[JBoss-dev] [ jboss-Patches-611452 ] Fix for tools.jar warning on OS X
Patches item #611452, was opened at 2002-09-18 23:54 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376687&aid=611452&group_id=22866 Category: JBossServer Group: CVS HEAD Status: Open Resolution: None Priority: 5 Submitted By: Kristoff Batizy (kbatizy) Assigned to: Nobody/Anonymous (nobody) Summary: Fix for tools.jar warning on OS X Initial Comment: Mac OS X comes with a full JDK by default with no option for a JRE only. Also, Mac OS X does not have a tools.jar--in favor of some other scheme, but the required classes are present. Therefore, the first two lines displayed after running run.sh are: run.sh: Missing file: /Library/Java/Home/lib/tools.jar run.sh: Unexpected results may occur. Make sure JAVA_HOME points to a JDK and not a JRE. I made the following changes to the run.sh file to remove this annoyance: --- run.sh Wed Sep 18 23:30:53 2002 +++ run.1.10Wed Sep 18 23:28:10 2002 @@ -118,11 +118,8 @@ JAVAC_JAR="$JAVA_HOME/lib/tools.jar" fi if [ ! -f "$JAVAC_JAR" ]; then -# Mac OS X does not have tools.jar in its JDK so only warn non-darwin OS -if [ "$darwin" != "true" ]; then - warn "Missing file: $JAVAC_JAR" - warn "Unexpected results may occur. Make sure JAVA_HOME points to a JDK and not a JRE." -fi + warn "Missing file: $JAVAC_JAR" + warn "Unexpected results may occur. Make sure JAVA_HOME points to a JDK and not a JRE." fi if [ "x$JBOSS_CLASSPATH" = "x" ]; then -- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376687&aid=611452&group_id=22866 --- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Build System... any ideas
> I think we should develop a new custom task to initialize the properties > and classpaths for the thirdparty packages. I wrote a hack to check > that directories are available before calling the task that declares the > classpath. We could write a task that takes the dir name properties to > set and paths to create, or we could load an xml file from the > thirdparty directory that had the above. I think either would be easier > to understand. Another possibility would be to make use of the script > task. I had looked into this, making a custom task, but dropped it... why... I can't remember. I think that making use of the script task would be a good idea. I have been thinking about using
RE: [JBoss-dev] Local home binding in Global JNDI namespace?
Not really... it would just make more sense if they were bound to java:/... don't you think? --jason > -Original Message- > From: [EMAIL PROTECTED] [mailto:jboss- > [EMAIL PROTECTED]] On Behalf Of Scott M Stark > Sent: Wednesday, September 18, 2002 7:17 PM > To: [EMAIL PROTECTED] > Subject: Re: [JBoss-dev] Local home binding in Global JNDI namespace? > > The local interfaces are not usable outside of the vm since they have > no invoker capable of handling the transport. Does it matter where > the default binding is? > > > Scott Stark > Chief Technology Officer > JBoss Group, LLC > > > - Original Message - > From: "Jason Dillon" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Wednesday, September 18, 2002 6:16 PM > Subject: RE: [JBoss-dev] Local home binding in Global JNDI namespace? > > > > Okay... though I think some code still depends on EJB names being > > unique. > > > > What about binging into the global JNDI namespace? Why not into > > java:/ejb/* or something? Or is there a reason that a remote client > > might need to talk to a EJBLocalHome? > > > > --jason > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development --- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Local home binding in Global JNDI namespace?
The local interfaces are not usable outside of the vm since they have no invoker capable of handling the transport. Does it matter where the default binding is? Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "Jason Dillon" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 6:16 PM Subject: RE: [JBoss-dev] Local home binding in Global JNDI namespace? > Okay... though I think some code still depends on EJB names being > unique. > > What about binging into the global JNDI namespace? Why not into > java:/ejb/* or something? Or is there a reason that a remote client > might need to talk to a EJBLocalHome? > > --jason --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Local home binding in Global JNDI namespace?
Okay... though I think some code still depends on EJB names being unique. What about binging into the global JNDI namespace? Why not into java:/ejb/* or something? Or is there a reason that a remote client might need to talk to a EJBLocalHome? --jason > -Original Message- > From: [EMAIL PROTECTED] [mailto:jboss- > [EMAIL PROTECTED]] On Behalf Of Scott M Stark > Sent: Wednesday, September 18, 2002 5:43 PM > To: [EMAIL PROTECTED] > Subject: Re: [JBoss-dev] Local home binding in Global JNDI namespace? > > No. EJB names are unique in a deployment, but not globally. > > > Scott Stark > Chief Technology Officer > JBoss Group, LLC > > > - Original Message - > From: "Jason Dillon" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Wednesday, September 18, 2002 5:38 PM > Subject: [JBoss-dev] Local home binding in Global JNDI namespace? > > > > Why are local references bound into the Global JNDI namespace? > > > > Also, why do the containers register using the JNDI name for uniqueness > > and not the EJB name? Don't EJB names have to be unique anyways? > > > > --jason > > > > > > > > --- > > This SF.NET email is sponsored by: AMD - Your access to the experts > > on Hammer Technology! Open Source & Linux Developers, register now > > for the AMD Developer Symposium. Code: EX8664 > > http://www.developwithamd.com/developerlab > > ___ > > Jboss-development mailing list > > [EMAIL PROTECTED] > > https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] Automated JBoss(HEAD Matrix2) Testsuite Results: 19-September-2002
Number of tests run: 932 Successful tests: 907 Errors:20 Failures: 5 [time of test: 19 September 2002 1:48 GMT] [java.version: 1.3.1_03] [java.vendor: Sun Microsystems Inc.] [java.vm.version: 1.3.1_03-b03] [java.vm.name: Java HotSpot(TM) Client VM] [java.vm.info: mixed mode] [os.name: Linux] [os.arch: i386] [os.version: 2.4.9-34] Useful resources: - http://lubega.com/testarchive/sun_jdk131_03 for the junit report of this test. - http://lubega.com/testarchive/sun_jdk131_03/logs/ for the logs for this test. - http://lubega.com for general test information. NOTE: If there are any errors shown above - this mail is only highlighting them - it is NOT indicating that they are being looked at by anyone. Remember - if a test becomes broken after your changes - fix it or fix the test! Oh dear - still got some errors! Thanks for all your effort - we really do love you! --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] property editors for the jmx html adaptor
Yes. Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "Jason Dillon" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 1:43 PM Subject: RE: [JBoss-dev] property editors for the jmx html adaptor > Does the new (Jboss) JMX HTML Adapter use PropertyEditors? > > --jason --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Local home binding in Global JNDI namespace?
No. EJB names are unique in a deployment, but not globally. Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "Jason Dillon" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 5:38 PM Subject: [JBoss-dev] Local home binding in Global JNDI namespace? > Why are local references bound into the Global JNDI namespace? > > Also, why do the containers register using the JNDI name for uniqueness > and not the EJB name? Don't EJB names have to be unique anyways? > > --jason > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] Local home binding in Global JNDI namespace?
Why are local references bound into the Global JNDI namespace? Also, why do the containers register using the JNDI name for uniqueness and not the EJB name? Don't EJB names have to be unique anyways? --jason --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Deploy/Redeploy broken in 3.0.3RC1?
Okay, so one problem down... did not realize that jndi-name was not used
for local home's...
--jason
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:jboss-
> [EMAIL PROTECTED]] On Behalf Of Jason Dillon
> Sent: Wednesday, September 18, 2002 5:22 PM
> To: [EMAIL PROTECTED]
> Subject: [JBoss-dev] Deploy/Redeploy broken in 3.0.3RC1?
>
> I am seeing exceptions about beans that have already been registered
> when redeploying an EAR:
>
>
>
> 17:11:25,819 WARN [ServiceController] Problem creating service
>
jboss.j2ee:service=EjbModule,url=file%/home/jason/ws/boldfish/ben/build/
>
output/ben-1.5.4beta/run/server/manager/tmp/deploy/home/jason/ws/boldfis
>
h/ben/build/output/ben-1.5.4beta/deploy/manager.ear/77.manager.ear-conte
> nts/requestspooler.jar
> javax.management.InstanceAlreadyExistsException:
>
jboss.j2ee:service=EJB,jndiName=com.boldfish.does.job.workflow.RequestSp
> oolerHome already registered.
> at
>
org.jboss.mx.server.registry.BasicMBeanRegistry.add(BasicMBeanRegistry.j
> ava:616)
> at
>
org.jboss.mx.server.registry.BasicMBeanRegistry.registerMBean(BasicMBean
> Registry.java:218)
> at
>
org.jboss.mx.server.MBeanServerImpl.registerMBean(MBeanServerImpl.java:9
> 75)
> at
>
org.jboss.mx.server.MBeanServerImpl.registerMBean(MBeanServerImpl.java:3
> 02)
> at org.jboss.ejb.EjbModule.createService(EjbModule.java:383)
> at
>
org.jboss.system.ServiceMBeanSupport.create(ServiceMBeanSupport.java:136
> )
> at java.lang.reflect.Method.invoke(Native Method)
> at
>
org.jboss.mx.capability.ReflectedMBeanDispatcher.invoke(ReflectedMBeanDi
> spatcher.java:284)
> at
> org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:517)
> at
>
org.jboss.system.ServiceController$ServiceProxy.invoke(ServiceController
> .java:976)
> at $Proxy7.create(Unknown Source)
> at
> org.jboss.system.ServiceController.create(ServiceController.java:315)
> at
> org.jboss.system.ServiceController.create(ServiceController.java:243)
> at java.lang.reflect.Method.invoke(Native Method)
> at
>
org.jboss.mx.capability.ReflectedMBeanDispatcher.invoke(ReflectedMBeanDi
> spatcher.java:284)
> at
> org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:517)
> at org.jboss.util.jmx.MBeanProxy.invoke(MBeanProxy.java:174)
> at $Proxy5.create(Unknown Source)
> at org.jboss.ejb.EJBDeployer.create(EJBDeployer.java:376)
> at
> org.jboss.deployment.MainDeployer.create(MainDeployer.java:755)
> at
> org.jboss.deployment.MainDeployer.create(MainDeployer.java:747)
> at
> org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:615)
> at
> org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:580)
> at java.lang.reflect.Method.invoke(Native Method)
> at
>
org.jboss.mx.capability.ReflectedMBeanDispatcher.invoke(ReflectedMBeanDi
> spatcher.java:284)
> at
> org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:517)
> at org.jboss.util.jmx.MBeanProxy.invoke(MBeanProxy.java:174)
> at $Proxy4.deploy(Unknown Source)
> at
>
org.jboss.deployment.scanner.URLDeploymentScanner.deploy(URLDeploymentSc
> anner.java:427)
> at
>
org.jboss.deployment.scanner.URLDeploymentScanner.scan(URLDeploymentScan
> ner.java:553)
> at
>
org.jboss.deployment.scanner.AbstractDeploymentScanner$ScannerThread.doS
> can(AbstractDeploymentScanner.java:212)
> at
>
org.jboss.deployment.scanner.AbstractDeploymentScanner$ScannerThread.loo
> p(AbstractDeploymentScanner.java:225)
> at
>
org.jboss.deployment.scanner.AbstractDeploymentScanner$ScannerThread.run
> (AbstractDeploymentScanner.java:202)
>
>
>
> Followed by a NPE:
>
>
>
> 17:11:26,302 ERROR [URLDeploymentScanner] Failed to deploy:
>
org.jboss.deployment.scanner.URLDeploymentScanner$DeployedURL@61cb4f20{
>
url=file:/home/jason/ws/boldfish/ben/build/output/ben-1.5.4beta/deploy/m
> anager.ear, deployedLastModified=1032393706000 }
> org.jboss.deployment.DeploymentException: Could not create deployment:
>
file:/home/jason/ws/boldfish/ben/build/output/ben-1.5.4beta/run/server/m
>
anager/tmp/deploy/home/jason/ws/boldfish/ben/build/output/ben-1.5.4beta/
> deploy/manager.ear/77.manager.ear-contents/customer.jar; - nested
> throwable: (RuntimeMBeanException: RuntimeException in MBean operation
>
'validateDeploymentState(,org.jboss.deployment.DeploymentInfo,org.jboss.
> deployment.DeploymentState)'
> Cause: java.lang.NullPointerException)
> at
> org.jboss.deployment.MainDeployer.start(MainDeployer.java:822)
> at
> org.jboss.deployment.MainDeployer.start(MainDeployer.java:794)
> at
> org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:616)
> at
> org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:580)
> at java.lang.reflect.Method.invoke(Native Method)
> at
>
org.jboss.mx.capabil
[JBoss-dev] Deploy/Redeploy broken in 3.0.3RC1?
I am seeing exceptions about beans that have already been registered
when redeploying an EAR:
17:11:25,819 WARN [ServiceController] Problem creating service
jboss.j2ee:service=EjbModule,url=file%/home/jason/ws/boldfish/ben/build/
output/ben-1.5.4beta/run/server/manager/tmp/deploy/home/jason/ws/boldfis
h/ben/build/output/ben-1.5.4beta/deploy/manager.ear/77.manager.ear-conte
nts/requestspooler.jar
javax.management.InstanceAlreadyExistsException:
jboss.j2ee:service=EJB,jndiName=com.boldfish.does.job.workflow.RequestSp
oolerHome already registered.
at
org.jboss.mx.server.registry.BasicMBeanRegistry.add(BasicMBeanRegistry.j
ava:616)
at
org.jboss.mx.server.registry.BasicMBeanRegistry.registerMBean(BasicMBean
Registry.java:218)
at
org.jboss.mx.server.MBeanServerImpl.registerMBean(MBeanServerImpl.java:9
75)
at
org.jboss.mx.server.MBeanServerImpl.registerMBean(MBeanServerImpl.java:3
02)
at org.jboss.ejb.EjbModule.createService(EjbModule.java:383)
at
org.jboss.system.ServiceMBeanSupport.create(ServiceMBeanSupport.java:136
)
at java.lang.reflect.Method.invoke(Native Method)
at
org.jboss.mx.capability.ReflectedMBeanDispatcher.invoke(ReflectedMBeanDi
spatcher.java:284)
at
org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:517)
at
org.jboss.system.ServiceController$ServiceProxy.invoke(ServiceController
.java:976)
at $Proxy7.create(Unknown Source)
at
org.jboss.system.ServiceController.create(ServiceController.java:315)
at
org.jboss.system.ServiceController.create(ServiceController.java:243)
at java.lang.reflect.Method.invoke(Native Method)
at
org.jboss.mx.capability.ReflectedMBeanDispatcher.invoke(ReflectedMBeanDi
spatcher.java:284)
at
org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:517)
at org.jboss.util.jmx.MBeanProxy.invoke(MBeanProxy.java:174)
at $Proxy5.create(Unknown Source)
at org.jboss.ejb.EJBDeployer.create(EJBDeployer.java:376)
at
org.jboss.deployment.MainDeployer.create(MainDeployer.java:755)
at
org.jboss.deployment.MainDeployer.create(MainDeployer.java:747)
at
org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:615)
at
org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:580)
at java.lang.reflect.Method.invoke(Native Method)
at
org.jboss.mx.capability.ReflectedMBeanDispatcher.invoke(ReflectedMBeanDi
spatcher.java:284)
at
org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:517)
at org.jboss.util.jmx.MBeanProxy.invoke(MBeanProxy.java:174)
at $Proxy4.deploy(Unknown Source)
at
org.jboss.deployment.scanner.URLDeploymentScanner.deploy(URLDeploymentSc
anner.java:427)
at
org.jboss.deployment.scanner.URLDeploymentScanner.scan(URLDeploymentScan
ner.java:553)
at
org.jboss.deployment.scanner.AbstractDeploymentScanner$ScannerThread.doS
can(AbstractDeploymentScanner.java:212)
at
org.jboss.deployment.scanner.AbstractDeploymentScanner$ScannerThread.loo
p(AbstractDeploymentScanner.java:225)
at
org.jboss.deployment.scanner.AbstractDeploymentScanner$ScannerThread.run
(AbstractDeploymentScanner.java:202)
Followed by a NPE:
17:11:26,302 ERROR [URLDeploymentScanner] Failed to deploy:
org.jboss.deployment.scanner.URLDeploymentScanner$DeployedURL@61cb4f20{
url=file:/home/jason/ws/boldfish/ben/build/output/ben-1.5.4beta/deploy/m
anager.ear, deployedLastModified=1032393706000 }
org.jboss.deployment.DeploymentException: Could not create deployment:
file:/home/jason/ws/boldfish/ben/build/output/ben-1.5.4beta/run/server/m
anager/tmp/deploy/home/jason/ws/boldfish/ben/build/output/ben-1.5.4beta/
deploy/manager.ear/77.manager.ear-contents/customer.jar; - nested
throwable: (RuntimeMBeanException: RuntimeException in MBean operation
'validateDeploymentState(,org.jboss.deployment.DeploymentInfo,org.jboss.
deployment.DeploymentState)'
Cause: java.lang.NullPointerException)
at
org.jboss.deployment.MainDeployer.start(MainDeployer.java:822)
at
org.jboss.deployment.MainDeployer.start(MainDeployer.java:794)
at
org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:616)
at
org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:580)
at java.lang.reflect.Method.invoke(Native Method)
at
org.jboss.mx.capability.ReflectedMBeanDispatcher.invoke(ReflectedMBeanDi
spatcher.java:284)
at
org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:517)
at org.jboss.util.jmx.MBeanProxy.invoke(MBeanProxy.java:174)
at $Proxy4.deploy(Unknown Source)
at
org.jboss.deployment.scanner.URLDeploymentScanner.deploy(URLDeploymentSc
anner.java:427)
at
org.jboss.deployment.scanner.URLDeploymentScanner.scan(URLDeploymentScan
ner.java:553)
at
org.jboss.deployment.scanner.AbstractDeploymentScanner$ScannerThread.doS
can(AbstractDeploymentScanner.java:
[JBoss-dev] URGENT: JBoss 3.0.1/2 with Jetty - Security Exploit report & fix
A security hole has been found that allows JSP source to be viewed remotely. This problem is present in the Jetty versions packaged as JBossWeb in JBoss versions 3.0.1 and 3.0.2. JBoss/Tomcat users need take NO action. Two upgrade paths are available : EITHER: Upgrade to JBoss 3.2 http://sourceforge.net/project/showfiles.php?group_id=22866 OR: download Jetty-4.1.0RC5 or above from : http://sourceforge.net/project/showfiles.php?group_id=7322 and replace the org.mortbay.jetty.jar and the org.mortbay.jmx.jars in your JBoss distrib's jbossweb.sar directories with the ones included in the lib/ directory of this package. Jules --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] property editors for the jmx html adaptor
Hi, To add more Property Editor at runtime you can : - add one or several package names to the PropertyEditorManager search path ( PropertyEditorManager.setEditorSearchPath(String[] path) ). - explicitly register a PropertyEditor for a class ( PropertyEditorManager.registerEditor(Class targetType, Class editorClass) ). Regards. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of David Jencks Sent: mercredi 18 septembre 2002 19:46 To: [EMAIL PROTECTED] Subject: Re: [JBoss-dev] property editors for the jmx html adaptor I didn't look at that part of the jmx-console, maybe it is already using this stuff. Jason wrote some nice property editor manager extensions that are used in at least ServiceController, RARDeployer, and the jmx ant task in varia. Using all our property editors would be very nice! I think there's an easy way to add more at runtime but I dont remember what it is. thanks david jencks On 2002.09.18 13:01:37 -0400 Matt Munz wrote: > > Hi all, > > I'm thinking of making my MBean more accessible through the jmx > console. > Specifically, I'd like to allow the user to enter values for complex > objects > (not primitives, Strings, etc.). Is there currently a way to add a > property > editor for an object so that I can enter it (as a String) in the jmx > console? > > - Matt > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Continuous Integration
I gave this a lot of thought a year or so ago, and even began to write a basic skeleton. I came up with a design as follows: A master which watches a [CVS] repository, maintains a list of builds-to-be-done (branch/label, jvm, OS etc), and a list of slave machines (and their OS). The master sends ant scripts to slaves which are capable of upgrading the Java environment on the slave, checking out src trees, doing builds etc. A number of slaves. These can be anywhere on the web. They simply register with the master and inform it when it may make use of their spare cycles. They run whatever ant files are sent to them and report results back to the master. The master correlates results, notifies developers who can be traced to checkins that break builds, maintains a hall of shame etc. Anyone can hit the presentation layer of the master over the web and see lots of nice graphs, the most important of which shows them exactly which timestamp they should use to get a working build for their platform etc... I thought it should be written as a nice Java compile/test farm which could plug into e.g. SourceForge, so even if you can't contribute code you can contribute cycles - an SF@Home. Diff this with current CC fn-ality - the result is my wish list. I'd love to write it, but just don't have the time. Jules Michael Rettig wrote: > Developers, > > I'm looking to setup a continuous integration enviroment for the JBoss project. Does >anyone know of one that exists already? > > If you don't know what continuous integration and what it can do for a project. You >can read more about it: > > http://martinfowler.com/articles/continuousIntegration.html > > I'm planning to setup a box that will detect changes to the CVS repo then kick off >an automated build to do a clean build and test of JBoss. A public webpage will be >generated that will be available to see the current status of the build, and past >builds. > > This process will help eliminate problems caused by developers checking in code that >simply doesn't compile or breaks tests. When the build does get broken, it only takes >a quick look at the web page to figure who is responsible. > > Also, the build can be configured to automatically email anyone that breaks the >build. > > CruiseControl will be used to build the project and generate the results. It's >another sourceforge OS project hosted on sourceforge. > > http://sourceforge.net/projects/cruisecontrol > > I'm looking for some feedback on this. What do you want the build to do? Do you want >it to email you when the build breaks? What information do you want displayed on the >webpage? > > Any feedback is appreciated. > > Thanks, > > Mike Rettig --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] error when buiding
hi jason, sorry for asking you again, but : how do I set this up??? through env variables? could you send the steps, or commands??? Jason Dillon wrote: > Hey. It is needed because the build scripts are using the 'resource' > protocol to include files. Since Ant does not have a mechanism to > include scripts we are forced to use the XML mechanism. Ant does > provide a method to load an entity from a URL though, so we use a URL > protocol handler to read a resource file to include custom bits. > > --jason > > > >>-Original Message- >>From: [EMAIL PROTECTED] [mailto:jboss- >>[EMAIL PROTECTED]] On Behalf Of Emerson Cargnin > > - > >>SICREDI Serviços >>Sent: Wednesday, September 18, 2002 12:23 PM >>To: [EMAIL PROTECTED] >>Subject: Re: [JBoss-dev] error when buiding >> >>hey jason, I still can't build jboss. Could you explain how to set up >>the url search path and why it's needed to the dev do that? >> >>Jason Dillon wrote: >> >>>You need to set the url search path to include >> > org.jboss.net.protocol. > >>>Look at one of the build.sh or build.bat scripts for details. >>> >>>--jason >>> >>> >>> >>> -Original Message- From: [EMAIL PROTECTED] [mailto:jboss- [EMAIL PROTECTED]] On Behalf Of Emerson >>> > Cargnin > >>>- >>> >>> SICREDI Serviços Sent: Wednesday, September 11, 2002 9:11 AM To: [EMAIL PROTECTED] Subject: [JBoss-dev] error when buiding why do i get this error when building jboss-all?? ./build.sh Searching for build.xml ... Buildfile: /home/emersonc/eclipse/workspace/jboss/build.xml BUILD FAILED Error reading project file: unknown protocol: resource -- Emerson Cargnin SICREDI - Tel : 3358-4860 --- In remembrance www.osdn.com/911/ ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development >>> >>> >>> >>> >>>--- >>>In remembrance >>>www.osdn.com/911/ >>>___ >>>Jboss-development mailing list >>>[EMAIL PROTECTED] >>>https://lists.sourceforge.net/lists/listinfo/jboss-development >>> >> >> >>-- >>Emerson Cargnin >>SICREDI - Tel : 3358-4860 >> >> >> >>--- >>This SF.NET email is sponsored by: AMD - Your access to the experts >>on Hammer Technology! Open Source & Linux Developers, register now >>for the AMD Developer Symposium. Code: EX8664 >>http://www.developwithamd.com/developerlab >>___ >>Jboss-development mailing list >>[EMAIL PROTECTED] >>https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > -- Emerson Cargnin SICREDI - Tel : 3358-4860 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Build System... any ideas
I think we should develop a new custom task to initialize the properties and classpaths for the thirdparty packages. I wrote a hack to check that directories are available before calling the task that declares the classpath. We could write a task that takes the dir name properties to set and paths to create, or we could load an xml file from the thirdparty directory that had the above. I think either would be easier to understand. Another possibility would be to make use of the script task. Other then that I think we should use the parallel task in the testsuite to speed up the xdoclet and jar tasks. I'm not sure if it would really speed it up but doing a one-test takes forever because of the xdoclet tasks. Also the default test suite takes so long that no one runs it anymore and most have created "smaller" sub suites, but I don't think that is a build system problem. -dain Jason Dillon wrote: > Can I get anyone who knows anything about Ant based build systems > (extensions, helpers, whatever) to send me some feedback on both > positive and negative experiences they have had. > > It is becoming very apparent that we need to overhaul the build system > to meet the current and future needs. > > I would appreciate any input you have. > > --jason --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
> The new 2.0 version that will be out soon is significantly > better. Even so, I never had much trouble with the old > version. Glad to hear you had better success, and I'm eager for a new version to give it another shot. > Besides CruiseControl is Open Source. If there are > problems, get the source and fix them. > Uh, we did that. That's why I said "you end up doing a lot of your own coding and config work to make cruise control happy". Thus, we did extend it. In fact, we have a version that is so far from the original with our own customizations, we may release it to the public if anyone thinks its worthwhile to their project. Lots of configuration additions mostly. > Are you going to tell me I should buy Weblogic too? Um, no, but feel free to if you feel so inclined. I prefer Jboss. Best Regards, James --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] Build System... any ideas
Can I get anyone who knows anything about Ant based build systems (extensions, helpers, whatever) to send me some feedback on both positive and negative experiences they have had. It is becoming very apparent that we need to overhaul the build system to meet the current and future needs. I would appreciate any input you have. --jason --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] error when buiding
Hey. It is needed because the build scripts are using the 'resource' protocol to include files. Since Ant does not have a mechanism to include scripts we are forced to use the XML mechanism. Ant does provide a method to load an entity from a URL though, so we use a URL protocol handler to read a resource file to include custom bits. --jason > -Original Message- > From: [EMAIL PROTECTED] [mailto:jboss- > [EMAIL PROTECTED]] On Behalf Of Emerson Cargnin - > SICREDI Serviços > Sent: Wednesday, September 18, 2002 12:23 PM > To: [EMAIL PROTECTED] > Subject: Re: [JBoss-dev] error when buiding > > hey jason, I still can't build jboss. Could you explain how to set up > the url search path and why it's needed to the dev do that? > > Jason Dillon wrote: > > You need to set the url search path to include org.jboss.net.protocol. > > Look at one of the build.sh or build.bat scripts for details. > > > > --jason > > > > > > > >>-Original Message- > >>From: [EMAIL PROTECTED] [mailto:jboss- > >>[EMAIL PROTECTED]] On Behalf Of Emerson Cargnin > > > > - > > > >>SICREDI Serviços > >>Sent: Wednesday, September 11, 2002 9:11 AM > >>To: [EMAIL PROTECTED] > >>Subject: [JBoss-dev] error when buiding > >> > >>why do i get this error when building jboss-all?? > >> > >> > >> ./build.sh > >>Searching for build.xml ... > >>Buildfile: /home/emersonc/eclipse/workspace/jboss/build.xml > >> > >> > >>BUILD FAILED > >>Error reading project file: unknown protocol: resource > >> > >> > >> > >>-- > >>Emerson Cargnin > >>SICREDI - Tel : 3358-4860 > >> > >> > >> > >>--- > >>In remembrance > >>www.osdn.com/911/ > >>___ > >>Jboss-development mailing list > >>[EMAIL PROTECTED] > >>https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > > > > > > > --- > > In remembrance > > www.osdn.com/911/ > > ___ > > Jboss-development mailing list > > [EMAIL PROTECTED] > > https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > > -- > Emerson Cargnin > SICREDI - Tel : 3358-4860 > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] property editors for the jmx html adaptor
Does the new (Jboss) JMX HTML Adapter use PropertyEditors? --jason > -Original Message- > From: [EMAIL PROTECTED] [mailto:jboss- > [EMAIL PROTECTED]] On Behalf Of Scott M Stark > Sent: Wednesday, September 18, 2002 11:24 AM > To: [EMAIL PROTECTED] > Subject: Re: [JBoss-dev] property editors for the jmx html adaptor > > You need to add the property editors to the PropertyEditorManager > Try using the PropertyEditorManagerService mbean in the properties- > service.xml > file: > >name="jboss:type=Service,name=PropertyEditorManager"> > > > > Either that or add the package your editors are in to the > PropertyEditorManager > EditorSearchPath, taking care to maintain the existing path. > > > Scott Stark > Chief Technology Officer > JBoss Group, LLC > > > - Original Message - > From: "Matt Munz" <[EMAIL PROTECTED]> > To: "JBoss Developers Group" <[EMAIL PROTECTED]> > Sent: Wednesday, September 18, 2002 10:01 AM > Subject: [JBoss-dev] property editors for the jmx html adaptor > > > > > > Hi all, > > > > I'm thinking of making my MBean more accessible through the jmx > console. > > Specifically, I'd like to allow the user to enter values for complex > objects > > (not primitives, Strings, etc.). Is there currently a way to add a > property > > editor for an object so that I can enter it (as a String) in the jmx > > console? > > > > - Matt > > > > > > > > --- > > This SF.NET email is sponsored by: AMD - Your access to the experts > > on Hammer Technology! Open Source & Linux Developers, register now > > for the AMD Developer Symposium. Code: EX8664 > > http://www.developwithamd.com/developerlab > > ___ > > Jboss-development mailing list > > [EMAIL PROTECTED] > > https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] build of clean most does not work in 3.2
I noticed this awhile ago. I am not sure why Ant was upgraded really... especially with out modify the buildmagic bits to be compatible. Blah... looks like it is time to revisit the build system =( --jason > -Original Message- > From: [EMAIL PROTECTED] [mailto:jboss- > [EMAIL PROTECTED]] On Behalf Of Scott M Stark > Sent: Wednesday, September 18, 2002 11:29 AM > To: [EMAIL PROTECTED] > Subject: [JBoss-dev] build of clean most does not work in 3.2 > > I cannot do a combined clean and build in 3.2. I'm guessing this is > a change in behavior between buildmagic and ant1.5. The build seems > to get stuck in the last module directory, which currently is catalina. > Here > is the output from the start of the jmx module most and it shows that the > module root is jboss-all/catalina rather than jboss-all/jmx: > > starksm@ironmaiden[build] 555>build.sh clean most > Searching for build.xml ... > > ... > > _buildmagic:init: > [available] DEPRECATED - used to override an existing > property. > [available] Build file should not reuse the same property name for > different values. > [available] DEPRECATED - used to override an existing > property. > [available] Build file should not reuse the same property name for > different values. > [available] DEPRECATED - used to override an existing > property. > [available] Build file should not reuse the same property name for > different values. > [available] DEPRECATED - used to override an existing > property. > [available] Build file should not reuse the same property name for > different values. > > _buildmagic:init:local-properties: > > _buildmagic:init:buildlog: > > configure: > [echo] groups: default > [echo] modules: > jmx,common,system,j2ee,naming,management,transaction,server,security,mes sa > ging,connector,cluster,jetty,varia,jboss.net,iiop,ca > talina > > init: > > _buildmagic:modules:most: > [execmodules] > [execmodules] > == > [execmodules] == Executing 'most' in module 'jmx'... > [execmodules] == > > _buildmagic:init: > > configure: > > init: > > compile-classes: > > compile-resources: > [mkdir] Created dir: /usr/local/jboss/JBoss3.2/jboss- > all/catalina/output/resources > [copy] Copying 7 files to /usr/local/jboss/JBoss3.2/jboss- > all/catalina/output/resources > > compile-etc: > [mkdir] Created dir: /usr/local/jboss/JBoss3.2/jboss- > all/catalina/output/etc > [copy] Copying 1 file to /usr/local/jboss/JBoss3.2/jboss- > all/catalina/output/etc > [copy] Copied 1 empty directory to /usr/local/jboss/JBoss3.2/jboss- > all/catalina/output/etc > > compile: > > jars: > > most: > [execmodules] > [execmodules] == > [execmodules] == Finished with 'most' in module 'jmx'. > [execmodules] > == > > > Scott Stark > Chief Technology Officer > JBoss Group, LLC > > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-611270 ] No longer boots form dirs with spaces
Bugs item #611270, was opened at 2002-09-18 15:33
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=611270&group_id=22866
Category: JBossServer
Group: v3.0 Rabbit Hole
Status: Open
Resolution: None
Priority: 5
Submitted By: Nathan W. Phelps (nphelps)
Assigned to: Nobody/Anonymous (nobody)
Summary: No longer boots form dirs with spaces
Initial Comment:
I'm using JBoss 3.0.2 on Windows XP and JDK 1.4.1
and I am unable to get it to boot when JBOSS_HOME
is "C:\Program Files\JBoss." I used JBoss 3.0 from
this exact same directory, but evidenlty this is broken
now. This was one of the things I really liked about
JBoss. I hate putting things at root. Here is the server
trace:
15:30:59,941 INFO [Server] Starting JBoss (MX
MicroKernel)...
15:30:59,941 INFO [Server] Release ID: JBoss
[WonderLand] 3.2.0beta (build: JBoss_3_2_0_beta
200209161335)
15:30:59,941 DEBUG [Server] Using config:
org.jboss.system.server.ServerConfigImpl@7bd9f2
15:30:59,941 DEBUG [Server] Server type: class
org.jboss.system.server.ServerImpl
15:30:59,957 INFO [Server] Home Dir: C:\Program%
20Files\jboss-3.2.0beta
15:30:59,957 INFO [Server] Home URL:
file:/C:/Program%20Files/jboss-3.2.0beta/
15:30:59,957 INFO [Server] Library URL:
file:/C:/Program%20Files/jboss-3.2.0beta/lib/
15:30:59,973 INFO [Server] Patch URL: null
15:30:59,973 INFO [Server] Server Name: default
15:30:59,973 INFO [Server] Server Home Dir:
C:\Program%20Files\jboss-3.2.0beta\server\default
15:30:59,973 INFO [Server] Server Home URL:
file:/C:/Program%20Files/jboss-3.2.0beta/server/default/
15:30:59,988 INFO [Server] Server Data Dir:
C:\Program%20Files\jboss-3.2.0beta\server\default\data
15:30:59,988 INFO [Server] Server Temp Dir:
C:\Program%20Files\jboss-3.2.0beta\server\default\tmp
15:30:59,988 INFO [Server] Server Config URL:
file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/conf/
15:30:59,988 INFO [Server] Server Library URL:
file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/lib/
15:30:59,988 INFO [Server] Root Deployemnt
Filename: jboss-service.xml
15:31:00,004 INFO [Server] Starting General Purpose
Architecture (GPA)...
15:31:00,270 DEBUG [Server] Created MBeanServer:
org.jboss.mx.server.MBeanServerImpl@2ce908
15:31:00,285 DEBUG [Server] Boot url list:
[file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/conf/]
15:31:00,285 DEBUG [Server] Creating loader for URL:
file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/conf/
15:31:00,301 DEBUG [UnifiedClassLoader] New jmx
UCL with url file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/conf/
15:31:00,301 DEBUG [UnifiedClassLoader] New jmx
UCL with url file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/conf/
15:31:00,301 DEBUG [UnifiedLoaderRepository2]
Adding
org.jboss.mx.loading.UnifiedClassLoader@d80be3{
url=file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/conf/ }
15:31:00,317 DEBUG [UnifiedLoaderRepository2] Failed
to update pkgs for
cl=org.jboss.mx.loading.UnifiedClassLoader@d80be3{
url=file:/C:/Program%20Files/jboss-
3.2.0beta/server/default/conf/ }
java.io.FileNotFoundException: C:\Program%
20Files\jboss-3.2.0beta\server\default\conf (The system
cannot find the file specified)
at java.io.FileInputStream.open(Native
Method)
at java.io.FileInputStream.
(FileInputStream.java:103)
at
org.jboss.mx.loading.ClassLoaderUtils$ClassPathIterato
r.(ClassLoaderUtils.java:254)
at
org.jboss.mx.loading.ClassLoaderUtils.updatePackageM
ap(ClassLoaderUtils.java:42)
at
org.jboss.mx.loading.UnifiedLoaderRepository2.updateP
ackageMap(UnifiedLoaderRepository2.java:581)
at
org.jboss.mx.loading.UnifiedLoaderRepository2.addUnifie
dClassLoader(UnifiedLoaderRepository2.java:566)
at
org.jboss.mx.loading.UnifiedLoaderRepository2.addClas
sLoader(UnifiedLoaderRepository2.java:526)
at
org.jboss.mx.loading.UnifiedLoaderRepository2.newClas
sLoader(UnifiedLoaderRepository2.java:114)
at
sun.reflect.NativeMethodAccessorImpl.invoke0(Native
Method)
at
sun.reflect.NativeMethodAccessorImpl.invoke
(NativeMethodAccessorImpl.java:39)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke
(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke
(Method.java:324)
at
org.jboss.mx.capability.ReflectedMBeanDispatcher.invok
e(ReflectedMBeanDispatcher.java:284)
at
org.jboss.mx.server.MBeanServerImpl.invoke
(MBeanServerImpl.java:549)
at
org.jboss.system.server.ServerImpl.initBootLibraries
(ServerImpl.java:447)
at org.jboss.system.server.ServerImpl.doStart
(ServerImpl.java:276)
at org.jboss.system.server.ServerImpl.start
(ServerImpl.java:232)
at org.jboss.Main.boot(Main.java:146)
at org.jboss.Main$1.run(Main.java:379)
at java.lang.Thread.run(Thread.java:536)
15:31:00,332 INFO [ServerInfo] Java version: 1.4.1,Sun
Microsystems Inc.
15:31:00,332 INFO [Serv
[JBoss-dev] [ jboss-Bugs-610907 ] Using JMS in MDB/container-managed TX
Bugs item #610907, was opened at 2002-09-17 20:28
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610907&group_id=22866
Category: JBossMQ
Group: v3.0 Rabbit Hole
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Nobody/Anonymous (nobody)
Summary: Using JMS in MDB/container-managed TX
Initial Comment:
I've gotten MDB to work correctly with rolling-back
the database when I call mdc.setRollbackOnly().
However, I don't know how to create a QueueSession/Sender
which will send its messages after the onMessage() method
exists.
I have bought the JBoss documentation, but it does not have
an example of this case specifically. I suspect (but do not know)
that JBoss does not support automatic QueueSession
commit after an MDB exits its onMessage method. Looking
at the source after a few hours has gotten me confused.
I suspect this is NOT supported, if it is supported,
please forgive this bug submission and point me in
the right direction. This _should_ be documented,
at least document how this is correctly done.
The JMS tutorial:
http://java.sun.com/products/jms/tutorial/
has several examples of Container-managed TX
for MDB, which don't seem to work in JBoss.
Does anyone have an example of this sending out messages:
public class MessageDrivenBean implements MessageListener {
// Container-managed transactions with Required
// transaction attribute
public void onMessage(...) {
try {
QueueConnection qc = qcf.createQueueConnection();
QueueSession qs = qc.createQueueSession(true, 0);
QueueSender qsend = qs.createQueueSender("queue/A");
TextMessage m = qs.createTextMessage();
m.setText("test");
qsend.send(m);
/*
qs.close(); // JMS javadoc says this rollbacks
// In any case, doesn't seem to work
*/
qc.close();
} catch (Exception e) {
e.printStackTrace();
}
}
}
--
>Comment By: Elias Ross (genman)
Date: 2002-09-18 13:02
Message:
Logged In: YES
user_id=556458
Thanks for responding corby.
I guess the question in my mind, is a fix forthcoming?
(In the next month or so.) It should be easy to associate
a javax.jms.Session with a transaction, upon its creation,
and have it automatically commit when
onMessage() completes. If I could wrap my head around the
code, I could probably come up with a quick fix in a few
days. (My theory anyway.)
If you could tell me what's going on with JBoss,
I can either patiently wait for a fix, or go
ahead now and re-write parts of my application.
Again, I appreciate knowing MDB's have this limitation.
--
Comment By: Corby (corby)
Date: 2002-09-18 07:07
Message:
Logged In: YES
user_id=25032
Your bug submission is correct, Elias. This has been a long-
standing and frustrating issue with JBoss. MDB's with
container-managed transactions do not seem to have the
same transactional semantics as other EJB's, and you MUST
use bean-managed transactions if you want to enlist mutiple
resources in an MDB transaction.
--
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610907&group_id=22866
---
This SF.NET email is sponsored by: AMD - Your access to the experts
on Hammer Technology! Open Source & Linux Developers, register now
for the AMD Developer Symposium. Code: EX8664
http://www.developwithamd.com/developerlab
___
Jboss-development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] Automated JBoss(Branch_3_0) Testsuite Results: 18-September-2002
Number of tests run: 942
Successful tests: 926
Errors:10
Failures: 6
[time of test: 18 September 2002 12:52 GMT]
[java.version: 1.3.1]
[java.vendor: Apple Computer, Inc.]
[java.vm.version: 1.3.1_03-69]
[java.vm.name: Java HotSpot(TM) Client VM]
[java.vm.info: mixed mode]
[os.name: Mac OS X]
[os.arch: ppc]
[os.version: 10.2]
See http://lubega.com/testarchive/${build.uid} for details of this test.
See http://lubega.com for general test information.
NOTE: If there are any errors shown above - this mail is only highlighting
them - it is NOT indicating that they are being looked at by anyone.
Remember - if a test becomes broken after your changes - fix it or fix the test!
Oh dear - still got some errors!
Thanks for all your effort - we really do love you!
---
This SF.NET email is sponsored by: AMD - Your access to the experts
on Hammer Technology! Open Source & Linux Developers, register now
for the AMD Developer Symposium. Code: EX8664
http://www.developwithamd.com/developerlab
___
Jboss-development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
On 18 Sep, David Jencks wrote: > [...] > (I think) the jars in lib rather than server/[conf]/lib are not going to be > versionable since they are used in jboss startup to set up the versioning > mechanism. Stuff in server/[conf]/lib should be versionable by deploying > it in a .sar, although you may have problems with deployment order. Ok. I see that they are loaded by a normal URLClassLoader, and seems to be in the parent tree of the classloader. Ugh. All XML classes, log4j-boot, gnu-regexp will not be possible to version then. Have to overwite... > > I wasn't aware that there had been any changes in concurrent.jar for > years... can you provide details, we should probably upgrade. The oswego site seems to be down. The stuff i use here was downloaded 2002-02-03 and seems to contain new methods used in the stuff I try to integrate (XmlBlaster). I will check closer tomorrow. Thanks //Peter > > thanks > david jencks >> >> Thanks >> >> //Peter >> > >> > thanks >> > david jencks >> > >> >> >> >> >> >> Or have I gone completely nuts? >> >> >> >> //Peter >> >> -- >> >> >> >> Peter Antman Chief Systems Architect, Business Development >> >> Technology in Media, Box 34105 100 26 Stockholm >> >> WWW: http://www.tim.seWWW: http://www.backsource.org >> >> Email: [EMAIL PROTECTED] >> >> Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 >> >> >> >> >> >> >> >> >> >> --- >> >> This SF.NET email is sponsored by: AMD - Your access to the experts >> >> on Hammer Technology! Open Source & Linux Developers, register now >> >> for the AMD Developer Symposium. Code: EX8664 >> >> http://www.developwithamd.com/developerlab >> >> ___ >> >> Jboss-development mailing list >> >> [EMAIL PROTECTED] >> >> https://lists.sourceforge.net/lists/listinfo/jboss-development >> >> >> >> >> > >> > >> > --- >> > This SF.NET email is sponsored by: AMD - Your access to the experts >> > on Hammer Technology! Open Source & Linux Developers, register now >> > for the AMD Developer Symposium. Code: EX8664 >> > http://www.developwithamd.com/developerlab >> > ___ >> > Jboss-development mailing list >> > [EMAIL PROTECTED] >> > https://lists.sourceforge.net/lists/listinfo/jboss-development >> >> -- >> >> Peter Antman Chief Systems Architect, Business Development >> Technology in Media, Box 34105 100 26 Stockholm >> WWW: http://www.tim.se WWW: http://www.backsource.org >> Email: [EMAIL PROTECTED] >> Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 >> >> >> >> >> --- >> This SF.NET email is sponsored by: AMD - Your access to the experts >> on Hammer Technology! Open Source & Linux Developers, register now >> for the AMD Developer Symposium. Code: EX8664 >> http://www.developwithamd.com/developerlab >> ___ >> Jboss-development mailing list >> [EMAIL PROTECTED] >> https://lists.sourceforge.net/lists/listinfo/jboss-development >> >> > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development -- Peter Antman Technology in Media, Box 34105 100 26 Stockholm Systems ArchitectWWW: http://www.tim.se Email: [EMAIL PROTECTED]WWW: http://www.backsource.org Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] error when buiding
hey jason, I still can't build jboss. Could you explain how to set up the url search path and why it's needed to the dev do that? Jason Dillon wrote: > You need to set the url search path to include org.jboss.net.protocol. > Look at one of the build.sh or build.bat scripts for details. > > --jason > > > >>-Original Message- >>From: [EMAIL PROTECTED] [mailto:jboss- >>[EMAIL PROTECTED]] On Behalf Of Emerson Cargnin > > - > >>SICREDI Serviços >>Sent: Wednesday, September 11, 2002 9:11 AM >>To: [EMAIL PROTECTED] >>Subject: [JBoss-dev] error when buiding >> >>why do i get this error when building jboss-all?? >> >> >> ./build.sh >>Searching for build.xml ... >>Buildfile: /home/emersonc/eclipse/workspace/jboss/build.xml >> >> >>BUILD FAILED >>Error reading project file: unknown protocol: resource >> >> >> >>-- >>Emerson Cargnin >>SICREDI - Tel : 3358-4860 >> >> >> >>--- >>In remembrance >>www.osdn.com/911/ >>___ >>Jboss-development mailing list >>[EMAIL PROTECTED] >>https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > > --- > In remembrance > www.osdn.com/911/ > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > -- Emerson Cargnin SICREDI - Tel : 3358-4860 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Continuous Integration
If you create a setup that can be used for branches other than head I would look at using it to monitor the 3.0 and 3.2 branches. Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "Kevin Conner" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 8:09 AM Subject: RE: [JBoss-dev] Continuous Integration > It's not initiated on checkin though, it polls the CVS repository > for changes and will initiate the build once the changes have settled > down. This is all configurable in cruisecontrol. > > If the build fails then cruisecontrol will email the people who checked > in code since the previous checkout. The idea is that this should be > frequent enough to catch small amounts of changes which cause problems. > > I've introduced cruisecontrol here so that we can automate builds > and testing. It has been good for us as we are a small team. > > Having said all that, I'm not sure that there is any benefit of using > this in jboss if there is already a mechanism to build on a regular > interval (which there is). If you were starting from scratch then I > would certainly recommend cruisecontrol but why change if your current > mechanism is sufficient? > > Kev > > Kevin Conner > Orchard Information Systems Limited > Newcastle Technopole, Kings Manor > Newcastle Upon Tyne, NE1 6PA. United Kingdom > Registered in England, Number 1900078 > Tel: +44 (0) 191-2032536 Fax: +44 (0) 191 2302515 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] build of clean most does not work in 3.2
I cannot do a combined clean and build in 3.2. I'm guessing this is a change in behavior between buildmagic and ant1.5. The build seems to get stuck in the last module directory, which currently is catalina. Here is the output from the start of the jmx module most and it shows that the module root is jboss-all/catalina rather than jboss-all/jmx: starksm@ironmaiden[build] 555>build.sh clean most Searching for build.xml ... ... _buildmagic:init: [available] DEPRECATED - used to override an existing property. [available] Build file should not reuse the same property name for different values. [available] DEPRECATED - used to override an existing property. [available] Build file should not reuse the same property name for different values. [available] DEPRECATED - used to override an existing property. [available] Build file should not reuse the same property name for different values. [available] DEPRECATED - used to override an existing property. [available] Build file should not reuse the same property name for different values. _buildmagic:init:local-properties: _buildmagic:init:buildlog: configure: [echo] groups: default [echo] modules: jmx,common,system,j2ee,naming,management,transaction,server,security,messaging,connector,cluster,jetty,varia,jboss.net,iiop,ca talina init: _buildmagic:modules:most: [execmodules] [execmodules] == [execmodules] == Executing 'most' in module 'jmx'... [execmodules] == _buildmagic:init: configure: init: compile-classes: compile-resources: [mkdir] Created dir: /usr/local/jboss/JBoss3.2/jboss-all/catalina/output/resources [copy] Copying 7 files to /usr/local/jboss/JBoss3.2/jboss-all/catalina/output/resources compile-etc: [mkdir] Created dir: /usr/local/jboss/JBoss3.2/jboss-all/catalina/output/etc [copy] Copying 1 file to /usr/local/jboss/JBoss3.2/jboss-all/catalina/output/etc [copy] Copied 1 empty directory to /usr/local/jboss/JBoss3.2/jboss-all/catalina/output/etc compile: jars: most: [execmodules] [execmodules] == [execmodules] == Finished with 'most' in module 'jmx'. [execmodules] == Scott Stark Chief Technology Officer JBoss Group, LLC --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] property editors for the jmx html adaptor
You need to add the property editors to the PropertyEditorManager Try using the PropertyEditorManagerService mbean in the properties-service.xml file: Either that or add the package your editors are in to the PropertyEditorManager EditorSearchPath, taking care to maintain the existing path. Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "Matt Munz" <[EMAIL PROTECTED]> To: "JBoss Developers Group" <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 10:01 AM Subject: [JBoss-dev] property editors for the jmx html adaptor > > Hi all, > > I'm thinking of making my MBean more accessible through the jmx console. > Specifically, I'd like to allow the user to enter values for complex objects > (not primitives, Strings, etc.). Is there currently a way to add a property > editor for an object so that I can enter it (as a String) in the jmx > console? > > - Matt > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
On 18 Sep, Scott M Stark wrote: > In the 3.2RC1 release scoped class loaders will be the default for > ears and sars, Great news. >maybe wars but you get that with the servlet 2.3 > class loading model. Last I read in the jetty conf or something that was turned of by default. >You will be able to configure shared class loader > repositories across deployments if desired. Sounds terrific. Would it not also be greate to be able to have scoped repository (encapsulation) and then export stuff to the global classloader that should be available to other modules (interface). //Peter > > > Scott Stark > Chief Technology Officer > JBoss Group, LLC > > > - Original Message - > From: "David Jencks" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Wednesday, September 18, 2002 7:09 AM > Subject: Re: [JBoss-dev] ClassLoader frustration > > >> >> I set up the ulr stuff in 3.2 so it would be really easy to extend to other >> module types. Go ahead if you would like it for sars. In 3.0 branch it is >> a little harder. I didn't feel really pressed to do this since you can put >> any kind of package you want in an .ear and get the same effect. > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development -- Peter Antman Technology in Media, Box 34105 100 26 Stockholm Systems ArchitectWWW: http://www.tim.se Email: [EMAIL PROTECTED]WWW: http://www.backsource.org Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] property editors for the jmx html adaptor
I didn't look at that part of the jmx-console, maybe it is already using this stuff. Jason wrote some nice property editor manager extensions that are used in at least ServiceController, RARDeployer, and the jmx ant task in varia. Using all our property editors would be very nice! I think there's an easy way to add more at runtime but I dont remember what it is. thanks david jencks On 2002.09.18 13:01:37 -0400 Matt Munz wrote: > > Hi all, > > I'm thinking of making my MBean more accessible through the jmx > console. > Specifically, I'd like to allow the user to enter values for complex > objects > (not primitives, Strings, etc.). Is there currently a way to add a > property > editor for an object so that I can enter it (as a String) in the jmx > console? > > - Matt > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Continuous Integration
The more I think about this the more useful I think it would be. I'd say if you are willing to spend the time to set it up and maintain it go for it. You might want to see if anyone else has an opinion. david jencks On 2002.09.18 11:51:44 -0400 Michael Rettig wrote: > > - Original Message - > From: David Jencks <[EMAIL PROTECTED]> > > Aside from using cruise control, how does this differ from what chris > > kimpton already has set up(http://www.lubega.com/), which runs the > > testsuite at least once a day and I think does hourly builds? How does > > your proposed system distinguish a broken build from a lengthy > multi-part > > checkin where writing the comments on the next chunk takes 10 minutes? > > CruiseControl has a "quiet period" which means that it waits for a > developer to stop checking in code before it starts it's build. Once no > checkins have been detected for x amount of minutes, then the build > starts. > > I've been the build master for Java projects with 700k loc and 50+ > developers all actively checking in and it is extremely rare that the > build starts mid checkin. When it does, it usually fixes itself within a > few minutes. > > CruiseControl differs in that it ties the build to checkins, so you can > very easily address accountability for broken builds. The web page that > is generated will give you a list of files that were checked om. More > importantly, there are names attached to those files, so you can very > easily see who checked in Foo.java that does not compile or > SimpleTest.java that fails. > > In a big project when tests break, accountability needs to be established > and quickly - and this means when, how, and who. Broken windows cannot be > tolerated whatsoever. > > Although, http://lubega.com will tell you what is broken, and > approximately when it was broken. It can't link the broken build to a > specific checkin and ultimately to the offending party. > > > > > How would your system apportion blame when there are several checkins > at > > the same time (either interlaced or within less time than it takes to > > checkout and do a clean build) and the build breaks? > > The system cannot be perfect, and tell you exactly who broke what, > especially with interlaced checkins. However, these situations are often > rare, and continuous integration with CruiseControl will give you enough > information that you can figure out what happened and solve the problem > within a few minutes. I've been in these situations before, and its never > taken me more than 15 minutes to trace through the CruiseControl logs and > find the offending developer within a few minutes. > > Hope this helps. > > Mike > -- > __ > Sign-up for your own FREE Personalized E-mail at Mail.com > http://www.mail.com/?sr=signup > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
Right now in all jboss 3+ versions the only way to get scoped classloading is to put whatever should be in a scope in an .ear and include a jboss-app.xml specifying the loader repository object name next to application.xml in the ear. Many .ears can share the same loader repository. When the .ear is created, a new loader rep. will be created using the object name supplied if it does not already exist. THere are some examples of jboss-app.xml files in the testsuite. david jencks On 2002.09.18 11:39:33 -0400 Brian Macy wrote: > This thread may be indentifying my problem my EJBs are in a jar > file, my web client is in a war, and my mbean is in a sar. I want all my > deployments to have scoped class loaders. > > I'm gathering from this discussion: > a) The jar can not have a scoped class loader, needs to be an ear > b) By default the ?ar files do not use scoped classloaders and it such > behavior needs to be turned on in a config file in 3.0.2 > > Am I close? If b) is true I please for someone to tell me what the > option is and where it goes :) > > Brian Macy > > > Scott M Stark wrote: > > In the 3.2RC1 release scoped class loaders will be the default for > > ears and sars, maybe wars but you get that with the servlet 2.3 > > class loading model. You will be able to configure shared class loader > > repositories across deployments if desired. > > > > > > Scott Stark > > Chief Technology Officer > > JBoss Group, LLC > > > > > > - Original Message - > > From: "David Jencks" <[EMAIL PROTECTED]> > > To: <[EMAIL PROTECTED]> > > Sent: Wednesday, September 18, 2002 7:09 AM > > Subject: Re: [JBoss-dev] ClassLoader frustration > > > > > > > >>I set up the ulr stuff in 3.2 so it would be really easy to extend to > other > >>module types. Go ahead if you would like it for sars. In 3.0 branch > it is > >>a little harder. I didn't feel really pressed to do this since you can > put > >>any kind of package you want in an .ear and get the same effect. > > > > > > > > > > --- > > This SF.NET email is sponsored by: AMD - Your access to the experts > > on Hammer Technology! Open Source & Linux Developers, register now > > for the AMD Developer Symposium. Code: EX8664 > > http://www.developwithamd.com/developerlab > > ___ > > Jboss-development mailing list > > [EMAIL PROTECTED] > > https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] property editors for the jmx html adaptor
Hi all, I'm thinking of making my MBean more accessible through the jmx console. Specifically, I'd like to allow the user to enter values for complex objects (not primitives, Strings, etc.). Is there currently a way to add a property editor for an object so that I can enter it (as a String) in the jmx console? - Matt --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
- Original Message - From: "James Higginbotham" <[EMAIL PROTECTED]> > I have just recently started down the cruise control path on a personal > project, and my company has since dumped it in favor of an Austin-based > startup (Buildforge) which offers better support and feature set. The new 2.0 version that will be out soon is significantly better. Even so, I never had much trouble with the old version. Besides CruiseControl is Open Source. If there are problems, get the source and fix them. Are you going to tell me I should buy Weblogic too? Mike -- __ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
-> If you were starting from scratch then I > would certainly recommend cruisecontrol but why change if your current > mechanism is sufficient? The current mechanism is sufficient, but it could be much better. The key distinction here is that CruiseControl ties checkins to build failures and presents this information in a public web page. On prior projects, when I broke the build (yes, I admit it) the build itself would notify me via email within minutes, then if I didn't fix it, a steady stream of developers would call me to tell me what an idiot I am for breaking the build. They knew I broke the build because the very public web page had the results of the broken build and my list of checkins attached to it. No mystery there. In a very short amount of time, I learned not to break the build. This was true of all the other developers too. Do you have any current mechanisms that can give you this rapid feedback? When things break how do you address accountability? How long does it take? -- __ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Continuous Integration
The more I think about it the more I like the idea... Can you make it so only the relatively quick unit tests are run on each build and emails are sent to the list and recent committers if say the broken-test count increases by more than 5 or the number of tests decrease by more than 5? (breaking tests by preventing them from deploying results in a decrease in test count rather than in increase in broken test count). I think running the full test suite might take too long after each change... thanks david jencks On 2002.09.18 11:32:30 -0400 Matt Munz wrote: > From what I've seen of CruiseControl, I think it's a useful tool, and if > Michael or someone else wants to set it up, I'll use it. > > As far as I can tell, both the lubega and cruise control systems operate > remotely. I don't see any reason why they can't coexist. > > Nothing wrong with a little diversity, right? > > - Matt > > > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED]]On Behalf Of Kevin > Conner > Sent: Wednesday, September 18, 2002 11:09 AM > To: '[EMAIL PROTECTED]' > Subject: RE: [JBoss-dev] Continuous Integration > > > > Right, a build initiated on checkin is in general guarenteed > > to fail due > > to an inconsistent view of the repository. Start bitching more about > > specific problems due to checkins that developers are not fixing. If > > it continues unabated the offending developer is gone. > > It's not initiated on checkin though, it polls the CVS repository > for changes and will initiate the build once the changes have settled > down. This is all configurable in cruisecontrol. > > If the build fails then cruisecontrol will email the people who checked > in code since the previous checkout. The idea is that this should be > frequent enough to catch small amounts of changes which cause problems. > > I've introduced cruisecontrol here so that we can automate builds > and testing. It has been good for us as we are a small team. > > Having said all that, I'm not sure that there is any benefit of using > this in jboss if there is already a mechanism to build on a regular > interval (which there is). If you were starting from scratch then I > would certainly recommend cruisecontrol but why change if your current > mechanism is sufficient? > > Kev > > Kevin Conner > Orchard Information Systems Limited > Newcastle Technopole, Kings Manor > Newcastle Upon Tyne, NE1 6PA. United Kingdom > Registered in England, Number 1900078 > Tel: +44 (0) 191-2032536 Fax: +44 (0) 191 2302515 > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] Potential 3.2 Classloader Issue w/ WAR files
I originally posted this over on the Jetty list, and they suggested I cross-post to get you guys in the loop. I downloaded and checked out JBoss3.2.0 beta yesterday. I'm on MacOS X 10.2 / Apple's 1.3.1_03 JVM. I deployed 2 ears. Both ears work fine on 3.02. ** ALL EARS ARE DEPLOYED IN THEIR OWN CLASSLOADERS USING THE jboss-app.xml FILE TAGS ** This is because I deploy a similar content management system in different EARs and without this, they classes were conflicting... Anyway, it works great in 3.02. The first ear to deploy worked without problems. It contained one WAR and one JAR w/EJBs. The WAR contained a JAR with some classes in it used by servlets. Worked fine. The second ear to deploy contained two WARs and one JAR w/EJBs. Both WARs had similar structure. One worked fine, the other would act as if the stuff in its WEB-INF/lib directory were not added to the JSP compiler classpath. It would complain that included classes in the JSPs did not exist. I verified that they did exist by exploding the archive. Also, the same EAR functioned correctly on 3.02. Anyway, if anyone has any clues or would like some more information let me know. Hunter --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
Title: RE: [JBoss-dev] Continuous Integration FWIW: CC is interval based. It wakes up, say every 30 seconds (or 30 sec after last build), and does a check of the repos to see if any commits have happened. If a commit has happened 'recently' (based on quiet period setting), it will go back to sleep for some amount of time before trying again. It has worked very well for us - I can't remember any build failures due to inconsistent state. But, if the commits are frequent enough the quiet interval would be sufficiently long to give it little advantage over the lubega hourlies. Mike -Original Message- From: Scott M Stark [mailto:[EMAIL PROTECTED]] Sent: Wednesday, September 18, 2002 10:39 AM To: [EMAIL PROTECTED] Subject: Re: [JBoss-dev] Continuous Integration Right, a build initiated on checkin is in general guarenteed to fail due to an inconsistent view of the repository. Start bitching more about specific problems due to checkins that developers are not fixing. If it continues unabated the offending developer is gone. Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "David Jencks" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 7:26 AM Subject: Re: [JBoss-dev] Continuous Integration > Aside from using cruise control, how does this differ from what chris > kimpton already has set up(http://www.lubega.com/), which runs the > testsuite at least once a day and I think does hourly builds? How does > your proposed system distinguish a broken build from a lengthy multi-part > checkin where writing the comments on the next chunk takes 10 minutes? > > > I'm certainly not averse to a better system, however one of the drawbacks > of the current system is that the results are mostly ignored: people will > often fix problems that break the build the next day, but most people seem > to feel no obligation to do anything about changes that say only break 50 > or 100 tests (ok I'm probably exaggerating, but not by enough). > > > How would your system apportion blame when there are several checkins at > the same time (either interlaced or within less time than it takes to > checkout and do a clean build) and the build breaks? > > Although I might sound negative, I actually think what chris has set up has > been incredibly useful and improvements or a better system would be > wonderful. > > thanks > > david jencks --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Continuous Integration
- Original Message - From: David Jencks <[EMAIL PROTECTED]> > Aside from using cruise control, how does this differ from what chris > kimpton already has set up(http://www.lubega.com/), which runs the > testsuite at least once a day and I think does hourly builds? How does > your proposed system distinguish a broken build from a lengthy multi-part > checkin where writing the comments on the next chunk takes 10 minutes? CruiseControl has a "quiet period" which means that it waits for a developer to stop checking in code before it starts it's build. Once no checkins have been detected for x amount of minutes, then the build starts. I've been the build master for Java projects with 700k loc and 50+ developers all actively checking in and it is extremely rare that the build starts mid checkin. When it does, it usually fixes itself within a few minutes. CruiseControl differs in that it ties the build to checkins, so you can very easily address accountability for broken builds. The web page that is generated will give you a list of files that were checked om. More importantly, there are names attached to those files, so you can very easily see who checked in Foo.java that does not compile or SimpleTest.java that fails. In a big project when tests break, accountability needs to be established and quickly - and this means when, how, and who. Broken windows cannot be tolerated whatsoever. Although, http://lubega.com will tell you what is broken, and approximately when it was broken. It can't link the broken build to a specific checkin and ultimately to the offending party. > > How would your system apportion blame when there are several checkins at > the same time (either interlaced or within less time than it takes to > checkout and do a clean build) and the build breaks? The system cannot be perfect, and tell you exactly who broke what, especially with interlaced checkins. However, these situations are often rare, and continuous integration with CruiseControl will give you enough information that you can figure out what happened and solve the problem within a few minutes. I've been in these situations before, and its never taken me more than 15 minutes to trace through the CruiseControl logs and find the offending developer within a few minutes. Hope this helps. Mike -- __ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
This thread may be indentifying my problem my EJBs are in a jar file, my web client is in a war, and my mbean is in a sar. I want all my deployments to have scoped class loaders. I'm gathering from this discussion: a) The jar can not have a scoped class loader, needs to be an ear b) By default the ?ar files do not use scoped classloaders and it such behavior needs to be turned on in a config file in 3.0.2 Am I close? If b) is true I please for someone to tell me what the option is and where it goes :) Brian Macy Scott M Stark wrote: > In the 3.2RC1 release scoped class loaders will be the default for > ears and sars, maybe wars but you get that with the servlet 2.3 > class loading model. You will be able to configure shared class loader > repositories across deployments if desired. > > > Scott Stark > Chief Technology Officer > JBoss Group, LLC > > > - Original Message - > From: "David Jencks" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Wednesday, September 18, 2002 7:09 AM > Subject: Re: [JBoss-dev] ClassLoader frustration > > > >>I set up the ulr stuff in 3.2 so it would be really easy to extend to other >>module types. Go ahead if you would like it for sars. In 3.0 branch it is >>a little harder. I didn't feel really pressed to do this since you can put >>any kind of package you want in an .ear and get the same effect. > > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
> I have just recently started down the cruise control path on > a personal > project, and my company has since dumped it in favor of an > Austin-based > startup (Buildforge) which offers better support and feature set. > Basically, I have found that if you don't use the exact version of Ant > and scripts that they use, you won't get the behavior as > expected. Thus, > you end up doing a lot of your own coding and config work to > make cruise > control happy. Some is acceptable, but we practically rewrote > the thing > and some of our ant scripts to get it all to play nice. We must have been lucky I guess :-). We're using version 1.2.1a and all we've had to do is write our own label incrementer. I'll watch out for this in future though. Thanks, Kev Kevin Conner Orchard Information Systems Limited Newcastle Technopole, Kings Manor Newcastle Upon Tyne, NE1 6PA. United Kingdom Registered in England, Number 1900078 Tel: +44 (0) 191-2032536 Fax: +44 (0) 191 2302515 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610988 ] Gangster-Example and MSSQL 2000: error
Bugs item #610988, was opened at 2002-09-18 03:40 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610988&group_id=22866 Category: JBossCMP Group: v3.0 Rabbit Hole >Status: Closed >Resolution: Duplicate Priority: 5 Submitted By: Halil-C. Gürsoy (che---) >Assigned to: Dain Sundstrom (dsundstrom) Summary: Gangster-Example and MSSQL 2000: error Initial Comment: Hi Folks, while testing a bit with my local JBoss 3.0.2/Catalina 4.0.4 and a MSSQL 2000 installation i trapped into an error. I have installed the gangster cmp 2.0 example and deployed it with the build file. During the deployment an SQLException occured (see attached log file) after some tables are created. Details: - Win 2k prof - JDK 1.4.0 - JBoss 3.0.2 / Tomcat 4.0.4-Bundle - MS SQL Server 2000 (on a another server, 8.00.194) - M$ JDBC Drivers dated to 19.04.2002 - error log: see attached file -- >Comment By: Dain Sundstrom (dsundstrom) Date: 2002-09-18 10:31 Message: Logged In: YES user_id=251431 Duplicate of [ 604916 ] JBoss.3.0TemplateAndExample w/ sqlserv2k -- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610988&group_id=22866 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
>From what I've seen of CruiseControl, I think it's a useful tool, and if Michael or someone else wants to set it up, I'll use it. As far as I can tell, both the lubega and cruise control systems operate remotely. I don't see any reason why they can't coexist. Nothing wrong with a little diversity, right? - Matt -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Kevin Conner Sent: Wednesday, September 18, 2002 11:09 AM To: '[EMAIL PROTECTED]' Subject: RE: [JBoss-dev] Continuous Integration > Right, a build initiated on checkin is in general guarenteed > to fail due > to an inconsistent view of the repository. Start bitching more about > specific problems due to checkins that developers are not fixing. If > it continues unabated the offending developer is gone. It's not initiated on checkin though, it polls the CVS repository for changes and will initiate the build once the changes have settled down. This is all configurable in cruisecontrol. If the build fails then cruisecontrol will email the people who checked in code since the previous checkout. The idea is that this should be frequent enough to catch small amounts of changes which cause problems. I've introduced cruisecontrol here so that we can automate builds and testing. It has been good for us as we are a small team. Having said all that, I'm not sure that there is any benefit of using this in jboss if there is already a mechanism to build on a regular interval (which there is). If you were starting from scratch then I would certainly recommend cruisecontrol but why change if your current mechanism is sufficient? Kev Kevin Conner Orchard Information Systems Limited Newcastle Technopole, Kings Manor Newcastle Upon Tyne, NE1 6PA. United Kingdom Registered in England, Number 1900078 Tel: +44 (0) 191-2032536 Fax: +44 (0) 191 2302515 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
In the 3.2RC1 release scoped class loaders will be the default for ears and sars, maybe wars but you get that with the servlet 2.3 class loading model. You will be able to configure shared class loader repositories across deployments if desired. Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "David Jencks" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 7:09 AM Subject: Re: [JBoss-dev] ClassLoader frustration > > I set up the ulr stuff in 3.2 so it would be really easy to extend to other > module types. Go ahead if you would like it for sars. In 3.0 branch it is > a little harder. I didn't feel really pressed to do this since you can put > any kind of package you want in an .ear and get the same effect. --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
I have just recently started down the cruise control path on a personal project, and my company has since dumped it in favor of an Austin-based startup (Buildforge) which offers better support and feature set. Basically, I have found that if you don't use the exact version of Ant and scripts that they use, you won't get the behavior as expected. Thus, you end up doing a lot of your own coding and config work to make cruise control happy. Some is acceptable, but we practically rewrote the thing and some of our ant scripts to get it all to play nice. On my plate still is a task to try to use their modification-set tag and from a 10 min cron job, run a target that uses that tag to determine if a build is needed. If not, it will just return quietly. Hokey, but I haven't found anything better out there unless I just use their code to create something better. I'll look into that other link that was sent out - lubega.com was it? James > -Original Message- > From: Kevin Conner [mailto:[EMAIL PROTECTED]] > Sent: Wednesday, September 18, 2002 10:09 AM > To: '[EMAIL PROTECTED]' > Subject: RE: [JBoss-dev] Continuous Integration > > > > Right, a build initiated on checkin is in general guarenteed > > to fail due > > to an inconsistent view of the repository. Start bitching more about > > specific problems due to checkins that developers are not fixing. If > > it continues unabated the offending developer is gone. > > It's not initiated on checkin though, it polls the CVS > repository for changes and will initiate the build once the > changes have settled down. This is all configurable in cruisecontrol. > > If the build fails then cruisecontrol will email the people > who checked in code since the previous checkout. The idea is > that this should be frequent enough to catch small amounts of > changes which cause problems. > > I've introduced cruisecontrol here so that we can automate > builds and testing. It has been good for us as we are a small team. > > Having said all that, I'm not sure that there is any benefit > of using this in jboss if there is already a mechanism to > build on a regular interval (which there is). If you were > starting from scratch then I would certainly recommend > cruisecontrol but why change if your current mechanism is sufficient? > > Kev > > Kevin Conner > Orchard Information Systems Limited > Newcastle Technopole, Kings Manor > Newcastle Upon Tyne, NE1 6PA. United Kingdom > Registered in England, Number 1900078 > Tel: +44 (0) 191-2032536 Fax: +44 (0) 191 2302515 > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the > experts on Hammer Technology! Open Source & Linux Developers, > register now for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > > ___ > Jboss-development mailing list [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] Continuous Integration
> Right, a build initiated on checkin is in general guarenteed > to fail due > to an inconsistent view of the repository. Start bitching more about > specific problems due to checkins that developers are not fixing. If > it continues unabated the offending developer is gone. It's not initiated on checkin though, it polls the CVS repository for changes and will initiate the build once the changes have settled down. This is all configurable in cruisecontrol. If the build fails then cruisecontrol will email the people who checked in code since the previous checkout. The idea is that this should be frequent enough to catch small amounts of changes which cause problems. I've introduced cruisecontrol here so that we can automate builds and testing. It has been good for us as we are a small team. Having said all that, I'm not sure that there is any benefit of using this in jboss if there is already a mechanism to build on a regular interval (which there is). If you were starting from scratch then I would certainly recommend cruisecontrol but why change if your current mechanism is sufficient? Kev Kevin Conner Orchard Information Systems Limited Newcastle Technopole, Kings Manor Newcastle Upon Tyne, NE1 6PA. United Kingdom Registered in England, Number 1900078 Tel: +44 (0) 191-2032536 Fax: +44 (0) 191 2302515 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
On 2002.09.18 10:28:19 -0400 Peter Antman wrote: > On 18 Sep, David Jencks wrote: > > On 2002.09.18 09:33:40 -0400 Peter Antman wrote: > >> Hi, i am sorry if I have slept under a stone the last 6 month or so, > but > >> I have to say I find the new classloader aproach somewhat frustrating > >> (if I am not wrong in what follows). > >> > >> At least in 3.0 the following scenario (among others) would apply: > >> > >> Say we wanted to integrate two system services (as mbeans). > >> > >> a) Both uses a common library, but of differents versions. > >> b) Both uses a lib that also JBoss uses, but of later version. > >> > >> What happens in this case with the current CL aproach, well. > >> > >> b) is impossible if you do not change the lib used by JBoss (speek > about > >> cost of integration to verify that that works). > >> > >> a) is stocastic. The Mbean getting deployed first will rule wich > classes > >> are loaded and cached globaly. > >> > >> As far as I have found out the only type that do not have this problem > >> is *.ear when used with the Heir...Repository. But no other packages > >> types has this option. > >> > >> Would it not be good to, either > >> 1) Have the same optional Heir..Rep for at least SAR modules, so they > >> can load local classes that are different from global ones. > > > > I set up the ulr stuff in 3.2 so it would be really easy to extend to > other > > module types. Go ahead if you would like it for sars. In 3.0 branch > it is > > a little harder. I didn't feel really pressed to do this since you can > put > > any kind of package you want in an .ear and get the same effect. > > Ok. Just seams a little waird to deploy server components inside emtry > sars. But ok, that is maybe the way to go. I agree, but at least there is some way to do it. > > I will starts looking into 3.2 when I have finalized som more of my > stuff. ulr - what is that? ;-) unified loader repository. I should have said loader repository. > > > >> > >> or > >> 2) Change the loading algoritms so its looks something like this: > >> > >> - Check cache connected to classloader > >> - Check local classloader (cache localy, and globaly if its not there) > >> - Check global cache > >> - Check global repository > >> > >> The later does, however, not fully solve the problem with dependant > jars > >> embedded in a sar, since - if I understand it correctly - they are > >> loaded as equal sitizens of the global repository. Here I guess only > the > >> HeirarchicalLoaderRepository2 is a solution (but with that aproach > >> classes will not become available globaly, only in the Heir...Rep). > > > > I think this is apt to cause more problems than it solves. To solve > the "I > > want a different version of a lib used internally by jboss" you'd have > to > > find a way of making that part of jboss use a non-default hier.loader > > repository. > > ? > > I guess it will work out of the box with the Heir..Rep, since classes > are loaded locally firts. > > >This is presumably possible but sounds like a big hassle, in > > particular to maintain and test continually. In this case, didn't we > have > > the same problem in 2.4? > > Yea I know, which is part of my disapointment. I reallt thought that hod > deploy of server components would solve the hazzle of beeing forced to > have all your custom libs in lib/ext and overwriting jboss system libs > and restarting the server. In the 2.4 line you do not just install > JBoss, you have to install your application with a specific JBoss ;-). > > > Which library are you having problems with in > > particular? > > In this case its concurrent.jar, but I have not yet checked which > version is which. I see it mostly as a symptom. (I think) the jars in lib rather than server/[conf]/lib are not going to be versionable since they are used in jboss startup to set up the versioning mechanism. Stuff in server/[conf]/lib should be versionable by deploying it in a .sar, although you may have problems with deployment order. I wasn't aware that there had been any changes in concurrent.jar for years... can you provide details, we should probably upgrade. thanks david jencks > > Thanks > > //Peter > > > > thanks > > david jencks > > > >> > >> > >> Or have I gone completely nuts? > >> > >> //Peter > >> -- > >> > >> Peter Antman Chief Systems Architect, Business Development > >> Technology in Media, Box 34105 100 26 Stockholm > >> WWW: http://www.tim.se WWW: http://www.backsource.org > >> Email: [EMAIL PROTECTED] > >> Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 > >> > >> > >> > >> > >> --- > >> This SF.NET email is sponsored by: AMD - Your access to the experts > >> on Hammer Technology! Open Source & Linux Developers, register now > >> for the AMD Developer Symposium.
Re: [JBoss-dev] Continuous Integration
Right, a build initiated on checkin is in general guarenteed to fail due to an inconsistent view of the repository. Start bitching more about specific problems due to checkins that developers are not fixing. If it continues unabated the offending developer is gone. Scott Stark Chief Technology Officer JBoss Group, LLC - Original Message - From: "David Jencks" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Wednesday, September 18, 2002 7:26 AM Subject: Re: [JBoss-dev] Continuous Integration > Aside from using cruise control, how does this differ from what chris > kimpton already has set up(http://www.lubega.com/), which runs the > testsuite at least once a day and I think does hourly builds? How does > your proposed system distinguish a broken build from a lengthy multi-part > checkin where writing the comments on the next chunk takes 10 minutes? > > > I'm certainly not averse to a better system, however one of the drawbacks > of the current system is that the results are mostly ignored: people will > often fix problems that break the build the next day, but most people seem > to feel no obligation to do anything about changes that say only break 50 > or 100 tests (ok I'm probably exaggerating, but not by enough). > > > How would your system apportion blame when there are several checkins at > the same time (either interlaced or within less time than it takes to > checkout and do a clean build) and the build breaks? > > Although I might sound negative, I actually think what chris has set up has > been incredibly useful and improvements or a better system would be > wonderful. > > thanks > > david jencks --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610008 ] Verifier too strict for 2.0 cmp 1.1 home
Bugs item #610008, was opened at 2002-09-16 18:29 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610008&group_id=22866 Category: JBossServer Group: v3.2 Status: Open Resolution: Fixed Priority: 5 Submitted By: Scott M Stark (starksm) Assigned to: Christian Riege (lqd) Summary: Verifier too strict for 2.0 cmp 1.1 home Initial Comment: The verifier is validating 2.0 EJBs that use 1.1 cmp as EJB 1.1 beans. This fails when the home includes method forms from the EJB 2.0 spec, which should be allowed. See the current failures to deploy the org.jboss.test.testbean.test.BeanUnitTestCase testbean.jar in either 3.2 or main. -- >Comment By: Christian Riege (lqd) Date: 2002-09-18 16:37 Message: Logged In: YES user_id=176671 Thanks for the clarification. I will see what I can do but this is a bit tricky with the current architecture. I think I have a solution but its not very elegant. Will commit later after I have banged it around a bit. -- Comment By: Scott M Stark (starksm) Date: 2002-09-18 13:52 Message: Logged In: YES user_id=175228 Yes, if you can, verifier the cmp 1.1 bean and if it fails, instead of failing the deployment issue a warning saying that the bean is suspect but allow it to deploy since there are valid constructs they verifier cannot handle currently. -- Comment By: Christian Riege (lqd) Date: 2002-09-18 10:08 Message: Logged In: YES user_id=176671 scott, i'm sorry but i don't get the direction you're leaning to. from my understanding you want the CMP 1.1 Bean to be verified but when there is an error condition encountered, the bean shouild still be marked as valid?! please clarify, i will fix this to your likings. -- Comment By: Scott M Stark (starksm) Date: 2002-09-17 19:00 Message: Logged In: YES user_id=175228 The let's change the verifier error to a warning if a 2.0 DTD is seen and the entity bean is using cmp 1.1 -- Comment By: Christian Riege (lqd) Date: 2002-09-17 17:48 Message: Logged In: YES user_id=176671 CMP 1.x Entity Beans are not checked anymore when encountered in a 2.0 JAR file. This is also noted in the logging infrastructure, although on a DEBUG level (see org.jboss.ejb.EJBDeployer for more info on this). This is certainly not the most elegant solution but its the most viable workaround I can see ATM w/o opening up too many construction sites in the current code. The whole Verifier infrastructure will undergo a rewrite for JBoss 4.0; once that is ready and tested a backport into 3.x Branch should be possible w/o many problems. I just need a quiet weekend to get that done :). Stay tuned. Oh yeah, this is in HEAD and 3.2; does it make sense to bring this into 3.0, too? -- Comment By: Christian Riege (lqd) Date: 2002-09-17 17:09 Message: Logged In: YES user_id=176671 The current Verifier doesn't have a notion of different levels of Spec Violations (i.e. Warning/Error). This is on my TODO for the next generation of the Verifier. I think what I will do is a work-around in the current incarnation. EJB 1.1 Beans inside a 2.0 jar won't get verified at all and a warning message will be issued. -- Comment By: Scott M Stark (starksm) Date: 2002-09-17 16:42 Message: Logged In: YES user_id=175228 The let's change the verifier error to a warning if a 2.0 DTD is seen and the entity bean is using cmp 1.1 -- Comment By: Christian Riege (lqd) Date: 2002-09-17 10:56 Message: Logged In: YES user_id=176671 I will have a look at this but from my memory the current architecture of the Verifier doesn't allow for an easy fix for this -- it's either 2.0 or 1.1 all the way. -- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610008&group_id=22866 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
On 18 Sep, David Jencks wrote: > On 2002.09.18 09:33:40 -0400 Peter Antman wrote: >> Hi, i am sorry if I have slept under a stone the last 6 month or so, but >> I have to say I find the new classloader aproach somewhat frustrating >> (if I am not wrong in what follows). >> >> At least in 3.0 the following scenario (among others) would apply: >> >> Say we wanted to integrate two system services (as mbeans). >> >> a) Both uses a common library, but of differents versions. >> b) Both uses a lib that also JBoss uses, but of later version. >> >> What happens in this case with the current CL aproach, well. >> >> b) is impossible if you do not change the lib used by JBoss (speek about >> cost of integration to verify that that works). >> >> a) is stocastic. The Mbean getting deployed first will rule wich classes >> are loaded and cached globaly. >> >> As far as I have found out the only type that do not have this problem >> is *.ear when used with the Heir...Repository. But no other packages >> types has this option. >> >> Would it not be good to, either >> 1) Have the same optional Heir..Rep for at least SAR modules, so they >> can load local classes that are different from global ones. > > I set up the ulr stuff in 3.2 so it would be really easy to extend to other > module types. Go ahead if you would like it for sars. In 3.0 branch it is > a little harder. I didn't feel really pressed to do this since you can put > any kind of package you want in an .ear and get the same effect. Ok. Just seams a little waird to deploy server components inside emtry sars. But ok, that is maybe the way to go. I will starts looking into 3.2 when I have finalized som more of my stuff. ulr - what is that? ;-) > >> >> or >> 2) Change the loading algoritms so its looks something like this: >> >> - Check cache connected to classloader >> - Check local classloader (cache localy, and globaly if its not there) >> - Check global cache >> - Check global repository >> >> The later does, however, not fully solve the problem with dependant jars >> embedded in a sar, since - if I understand it correctly - they are >> loaded as equal sitizens of the global repository. Here I guess only the >> HeirarchicalLoaderRepository2 is a solution (but with that aproach >> classes will not become available globaly, only in the Heir...Rep). > > I think this is apt to cause more problems than it solves. To solve the "I > want a different version of a lib used internally by jboss" you'd have to > find a way of making that part of jboss use a non-default hier.loader > repository. ? I guess it will work out of the box with the Heir..Rep, since classes are loaded locally firts. >This is presumably possible but sounds like a big hassle, in > particular to maintain and test continually. In this case, didn't we have > the same problem in 2.4? Yea I know, which is part of my disapointment. I reallt thought that hod deploy of server components would solve the hazzle of beeing forced to have all your custom libs in lib/ext and overwriting jboss system libs and restarting the server. In the 2.4 line you do not just install JBoss, you have to install your application with a specific JBoss ;-). > Which library are you having problems with in > particular? In this case its concurrent.jar, but I have not yet checked which version is which. I see it mostly as a symptom. Thanks //Peter > > thanks > david jencks > >> >> >> Or have I gone completely nuts? >> >> //Peter >> -- >> >> Peter Antman Chief Systems Architect, Business Development >> Technology in Media, Box 34105 100 26 Stockholm >> WWW: http://www.tim.se WWW: http://www.backsource.org >> Email: [EMAIL PROTECTED] >> Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 >> >> >> >> >> --- >> This SF.NET email is sponsored by: AMD - Your access to the experts >> on Hammer Technology! Open Source & Linux Developers, register now >> for the AMD Developer Symposium. Code: EX8664 >> http://www.developwithamd.com/developerlab >> ___ >> Jboss-development mailing list >> [EMAIL PROTECTED] >> https://lists.sourceforge.net/lists/listinfo/jboss-development >> >> > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development -- Peter AntmanChief Systems Architect, Business Development Technology in Media, B
Re: [JBoss-dev] Continuous Integration
Aside from using cruise control, how does this differ from what chris kimpton already has set up(http://www.lubega.com/), which runs the testsuite at least once a day and I think does hourly builds? How does your proposed system distinguish a broken build from a lengthy multi-part checkin where writing the comments on the next chunk takes 10 minutes? I'm certainly not averse to a better system, however one of the drawbacks of the current system is that the results are mostly ignored: people will often fix problems that break the build the next day, but most people seem to feel no obligation to do anything about changes that say only break 50 or 100 tests (ok I'm probably exaggerating, but not by enough). How would your system apportion blame when there are several checkins at the same time (either interlaced or within less time than it takes to checkout and do a clean build) and the build breaks? Although I might sound negative, I actually think what chris has set up has been incredibly useful and improvements or a better system would be wonderful. thanks david jencks On 2002.09.18 09:39:37 -0400 Michael Rettig wrote: > Developers, > > I'm looking to setup a continuous integration enviroment for the JBoss > project. Does anyone know of one that exists already? > > If you don't know what continuous integration and what it can do for a > project. You can read more about it: > > http://martinfowler.com/articles/continuousIntegration.html > > I'm planning to setup a box that will detect changes to the CVS repo then > kick off an automated build to do a clean build and test of JBoss. A > public webpage will be generated that will be available to see the > current status of the build, and past builds. > > This process will help eliminate problems caused by developers checking > in code that simply doesn't compile or breaks tests. When the build does > get broken, it only takes a quick look at the web page to figure who is > responsible. > > Also, the build can be configured to automatically email anyone that > breaks the build. > > CruiseControl will be used to build the project and generate the results. > It's another sourceforge OS project hosted on sourceforge. > > http://sourceforge.net/projects/cruisecontrol > > I'm looking for some feedback on this. What do you want the build to do? > Do you want it to email you when the build breaks? What information do > you want displayed on the webpage? > > Any feedback is appreciated. > > Thanks, > > Mike Rettig > -- > __ > Sign-up for your own FREE Personalized E-mail at Mail.com > http://www.mail.com/?sr=signup > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Continuous Integration
- Original Message - I wrote: > I'm looking to setup a continuous integration >enviroment for the JBoss project. >Does anyone know of >one that exists already? So I found the answer to my question at http://lubega.com. However, the continuous integration that I propose will not be based on Chron jobs, but will instead detect developer checkins and kickoff a build as a result. Also, it will be based upon CruiseControl which already has a full build reporting web page frontend. Thanks, Mike -- __ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] ClassLoader frustration
On 2002.09.18 09:33:40 -0400 Peter Antman wrote: > Hi, i am sorry if I have slept under a stone the last 6 month or so, but > I have to say I find the new classloader aproach somewhat frustrating > (if I am not wrong in what follows). > > At least in 3.0 the following scenario (among others) would apply: > > Say we wanted to integrate two system services (as mbeans). > > a) Both uses a common library, but of differents versions. > b) Both uses a lib that also JBoss uses, but of later version. > > What happens in this case with the current CL aproach, well. > > b) is impossible if you do not change the lib used by JBoss (speek about > cost of integration to verify that that works). > > a) is stocastic. The Mbean getting deployed first will rule wich classes > are loaded and cached globaly. > > As far as I have found out the only type that do not have this problem > is *.ear when used with the Heir...Repository. But no other packages > types has this option. > > Would it not be good to, either > 1) Have the same optional Heir..Rep for at least SAR modules, so they > can load local classes that are different from global ones. I set up the ulr stuff in 3.2 so it would be really easy to extend to other module types. Go ahead if you would like it for sars. In 3.0 branch it is a little harder. I didn't feel really pressed to do this since you can put any kind of package you want in an .ear and get the same effect. > > or > 2) Change the loading algoritms so its looks something like this: > > - Check cache connected to classloader > - Check local classloader (cache localy, and globaly if its not there) > - Check global cache > - Check global repository > > The later does, however, not fully solve the problem with dependant jars > embedded in a sar, since - if I understand it correctly - they are > loaded as equal sitizens of the global repository. Here I guess only the > HeirarchicalLoaderRepository2 is a solution (but with that aproach > classes will not become available globaly, only in the Heir...Rep). I think this is apt to cause more problems than it solves. To solve the "I want a different version of a lib used internally by jboss" you'd have to find a way of making that part of jboss use a non-default hier.loader repository. This is presumably possible but sounds like a big hassle, in particular to maintain and test continually. In this case, didn't we have the same problem in 2.4? Which library are you having problems with in particular? thanks david jencks > > > Or have I gone completely nuts? > > //Peter > -- > > Peter Antman Chief Systems Architect, Business Development > Technology in Media, Box 34105 100 26 Stockholm > WWW: http://www.tim.seWWW: http://www.backsource.org > Email: [EMAIL PROTECTED] > Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 > > > > > --- > This SF.NET email is sponsored by: AMD - Your access to the experts > on Hammer Technology! Open Source & Linux Developers, register now > for the AMD Developer Symposium. Code: EX8664 > http://www.developwithamd.com/developerlab > ___ > Jboss-development mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/jboss-development > > --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610907 ] Using JMS in MDB/container-managed TX
Bugs item #610907, was opened at 2002-09-18 03:28
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610907&group_id=22866
Category: JBossMQ
Group: v3.0 Rabbit Hole
Status: Open
Resolution: None
Priority: 5
Submitted By: Elias Ross (genman)
Assigned to: Nobody/Anonymous (nobody)
Summary: Using JMS in MDB/container-managed TX
Initial Comment:
I've gotten MDB to work correctly with rolling-back
the database when I call mdc.setRollbackOnly().
However, I don't know how to create a QueueSession/Sender
which will send its messages after the onMessage() method
exists.
I have bought the JBoss documentation, but it does not have
an example of this case specifically. I suspect (but do not know)
that JBoss does not support automatic QueueSession
commit after an MDB exits its onMessage method. Looking
at the source after a few hours has gotten me confused.
I suspect this is NOT supported, if it is supported,
please forgive this bug submission and point me in
the right direction. This _should_ be documented,
at least document how this is correctly done.
The JMS tutorial:
http://java.sun.com/products/jms/tutorial/
has several examples of Container-managed TX
for MDB, which don't seem to work in JBoss.
Does anyone have an example of this sending out messages:
public class MessageDrivenBean implements MessageListener {
// Container-managed transactions with Required
// transaction attribute
public void onMessage(...) {
try {
QueueConnection qc = qcf.createQueueConnection();
QueueSession qs = qc.createQueueSession(true, 0);
QueueSender qsend = qs.createQueueSender("queue/A");
TextMessage m = qs.createTextMessage();
m.setText("test");
qsend.send(m);
/*
qs.close(); // JMS javadoc says this rollbacks
// In any case, doesn't seem to work
*/
qc.close();
} catch (Exception e) {
e.printStackTrace();
}
}
}
--
Comment By: Corby (corby)
Date: 2002-09-18 14:07
Message:
Logged In: YES
user_id=25032
Your bug submission is correct, Elias. This has been a long-
standing and frustrating issue with JBoss. MDB's with
container-managed transactions do not seem to have the
same transactional semantics as other EJB's, and you MUST
use bean-managed transactions if you want to enlist mutiple
resources in an MDB transaction.
--
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610907&group_id=22866
---
This SF.NET email is sponsored by: AMD - Your access to the experts
on Hammer Technology! Open Source & Linux Developers, register now
for the AMD Developer Symposium. Code: EX8664
http://www.developwithamd.com/developerlab
___
Jboss-development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] Continuous Integration
Developers, I'm looking to setup a continuous integration enviroment for the JBoss project. Does anyone know of one that exists already? If you don't know what continuous integration and what it can do for a project. You can read more about it: http://martinfowler.com/articles/continuousIntegration.html I'm planning to setup a box that will detect changes to the CVS repo then kick off an automated build to do a clean build and test of JBoss. A public webpage will be generated that will be available to see the current status of the build, and past builds. This process will help eliminate problems caused by developers checking in code that simply doesn't compile or breaks tests. When the build does get broken, it only takes a quick look at the web page to figure who is responsible. Also, the build can be configured to automatically email anyone that breaks the build. CruiseControl will be used to build the project and generate the results. It's another sourceforge OS project hosted on sourceforge. http://sourceforge.net/projects/cruisecontrol I'm looking for some feedback on this. What do you want the build to do? Do you want it to email you when the build breaks? What information do you want displayed on the webpage? Any feedback is appreciated. Thanks, Mike Rettig -- __ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
RE: [JBoss-dev] [ jboss-Bugs-600435 ] deadlock detected erroneously
Bill, yes, I read your answers. I really appreciate them. I'm already doing everything you mentioned (read-only methods etc.). My guts say, that the weblogic default setting is, what you usually would expect. (10K$/CPU and no concurrent access would be too bad). Bad performance is still better than having no concurrent access to entity beans. Cheers Georg -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Bill Burke Sent: Friday, September 13, 2002 22:51 To: [EMAIL PROTECTED] Subject: RE: [JBoss-dev] [ jboss-Bugs-600435 ] deadlock detected erroneously Georg, Georg, Georg Did you read my last response to this? 1. You can declare beans read-only. There will be no pessimistic locking. 2. If the bean isn't entirely read-only, you can mark your read-only methods, and read-only methods will not cause a pessimistic lock. Yes, even with the above, this is SERIALIZED transaction isolation, but at the app-server level. Without pessimistic locking, we can't do caching because Entity beans are not required to be java.io.Serializable. If they were, we could copy beans from cache and do cache-versioning tricks. So, you can use Instance Per Transaction to avoid all transactional locking as you are right now. The downside to this is that you are required to use commit 'B' or 'C'. BTW, weblogic uses Instance Per Transaction and commit 'C' by default. Bill ...other messages deleted... --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] ClassLoader frustration
Hi, i am sorry if I have slept under a stone the last 6 month or so, but I have to say I find the new classloader aproach somewhat frustrating (if I am not wrong in what follows). At least in 3.0 the following scenario (among others) would apply: Say we wanted to integrate two system services (as mbeans). a) Both uses a common library, but of differents versions. b) Both uses a lib that also JBoss uses, but of later version. What happens in this case with the current CL aproach, well. b) is impossible if you do not change the lib used by JBoss (speek about cost of integration to verify that that works). a) is stocastic. The Mbean getting deployed first will rule wich classes are loaded and cached globaly. As far as I have found out the only type that do not have this problem is *.ear when used with the Heir...Repository. But no other packages types has this option. Would it not be good to, either 1) Have the same optional Heir..Rep for at least SAR modules, so they can load local classes that are different from global ones. or 2) Change the loading algoritms so its looks something like this: - Check cache connected to classloader - Check local classloader (cache localy, and globaly if its not there) - Check global cache - Check global repository The later does, however, not fully solve the problem with dependant jars embedded in a sar, since - if I understand it correctly - they are loaded as equal sitizens of the global repository. Here I guess only the HeirarchicalLoaderRepository2 is a solution (but with that aproach classes will not become available globaly, only in the Heir...Rep). Or have I gone completely nuts? //Peter -- Peter AntmanChief Systems Architect, Business Development Technology in Media, Box 34105 100 26 Stockholm WWW: http://www.tim.se WWW: http://www.backsource.org Email: [EMAIL PROTECTED] Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: [JBoss-dev] Thinlet and the JBoss Console GUI?
yes, I am looking at it from two angles a client based GUI and also a web application that uses thinlets in an applet. I will post more info and also a working version when I get a little more of it functional. >>> [EMAIL PROTECTED] 09/17/02 06:35PM >>>Has anyone looked into Thinlet (http://www.thinlet.com) as a possiblewidget set for a GUI console?Looks promising... fast... but I haven't looked into it very deeply.--jason---This SF.NET email is sponsored by: AMD - Your access to the expertson Hammer Technology! Open Source & Linux Developers, register nowfor the AMD Developer Symposium. Code: EX8664http://www.developwithamd.com/developerlab___Jboss-development mailing list[EMAIL PROTECTED]https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610843 ] Caching of JaasSecurityManager.DomainInf
Bugs item #610843, was opened at 2002-09-17 16:15
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610843&group_id=22866
Category: JBossSX
Group: v3.0 Rabbit Hole
>Status: Closed
Resolution: Out of Date
Priority: 5
Submitted By: Timo Warns (timow)
Assigned to: Scott M Stark (starksm)
Summary: Caching of JaasSecurityManager.DomainInf
Initial Comment:
We had problems with JaasSecurityManager. Sometimes it
returned wrong values for "isUserInRole".
Here is what we believe causes the problem:
JaasSecurityManager uses (as default)
TimedCachePolicy.get(Object) as
method to access cached JaasSecurityManager.DomainInfo
entries.
TimedCachePolicy.get(Object) checks whether entries
have expired and tries to
refresh them if this is the case. If refreshing fails,
the entry will be
removed and "null" will be returned.
The problem is that DomainInfo is not able to refresh
(refresh() just returns "false"). Thus if a DomainInfo
is expired and JaasSecurityManager.doesUserHaveRole(...)
is called, it will return "false", even if "true" would
be correct.
Sorry, if we misunderstood anything!
--
>Comment By: Scott M Stark (starksm)
Date: 2002-09-18 04:57
Message:
Logged In: YES
user_id=175228
I don't know where you are getting the 3.2 source as this is
not what I see.
@version $Revision: 1.27.2.1 $
public boolean doesUserHaveRole(Principal principal, Set
rolePrincipals)
{
boolean hasRole = false;
Subject subject = getActiveSubject();
if( subject != null )
{
DomainInfo info = getCacheInfo(principal, false);
...
private DomainInfo getCacheInfo(Principal principal,
boolean allowRefresh)
{
if( domainCache == null )
return null;
DomainInfo cacheInfo = null;
synchronized( domainCache )
{
if( allowRefresh == true )
cacheInfo = (DomainInfo) domainCache.get(principal);
else
cacheInfo = (DomainInfo) domainCache.peek
(principal);
}
return cacheInfo;
}
--
Comment By: Timo Warns (timow)
Date: 2002-09-18 01:51
Message:
Logged In: YES
user_id=606328
I just downloaded newest sources of JBoss-3.2, but the bug
still seems
to be present (as far as we understand):
In JaasSecurityManager (snipped to relevant parts):
[...]
public static class DomainInfo implements
TimedCachePolicy.TimedEntry
[...]
// refreshing always fails
public boolean refresh()
{
return false;
}
[...]
public boolean doesUserHaveRole(Principal principal, Set
rolePrincipals)
{
boolean hasRole = false;
[...]
// if entry is expired, "null" will be returned
DomainInfo info = getCacheInfo(principal);
Group roles = null;
if( info != null )
roles = info.roles;
if( roles != null )
[...]
return hasRole;
}
[...]
And in TimedCachePolicy:
[...]
public Object get(Object key)
{
[...]
if( entry.isCurrent(now) == false )
{ // Try to refresh the entry
// DomainInfo-entries cannot be refreshed
if( entry.refresh() == false )
{ // Failed, remove the entry and return null
entry.destroy();
entryMap.remove(key);
return null;
}
}
[...]
Thanks for your help!
--
Comment By: Scott M Stark (starksm)
Date: 2002-09-17 16:21
Message:
Logged In: YES
user_id=175228
You are looking at out of date code. Since 3.0.1 only
authentication requests will flush the cache and therefore
roles are always consistent with the authenticated user.
--
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610843&group_id=22866
---
This SF.NET email is sponsored by: AMD - Your access to the experts
on Hammer Technology! Open Source & Linux Developers, register now
for the AMD Developer Symposium. Code: EX8664
http://www.developwithamd.com/developerlab
___
Jboss-development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610008 ] Verifier too strict for 2.0 cmp 1.1 home
Bugs item #610008, was opened at 2002-09-16 09:29 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610008&group_id=22866 Category: JBossServer Group: v3.2 Status: Open Resolution: Fixed Priority: 5 Submitted By: Scott M Stark (starksm) Assigned to: Christian Riege (lqd) Summary: Verifier too strict for 2.0 cmp 1.1 home Initial Comment: The verifier is validating 2.0 EJBs that use 1.1 cmp as EJB 1.1 beans. This fails when the home includes method forms from the EJB 2.0 spec, which should be allowed. See the current failures to deploy the org.jboss.test.testbean.test.BeanUnitTestCase testbean.jar in either 3.2 or main. -- >Comment By: Scott M Stark (starksm) Date: 2002-09-18 04:52 Message: Logged In: YES user_id=175228 Yes, if you can, verifier the cmp 1.1 bean and if it fails, instead of failing the deployment issue a warning saying that the bean is suspect but allow it to deploy since there are valid constructs they verifier cannot handle currently. -- Comment By: Christian Riege (lqd) Date: 2002-09-18 01:08 Message: Logged In: YES user_id=176671 scott, i'm sorry but i don't get the direction you're leaning to. from my understanding you want the CMP 1.1 Bean to be verified but when there is an error condition encountered, the bean shouild still be marked as valid?! please clarify, i will fix this to your likings. -- Comment By: Scott M Stark (starksm) Date: 2002-09-17 10:00 Message: Logged In: YES user_id=175228 The let's change the verifier error to a warning if a 2.0 DTD is seen and the entity bean is using cmp 1.1 -- Comment By: Christian Riege (lqd) Date: 2002-09-17 08:48 Message: Logged In: YES user_id=176671 CMP 1.x Entity Beans are not checked anymore when encountered in a 2.0 JAR file. This is also noted in the logging infrastructure, although on a DEBUG level (see org.jboss.ejb.EJBDeployer for more info on this). This is certainly not the most elegant solution but its the most viable workaround I can see ATM w/o opening up too many construction sites in the current code. The whole Verifier infrastructure will undergo a rewrite for JBoss 4.0; once that is ready and tested a backport into 3.x Branch should be possible w/o many problems. I just need a quiet weekend to get that done :). Stay tuned. Oh yeah, this is in HEAD and 3.2; does it make sense to bring this into 3.0, too? -- Comment By: Christian Riege (lqd) Date: 2002-09-17 08:09 Message: Logged In: YES user_id=176671 The current Verifier doesn't have a notion of different levels of Spec Violations (i.e. Warning/Error). This is on my TODO for the next generation of the Verifier. I think what I will do is a work-around in the current incarnation. EJB 1.1 Beans inside a 2.0 jar won't get verified at all and a warning message will be issued. -- Comment By: Scott M Stark (starksm) Date: 2002-09-17 07:42 Message: Logged In: YES user_id=175228 The let's change the verifier error to a warning if a 2.0 DTD is seen and the entity bean is using cmp 1.1 -- Comment By: Christian Riege (lqd) Date: 2002-09-17 01:56 Message: Logged In: YES user_id=176671 I will have a look at this but from my memory the current architecture of the Verifier doesn't allow for an easy fix for this -- it's either 2.0 or 1.1 all the way. -- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610008&group_id=22866 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-611044 ] JBoss+Jetty: redirect bug on welcomefile
Bugs item #611044, was opened at 2002-09-18 13:36
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=611044&group_id=22866
Category: JBossServer
Group: v3.0 Rabbit Hole
Status: Open
Resolution: None
Priority: 5
Submitted By: Michael Pichler (mpichler)
Assigned to: Nobody/Anonymous (nobody)
Summary: JBoss+Jetty: redirect bug on welcomefile
Initial Comment:
JSP sendRedirect into subdirectory broken from JSP welcome-file
for JBoss 3.0.2 with Jetty.
Take a webapplication (the attached hello.war) that defines a
welcome-file Hello.jsp and page Hello.jsp does a redirect to a
a relative URL:
<% response.sendRedirect ("subdir/target.html"); %>
This works with Jetty and Tomcat alone, and also with
JBoss 3.0.2 with Tomcat 4.0.4, i.e. the redirect is performed
relative to the actual location of Hello.jsp. (The difference
is that tomcat uses a client redirect for the welcome page whereas
jetty directly serves the welcome page under the original URL).
When using JBoss 3.0.2 with Jetty 1.4 and accessing just
the URL of the webapplication (http://hostname:8080/hello/),
the redirect is done relative to a wrong base URL
and the wepappliction part disappears from the forwarded URL.
When accessing the welcome-file explicitly, it works in any case.
See also http://www.jboss.org/forums/thread.jsp?forum=50&thread=20933
--- JBoss+Jetty ---
GET /hello/ HTTP/1.0
HTTP/1.1 302 Moved Temporarily
Server: Jetty/RELEASE_MAJOR_MINOR (Linux 2.4.18-64GB-SMP i386)
Servlet-Engine: Jetty/RELEASE_MAJOR_MINOR (Servlet 2.3; JSP 1.2; java 1.4.0)
Location: http://192.168.x.y:8080/subdir/target.html
^--- web context /hello/ lost
Server Info: Jetty/RELEASE_MAJOR_MINOR
Request URI: /hello/Hello.jsp
Full URL: http://192.168.x.y:8080/hello/
--- JBoss+Tomcat ---
GET /hello/ HTTP/1.0
HTTP/1.1 302 Moved Temporarily
Location: http://192.168.x.y:8080/hello/Hello.jsp
(redirect to the host defined as defaultHost in tomcat4-service.xml)
GET /hello/Hello.jsp HTTP/1.0
HTTP/1.1 302 Moved Temporarily
Location: http://192.168.x.y:8080/hello/subdir/target.html
^--- OK
(contents of Hello.jsp not served by tomcat)
--
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=611044&group_id=22866
---
This SF.NET email is sponsored by: AMD - Your access to the experts
on Hammer Technology! Open Source & Linux Developers, register now
for the AMD Developer Symposium. Code: EX8664
http://www.developwithamd.com/developerlab
___
Jboss-development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development
AW: AW: [JBoss-dev] jboss.net and catalina
If I´m not totally out of mind, yes. CGJ -Ursprüngliche Nachricht- Von: Bruce Scharlau [mailto:[EMAIL PROTECTED]] Gesendet: Mittwoch, 18. September 2002 12:17 An: [EMAIL PROTECTED] Betreff: Re: AW: [JBoss-dev] jboss.net and catalina At 11:58 18/09/2002 +0200, you wrote: >Bruce, which jboss version do you refer to. It works for me from >3.2Beta on even in an exploded fashion. > >CGJ I'm building it all from jboss-all in cvs, ie jboss4.0.x So you got it going ok with the jboss-tomcat in 3.2 then? cheers, Bruce Dr. Bruce Scharlau Dept. of Computing Science University of Aberdeen Aberdeen AB24 3UE 01224 272193 http://www.csd.abdn.ac.uk/~bscharla mailto:[EMAIL PROTECTED] --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development ### This message has been scanned by F-Secure Anti-Virus for Microsoft Exchange. For more information, connect to http://www.F-Secure.com/ --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
Re: AW: [JBoss-dev] jboss.net and catalina
At 11:58 18/09/2002 +0200, you wrote: >Bruce, which jboss version do you refer to. It works for me from 3.2Beta on >even in an >exploded fashion. > >CGJ I'm building it all from jboss-all in cvs, ie jboss4.0.x So you got it going ok with the jboss-tomcat in 3.2 then? cheers, Bruce Dr. Bruce Scharlau Dept. of Computing Science University of Aberdeen Aberdeen AB24 3UE 01224 272193 http://www.csd.abdn.ac.uk/~bscharla mailto:[EMAIL PROTECTED] --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
AW: [JBoss-dev] jboss.net and catalina
Bruce, which jboss version do you refer to. It works for me from 3.2Beta on even in an exploded fashion. CGJ -Ursprüngliche Nachricht- Von: Bruce Scharlau [mailto:[EMAIL PROTECTED]] Gesendet: Mittwoch, 18. September 2002 11:31 An: [EMAIL PROTECTED] Betreff: [JBoss-dev] jboss.net and catalina Hello all, jboss.net with axis rc1 now seems to work with jboss and tomcat (catalina 4.0.4) Thanks Scott for catching the error with EmbeddedCatalinaSX! IF (and it seem only IF) you wrap the jboss-net.sar in an exploded ear file. Do this: 1. create unpacked jboss-net.ear 2. put jboss-net.sar into this 3. move jboss-net.war and jboss-net.wsr from under jboss-net.sar up a directory to be under jboss-net.ear 4. under jboss-net.ear create a META-INF directory with application.xml that contains this: JBoss-Example WebService jboss-net.sar jboss-net.war /jboss-net jboss-net.wsr As soon as I get a chance I'll update the jboss-net howto page at: http://www.csd.abdn.ac.uk/%7Ebscharla/soap.html and the accompanying example pages. cheers, Bruce Dr. Bruce Scharlau Dept. of Computing Science University of Aberdeen Aberdeen AB24 3UE 01224 272193 http://www.csd.abdn.ac.uk/~bscharla mailto:[EMAIL PROTECTED] --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development ### This message has been scanned by F-Secure Anti-Virus for Microsoft Exchange. For more information, connect to http://www.F-Secure.com/ --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] jboss.net and catalina
Hello all, jboss.net with axis rc1 now seems to work with jboss and tomcat (catalina 4.0.4) Thanks Scott for catching the error with EmbeddedCatalinaSX! IF (and it seem only IF) you wrap the jboss-net.sar in an exploded ear file. Do this: 1. create unpacked jboss-net.ear 2. put jboss-net.sar into this 3. move jboss-net.war and jboss-net.wsr from under jboss-net.sar up a directory to be under jboss-net.ear 4. under jboss-net.ear create a META-INF directory with application.xml that contains this: JBoss-Example WebService jboss-net.sar jboss-net.war /jboss-net jboss-net.wsr As soon as I get a chance I'll update the jboss-net howto page at: http://www.csd.abdn.ac.uk/%7Ebscharla/soap.html and the accompanying example pages. cheers, Bruce Dr. Bruce Scharlau Dept. of Computing Science University of Aberdeen Aberdeen AB24 3UE 01224 272193 http://www.csd.abdn.ac.uk/~bscharla mailto:[EMAIL PROTECTED] --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610843 ] Caching of JaasSecurityManager.DomainInf
Bugs item #610843, was opened at 2002-09-18 01:15
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610843&group_id=22866
Category: JBossSX
Group: v3.0 Rabbit Hole
>Status: Open
Resolution: Out of Date
Priority: 5
Submitted By: Timo Warns (timow)
Assigned to: Scott M Stark (starksm)
Summary: Caching of JaasSecurityManager.DomainInf
Initial Comment:
We had problems with JaasSecurityManager. Sometimes it
returned wrong values for "isUserInRole".
Here is what we believe causes the problem:
JaasSecurityManager uses (as default)
TimedCachePolicy.get(Object) as
method to access cached JaasSecurityManager.DomainInfo
entries.
TimedCachePolicy.get(Object) checks whether entries
have expired and tries to
refresh them if this is the case. If refreshing fails,
the entry will be
removed and "null" will be returned.
The problem is that DomainInfo is not able to refresh
(refresh() just returns "false"). Thus if a DomainInfo
is expired and JaasSecurityManager.doesUserHaveRole(...)
is called, it will return "false", even if "true" would
be correct.
Sorry, if we misunderstood anything!
--
>Comment By: Timo Warns (timow)
Date: 2002-09-18 10:51
Message:
Logged In: YES
user_id=606328
I just downloaded newest sources of JBoss-3.2, but the bug
still seems
to be present (as far as we understand):
In JaasSecurityManager (snipped to relevant parts):
[...]
public static class DomainInfo implements
TimedCachePolicy.TimedEntry
[...]
// refreshing always fails
public boolean refresh()
{
return false;
}
[...]
public boolean doesUserHaveRole(Principal principal, Set
rolePrincipals)
{
boolean hasRole = false;
[...]
// if entry is expired, "null" will be returned
DomainInfo info = getCacheInfo(principal);
Group roles = null;
if( info != null )
roles = info.roles;
if( roles != null )
[...]
return hasRole;
}
[...]
And in TimedCachePolicy:
[...]
public Object get(Object key)
{
[...]
if( entry.isCurrent(now) == false )
{ // Try to refresh the entry
// DomainInfo-entries cannot be refreshed
if( entry.refresh() == false )
{ // Failed, remove the entry and return null
entry.destroy();
entryMap.remove(key);
return null;
}
}
[...]
Thanks for your help!
--
Comment By: Scott M Stark (starksm)
Date: 2002-09-18 01:21
Message:
Logged In: YES
user_id=175228
You are looking at out of date code. Since 3.0.1 only
authentication requests will flush the cache and therefore
roles are always consistent with the authenticated user.
--
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610843&group_id=22866
---
This SF.NET email is sponsored by: AMD - Your access to the experts
on Hammer Technology! Open Source & Linux Developers, register now
for the AMD Developer Symposium. Code: EX8664
http://www.developwithamd.com/developerlab
___
Jboss-development mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610988 ] Gangster-Example and MSSQL 2000: error
Bugs item #610988, was opened at 2002-09-18 10:40 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610988&group_id=22866 Category: JBossCMP Group: v3.0 Rabbit Hole Status: Open Resolution: None Priority: 5 Submitted By: Halil-C. Gürsoy (che---) Assigned to: Nobody/Anonymous (nobody) Summary: Gangster-Example and MSSQL 2000: error Initial Comment: Hi Folks, while testing a bit with my local JBoss 3.0.2/Catalina 4.0.4 and a MSSQL 2000 installation i trapped into an error. I have installed the gangster cmp 2.0 example and deployed it with the build file. During the deployment an SQLException occured (see attached log file) after some tables are created. Details: - Win 2k prof - JDK 1.4.0 - JBoss 3.0.2 / Tomcat 4.0.4-Bundle - MS SQL Server 2000 (on a another server, 8.00.194) - M$ JDBC Drivers dated to 19.04.2002 - error log: see attached file -- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610988&group_id=22866 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-610008 ] Verifier too strict for 2.0 cmp 1.1 home
Bugs item #610008, was opened at 2002-09-16 18:29 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610008&group_id=22866 Category: JBossServer Group: v3.2 Status: Open Resolution: Fixed Priority: 5 Submitted By: Scott M Stark (starksm) Assigned to: Christian Riege (lqd) Summary: Verifier too strict for 2.0 cmp 1.1 home Initial Comment: The verifier is validating 2.0 EJBs that use 1.1 cmp as EJB 1.1 beans. This fails when the home includes method forms from the EJB 2.0 spec, which should be allowed. See the current failures to deploy the org.jboss.test.testbean.test.BeanUnitTestCase testbean.jar in either 3.2 or main. -- >Comment By: Christian Riege (lqd) Date: 2002-09-18 10:08 Message: Logged In: YES user_id=176671 scott, i'm sorry but i don't get the direction you're leaning to. from my understanding you want the CMP 1.1 Bean to be verified but when there is an error condition encountered, the bean shouild still be marked as valid?! please clarify, i will fix this to your likings. -- Comment By: Scott M Stark (starksm) Date: 2002-09-17 19:00 Message: Logged In: YES user_id=175228 The let's change the verifier error to a warning if a 2.0 DTD is seen and the entity bean is using cmp 1.1 -- Comment By: Christian Riege (lqd) Date: 2002-09-17 17:48 Message: Logged In: YES user_id=176671 CMP 1.x Entity Beans are not checked anymore when encountered in a 2.0 JAR file. This is also noted in the logging infrastructure, although on a DEBUG level (see org.jboss.ejb.EJBDeployer for more info on this). This is certainly not the most elegant solution but its the most viable workaround I can see ATM w/o opening up too many construction sites in the current code. The whole Verifier infrastructure will undergo a rewrite for JBoss 4.0; once that is ready and tested a backport into 3.x Branch should be possible w/o many problems. I just need a quiet weekend to get that done :). Stay tuned. Oh yeah, this is in HEAD and 3.2; does it make sense to bring this into 3.0, too? -- Comment By: Christian Riege (lqd) Date: 2002-09-17 17:09 Message: Logged In: YES user_id=176671 The current Verifier doesn't have a notion of different levels of Spec Violations (i.e. Warning/Error). This is on my TODO for the next generation of the Verifier. I think what I will do is a work-around in the current incarnation. EJB 1.1 Beans inside a 2.0 jar won't get verified at all and a warning message will be issued. -- Comment By: Scott M Stark (starksm) Date: 2002-09-17 16:42 Message: Logged In: YES user_id=175228 The let's change the verifier error to a warning if a 2.0 DTD is seen and the entity bean is using cmp 1.1 -- Comment By: Christian Riege (lqd) Date: 2002-09-17 10:56 Message: Logged In: YES user_id=176671 I will have a look at this but from my memory the current architecture of the Verifier doesn't allow for an easy fix for this -- it's either 2.0 or 1.1 all the way. -- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=610008&group_id=22866 --- This SF.NET email is sponsored by: AMD - Your access to the experts on Hammer Technology! Open Source & Linux Developers, register now for the AMD Developer Symposium. Code: EX8664 http://www.developwithamd.com/developerlab ___ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
[JBoss-dev] [ jboss-Bugs-562972 ] Error in PortableRemoteObject.narrow()
Bugs item #562972, was opened at 2002-05-31 09:30
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=376685&aid=562972&group_id=22866
Category: JBossServer
Group: v3.0 Rabbit Hole
Status: Open
Resolution: None
Priority: 8
Submitted By: Marius Kotsbak (mkotsbak)
Assigned to: Nobody/Anonymous (nobody)
Summary: Error in PortableRemoteObject.narrow()
Initial Comment:
18:30:30,517 ERROR [STDERR] java.lang.ClassCastException
18:30:30,518 ERROR [STDERR] at
com.sun.corba.se.internal.javax.rmi.PortableRemoteObject.narrow(PortableRemoteObject.java:293)
18:30:30,519 ERROR [STDERR] at
javax.rmi.PortableRemoteObject.narrow(PortableRemoteObject.java:134)
When running narrow after lookup from a servlet.
--
Comment By: Nobody/Anonymous (nobody)
Date: 2002-09-18 00:54
Message:
Logged In: NO
I investigated the problem as it arise on jboss-3.0.1_tomcat-4.0.4
on Win2000.
I use the following code to look-up a local home interface from a
servlet deployed in the same ear as my ejbs:
--- code begin ---
logger.debug("SomeEJBHome was loaded by: " +
SomeEJBHome.class.getClassLoader().toString());
InitialContext ic = new InitialContext();
Context ejbCtx = (Context) ic.lookup("java:comp/env/ejb");
Object o = ejbCtx.lookup(ejbRefName);
Class[] intfs = o.getClass().getInterfaces();
logger.debug("intf : " + intfs[0].toString() + " was
loaded by: " + intfs[0].getClassLoader().toString());
home = (SomeEJBHome)o;
--- code end ---
Log before hot redeploy:
12:41:30,866 DEBUG [SomeEJBFactory] SomeEJBHome was
loaded by:org.jboss.mx.loading.UnifiedClassLoader@2b9f14
{url=file:...
12:41:31,067 DEBUG [SomeEJBFactory] intf : interface
SomeEJBHome was loaded by:
org.jboss.mx.loading.UnifiedClassLoader@2b9f14{url=file:...
Log after hot redeploy:
12:42:37,142 DEBUG [SomeEJBFactory] SomeEJBHome was
loaded by: org.jboss.mx.loading.UnifiedClassLoader@c0d0a8 {
url=file:...
12:42:37,212 DEBUG [SomeEJBFactory] intf : interface
SomeEJBHome was loaded by:
org.jboss.mx.loading.UnifiedClassLoader@2b9f14{url=file:...
The JBoss-generated proxy-class that implements my home
interface is not reloaded after re-deploy. It therefore still
implements the home-interface that was loaded from the previous
deployment (or rather by the UnifiedClassLoader object created
during the first deployment). When my code tries to cast to the
home-interface loaded from this deployment I get the
ClassCastException.
I also see that the temporary files that JBoss
create ...server\default\tmp\deploy... are not removed, and are
still held by JBoss after my re-deploy.
Rune
--
Comment By: Brian Huang (kinmen)
Date: 2002-08-18 18:41
Message:
Logged In: YES
user_id=597050
Normally PortableRemoteObject will return a Proxy object.
However, in case it returns a MarshalledValuePair object, the
ClassCastException will be thrown. This is because you don't
have all the jar files in your ${JBOSS_HOME}/client directory
in your classpath. Please discover if you got a
MarshalledValuePair object back.
Brian Hunag
--
Comment By: Nobody/Anonymous (nobody)
Date: 2002-08-12 07:40
Message:
Logged In: NO
I get the same error as before with 3.0.0. Does the problem for you go away on both
IBM's 1.3 JVM and Sun's 1.4
JVM? Also I am on Sun's 1.4.0_01 JVM on W2K.
Matt
--
Comment By: Steve Harvey (sgharvey)
Date: 2002-08-12 00:50
Message:
Logged In: YES
user_id=593359
If you are using 3.0.0, have you tried 3.0.1?
I went through the same thing as Matt about a week ago,
except I also examined the class loader delegation chains
for both obj and the homeIF class with code like:
try {
// traverse the chain of delegated
classloaders // towards the bootstrap
Class currClass = clazz;
ClassLoader cl;
while ( (cl = currClass.getClassLoader()) != null ) {
System.out.println(label+":classloader is " + cl);
currClass = cl.getClass();
}
} catch ( Exception e ) {
System.out.println("looking for CLs, got " + e);
}
and got discrepancies (more details in my followup for bug
563120) which went away (with the ClassCastException) with
3.0.1 . I too was getting CCE whether or not I used
PortableRemoteObject.narrow.
- Steve
--
Comment By: Nobody/Anonymous (nobody)
Date: 2002-08-09 08:33
Message:
Logged In: NO
This bug is killing my development efforts, but it doesn't look like JBoss fault, but
rather the VM's. First a few
investigative notes. This ClassCastException happens on both remote and local
interfaces using
PortableRemoteObject and without. PortableRemoteObject has
