December 2007 - jboss-dev-forums - Jboss List Archives

[Design of JBoss Build System] - Re: AS5 / Java6 build

by dimitris＠jboss.org

I saw an org.jboss.resource.adapter.jdbc.jdk6 package introduced, that needs jdk6 to compile. How are we supposed to build/package the distro so that the extra jdk6 classes are included? a) Have different jdk5 / jdk6 distros ? b) Compile with jdk6, using -source 1.5 -target 1.5, expect for the jdk6 classes that should be compiled with -source 1.6 -target 1.6 ? c) other? Also the current jdk6 hudson run fails while testing the profileservice: http://hudson.qa.jboss.com/hudson/view/JBoss%20AS/job/JBoss-AS-5.0.x-Test... Do we need to set the "sun.lang.ClassLoader.allowArraySyntax" property under jdk6? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109660#4109660 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109660

17 years

1
0
0 / 0

[Design of JBoss Build System] - Re: AS5 / Java6 build

by dimitris＠jboss.org

Related jira issue http://jira.jboss.com/jira/browse/JBAS-5031 View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109659#4109659 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109659

17 years

1
0
0 / 0

[Design of JBoss jBPM] - Re: Improvig timer

by brittm

Ronald, I like it. So then we would have a duedate attribute with one of the following: 1) EL expression 2) jBPM calendar duration with an optional operator (+/-) 3) or both separated by a +/- operator Do you require the EL to evaluate to a java Date, java Calendar, String, or something else? Perhaps handling all three would be the right option. -Britt View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109658#4109658 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109658

17 years

1
0
0 / 0

[Design of JBoss Build System] - AS5 / Java6 build

by dimitris＠jboss.org

Copying here from the jboss-dev list. Adrian wrote:anonymous wrote : | I've got JBoss Head compiling with Java6. | | I managed to avoid using a preprocessor for the JDBC rar | but only by using reflection to decide which | version of the code is in the classpath, i.e. JDK5/JDK6 | | The reflection is only done once. but | there is more duplicate code that there would otherwise be. | | e.g. WrappedPreparedStatementJDK6 can't extend | both WrappedPreparedStatement and WrappedStatementJDK6 :-) | | I haven't implemented the Wrapper interface fully. | This will need a revisit where it needs to "recursively" | check whether the object we wrap is itself a wrapper. | | I also "commented out" the DataSourceInterceptor in varia. | I didn't convert it to JDK6. | | As far as I can tell, this is not included in the distribution, | not even in docs/examples/varia. | Alex you wrote this what is happening with this? | Alexey wrote: anonymous wrote : | The DataSourceInterceptor and other classes in that package were developed when I was working on the SPECjAppServer benchmark a few years ago. The goal was to collect statistics about SQL statements executed in a test run and per transaction. Which allowed to tune the CMP engine. | Here is a wiki page that describes how to use them http://wiki.jboss.org/wiki/Wiki.jsp?page=StatisticsCollector | View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109656#4109656 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109656

17 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: JBM 2.0 JDBCPersistenceManager - volunteers?

by alesj

"timfox" wrote : | The JDBCPersistenceManager implementation should use HIbernate (at least the dialects) Wouldn't straight JPA be better? btw: what's the main persistence mechanism? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109651#4109651 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109651

17 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: JBM 2.0 JDBCPersistenceManager - volunteers?

by timfox

Also you'll need a transaction table, for tracking XA transactions (same as current JBM_TX) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109615#4109615 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109615

17 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: JBM 2.0 Testing strategy

by timfox

Yes, for JBM2.0 I want us to be the model project w.r.t testing. We can be shown as an example of how to test code properly. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109614#4109614 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109614

17 years

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: ConcurrentModificationException / Classloading / Deploym

by ALRubinger

That one'll do 'er. I've updated the example test to include some logic which includes a wait (as this is an async call), which doesn't kill the race condition, but is enough to ensure the hare beats the tortoise 99% of the time. Don't run it twice though without server restart, may have uncovered another classloading issue on uninstall/install? ERROR [AbstractKernelController] Error installing to Real: name=vfsfile:/home/alrubinger/business/jboss/wc/jbossas/trunk/ejb3/output/test-lib/ejbthree1123.jar state=PostClassLoader mode=Manual requiredState=Real | org.jboss.deployers.spi.DeploymentException: Error deploying ejbthree1123.jar: Error creating ejb container TestStatelessBean: java.lang.ClassNotFoundException: Invalid use of destroyed classloader, UCL destroyed at: | at org.jboss.ejb3.deployers.Ejb3Deployer.deploy(Ejb3Deployer.java:158) S, ALR View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109606#4109606 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109606

17 years

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: ConcurrentModificationException / Classloading / Deploym

by scott.stark＠jboss.org

Its the TreeSets in the ConcurrentReaderHashMap that are being used and seeing the modifications. I have further updated the pkgSet copy on write updates for removal of class loaders and no longer seen any ConcurrentModificationException when running the test. The deployment is being torn down while the jms endpoint is still starting up so there are still race condition problems with the test. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109605#4109605 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109605

17 years

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: ConcurrentModificationException / Classloading / Deploym

by anil.saldhana＠jboss.com

Doesn't CopyOnWriteArrayList avoid ConcurrentModificationException by design? http://java.sun.com/j2se/1.5.0/docs/api/java/util/concurrent/CopyOnWriteA... It should. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4109603#4109603 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4109603

17 years

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 December 2007