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

[Design of Clustering on JBoss (Clusters/JBoss)] - Some cluster enhancements...

by ScottMarlowNovell

Hi, I have been working on a few small enhancements, some of them are complete and some need more work. I would like to check in the following changes into some 4.x or 5.x release if we agree that the changes are good: 1. new HAPartition methods for calling a specific cluster targetNode (identified with opaque server reference that is currently a org.jgroups.stack.IpAddress). This was useful for sending messages to a specific target node in the cluster. 2. A load balancing policy based on FirstAvailable that always favors a same machine choice. 3. A load balancing policy that always picks the ha-singleton server (depends on ha-jndi and a server side ha-singleton). The hapartition ehancement is minor and easy. The other changes are additions so they shouldn't break anything. I have a ejb enhancement as well but I'll post about that in ejb design forum. Thoughts? Scott View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101398#4101398 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101398

16 years, 6 months

1
0
0 / 0

[Design of EJB 3.0] - Re: metadata-related NPEs: empty collection vs null

by scott.stark＠jboss.org

Yes. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101395#4101395 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101395

16 years, 6 months

1
0
0 / 0

[Design of EJB 3.0] - Re: metadata-related NPEs: empty collection vs null

by alex.loubyansky＠jboss.com

You mean from the merged view we return empty collections instead of null? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101393#4101393 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101393

16 years, 6 months

1
0
0 / 0

[Design of AOP on JBoss (Aspects/JBoss)] - Hudson setup changes

by ryan.campbell＠jboss.com

The AOP jobs on Hudson have been modified such that jboss-aop-head-testsuite will automatically force the building of the integration test suites such as jboss-aop-jbossas-4.0.3SP1-jdk14 and jboss-aop-jbossas-4.0.3SP1-jdk14. This means that you no longer need to force each integration test job seperately, you only need to force the toplevel jboss-aop-head-testsuite job. So long as the jboss-aop-head-testsuite job does not fail, the other builds will be simultaneously started after the jboss-aop-head-testsuite job completes. This is intended to prevent the needless building of the integration test suites, only to find that the build is broken. When the build is broken, the integration testsuites are needlessly running only to fail once the checkout is complete. This ties up our Hudson nodes and prevents other jobs from running in a timely manner. You can see this presented graphically at: http://hudson.qa.jboss.com/hudson/view/AOP/job/jboss-aop-head-testsuite/ Please use the jboss-aop-testsuite-head to kick off the builds in preference to forcing each integration build seperately. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101392#4101392 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101392

16 years, 6 months

1
0
0 / 0

[Design of EJB 3.0] - Re: metadata-related NPEs: empty collection vs null

by scott.stark＠jboss.org

>From the merged view this is fine. The concern about doing this originally was that invalid objects would result for writing back out, and if you initialize fields to empty non-mutable collections this caused parse errors as jbossxb will use an existing collection from the binding target if one exists. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101388#4101388 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101388

16 years, 6 months

1
0
0 / 0

[Design of JBoss jBPM] - BAM-Queries & Task-Log

by camunda

Hi guys. For the simulation I currently write a Command with some HSQL Queries to get historical data (to propose it as simulation input). Basically I have 2 queries, which look like: | <query name="Simulation.calculateAverageTimeForStates"> | <![CDATA[ | select | nl.node.id, | nl.node.name, | count(nl.leave), | avg(nl.duration), | min(nl.duration), | max(nl.duration), | stddev(nl.duration) | from org.jbpm.graph.log.NodeLog nl | where nl.node.processDefinition = :processDefinition | and nl.node.class = org.jbpm.graph.node.State | and time(nl.enter) > time(:fromTime) and time(nl.enter) < time(:tillTime) | group by nl.node | ]]> | </query> | | <query name="Simulation.calculateAverageTimeForTaskInstances"> | <![CDATA[ | select | assignLog.taskInstance.task.id, | assignLog.taskInstance.task.name, | assignLog.taskInstance.task.taskNode.id, | assignLog.taskInstance.task.taskNode.name, | count(assignLog.date), | avg(endLog.date - assignLog.date), | min(endLog.date - assignLog.date), | max(endLog.date - assignLog.date), | stddev(endLog.date - assignLog.date) | from org.jbpm.taskmgmt.log.TaskEndLog as endLog, | org.jbpm.taskmgmt.log.TaskAssignLog as assignLog | where | endLog.taskInstance = assignLog.taskInstance | and assignLog.taskInstance.task.taskNode.processDefinition = :processDefinition | and time(assignLog.date) > time(:fromTime) and time(assignLog.date) < time(:tillTime) | group by assignLog.taskInstance.task | ]]> | </query> | They do their work. I figured out one problem, which is a bit ugly, but I am not sure how to really solve it. Maybe it makes sense to think about it more in PVM? I just wanted to write it down here... I don't get any information which outgoing transition the user selected while ending tasks. I only get the outgoing transition for the TaskNode. But these are two different pieces of information. For the states I use this HSQL. I now use it for the TaskInstances too, but this is not really correct (but I ignore this for the moment)... | <query name="Simulation.calculateLeavingTransitionProbability"> | <![CDATA[ | select | tl.transition.id, | tl.transition.name, | tl.destinationNode.id, | tl.destinationNode.name, | count(tl) | from org.jbpm.graph.log.TransitionLog tl | where tl.sourceNode.id = :nodeId | group by tl.transition | ]]> | </query> Cheers Bernd View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101387#4101387 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101387

16 years, 6 months

1
0
0 / 0

[Design of EJB 3.0] - metadata-related NPEs: empty collection vs null

by alex.loubyansky＠jboss.com

There is a number of NPEs in EJB3DescriptorHandler currently. The code doesn't expect null as a value of collection-valued metadata properties. I started adding checks for null for now. But it would probably be better to return empty collections? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101385#4101385 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101385

16 years, 6 months

1
0
0 / 0

[Design of JBoss ESB] - Re: NewProcessInstanceFacade bug?

by kadlecp

I have created the JIRA issue. Its address is http://jira.jboss.org/jira/browse/JBESB-1298 View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101246#4101246 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101246

16 years, 6 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: Accessing content of vfsfile url fails

by wolfc

VirtualFileURLConnection.urlCache is a memory leak. http://jira.jboss.com/jira/browse/JBMICROCONT-157 I don't have the object reference picture anymore. But it might be worth to build a new one, so we know where it is actually in use. This might be related to http://jira.jboss.com/jira/browse/JBMICROCONT-179. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101180#4101180 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101180

16 years, 6 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: Adding a new SchemaBindingInitializer

by scott.stark＠jboss.org

Use the org.jboss.xb.binding.sunday.unmarshalling.SchemaResolverConfig bean schemaInitializers property. You could add the following to the server metadata-beans.xml if that is where you need the initializer: | <bean name="SchemaResolverConfig" | class="org.jboss.xb.binding.sunday.unmarshalling.SchemaResolverConfig"> | |  | <property name="bindingClassesByLocations"> | <map keyClass="java.lang.String" valueClass="java.lang.String"> |  | <entry> | <key>application</key> | <value>org.jboss.metadata.ear.spec.Ear14MetaData</value> | </entry> | ... | </property> | </bean> | | <property name="schemaInitializers"> | <map keyClass="java.lang.String" valueClass="java.lang.String"> | <entry> | <key>location</key> | <value>initializer-class</value> | </entry> | </property> | </bean> | | </deployment> | | View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4101106#4101106 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4101106

16 years, 6 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 November 2007