss jBPM] - Re: to quartz or not to quartz

by tom.baeyens＠jboss.com

There were 2 main reasons why i opted to improve our own implementation over Quartz : 1) a very heavy integration layer was needed between quartz and jbpm. especially making sure that quartz was using our jdbc connection for user job operations required some tricks. apart from that, there was more jdbc connection management going on in quartz which was hard to tie to our hibernate integration. 2) quartz can't acquire job-locks be means of optimistic locking. only pessimistic locking is supported and that is not supported by all databases (i.e. hsqldb). the solution that we now have worked out can acquire exclusive locks on a job by using hibernate's optimistic locking. this should even work with isolation level 0 (READ_UNCOMMITTED), which is the only level supported by hsqldb View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968946#3968946 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968946

18 years, 2 months

1
0
0 / 0

[Design of JBoss jBPM] - checked in new job scheduler

by tom.baeyens＠jboss.com

After some investigations on how to use quartz, we ended up improving our own scheduler and messaging. Both the messaging and scheduler service were kept. in any case, these services will produce jobs in the JBPM_JOB table (tables JBPM_TIMER and JBPM_MESSAGE were removed) Both async messages and timers will end up as jobs in the job queue table. There will be one JobExecutor. One job executor can have multiple threads that are all competing for jobs. There is a locking mechanism in place that should quard against duplicate executions. Also besides the async="true", you now have an async="exclusive". this means that the job executor threads will execute all exclusive asynchronous jobs in a synchronized fashion. this might be handy when e.g. multiple concurrent paths of executions are created and in which all of these paths start with async behaviour. in that case, many jobs are offered to the job executor after one transaction. This might lead to contention since many job executor threads might start working on different jobs of the same process instance. That might lead to contention cause you'll get many optimistic locking exceptions. There are still some exceptions with respect to messages and timers. But the bulk of the test suite is running. But i already want to commit for backup reasons. I tried to make sure that other development is not impacted. Let me know if this is not the case and if you are experiencing problems. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968942#3968942 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968942

18 years, 2 months

1
0
0 / 0

[Design of JCA on JBoss] - Re: XAResourceWrapper for WebSphereMQ XA (JBAS-3183)

by weston.price＠jboss.com

anonymous wrote : | P.S. Since Oracle fixed the problem in their latest releases | this feature is only useful anyway for legacy Oracle. | Well, if you assume everything prior to release 10.2 to be 'legacy' then yeah, ok :-) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3968930#3968930 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3968930

18 years, 2 months

1
0
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jboss-dev-forums September 2006