March 2009 - jboss-dev-forums - Jboss List Archives

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: NettyConnector slow to shutdown

by timfox

I can see the Netty selector timeout is 500. I'm wondering if this can be interrupted at close to speed up the shutdown? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214541#4214541 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214541

17 years, 1 month

1
0
0 / 0

[Design of OSGi Integration] - Re: JBOSGI-38 and bundle deployment order

by thomas.diesler＠jboss.com

The current implementation of the BundleStartStopDeployer uses (and requiers) the PackageAdmin core service | 15:26:41,512 INFO [MainDeployer] deploy, url=file:/home/tdiesler/svn/jboss-osgi/trunk/testsuite/target/test-libs/jbosgi38-bundleA.jar | 15:26:41,550 INFO [OSGiDeployer] Installed: jbosgi38-bundleA,state=2 | 15:26:41,582 INFO [BundleStartStopDeployer] Unresolved: [jbosgi38-bundleA [5]] | 15:26:41,594 INFO [MainDeployer] deploy, url=file:/home/tdiesler/svn/jboss-osgi/trunk/testsuite/target/test-libs/jbosgi38-bundleB.jar | 15:26:41,609 INFO [OSGiDeployer] Installed: jbosgi38-bundleB,state=2 | 15:26:41,619 INFO [BundleStartStopDeployer] Unresolved: [jbosgi38-bundleA [5], jbosgi38-bundleB [6]] | 15:26:41,627 INFO [MainDeployer] deploy, url=file:/home/tdiesler/svn/jboss-osgi/trunk/testsuite/target/test-libs/jbosgi38-bundleX.jar | 15:26:41,647 INFO [OSGiDeployer] Installed: jbosgi38-bundleX,state=2 | 15:26:41,679 INFO [BundleStartStopDeployer] Started: jbosgi38-bundleA,state=32 | 15:26:41,688 INFO [BundleStartStopDeployer] Started: jbosgi38-bundleB,state=32 | 15:26:41,689 INFO [BundleStartStopDeployer] Started: jbosgi38-bundleX,state=32 | View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214535#4214535 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214535

17 years, 1 month

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - NettyConnector slow to shutdown

by timfox

I've been investigating why it seems to take at least 500 ms between closing the last session and the session close returning. This is one of the things that has really been slowing down the test suite. It seems that after calling NettyConnector.close() it always pauses 500ms before actually closing. I'm thinking there's a Thread.sleep(500) or 500 ms timeout somewhere in the Netty code that's causing this problem. Any got any ideas? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214534#4214534 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214534

17 years, 1 month

1
0
0 / 0

[Design of JBoss Profiler] - Re: Thoughts on jboss-profiler-2.0.Beta3.SP1 install/usage i

by jesper.pedersen

Thanks for the feedback. Documentation is a known area that needs improving - see https://jira.jboss.org/jira/browse/JBPROFILER-59 - and since this is a community project any patches to this area is most welcomed. anonymous wrote : - "java -jar jboss-profiler-client.jar getSnapshot 1" And then, what? There is no Web GUI anymore? Now, it's all about TXT files? The 2.x release series is different from the 1.x release series - as the 2.x uses 100% Java code for the profiling. There is an effort to integrate native code in the 2.x release series also - see https://jira.jboss.org/jira/browse/JBPROFILER-60 - again, feel free to submit patches in this area. As to the web GUI - there is an initial implementation in SVN | ant web | Again - patches welcome ! As you can tell - this is a community project, so any patches that will help making it better is most welcomed. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214521#4214521 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214521

17 years, 1 month

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: ByteBuffer.allocateDirect ridiculously slow

by clebert.suconic＠jboss.com

Jason told me on a pvt email that there is a Sleep(100ms) and a System.gc call for every direct buffer you create. wow! View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214515#4214515 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214515

17 years, 1 month

1
0
0 / 0

[Design of JBoss Profiler] - Thoughts on jboss-profiler-2.0.Beta3.SP1 install/usage instr

by alegomes

Hi Cleber et. al. This is my first time running jboss-profiler-2.0.Beta3.SP1. I've just downloaded it, started investigating README.txt and there comes my feedback. - In the begining, README.txt mentions 'command line client'. In my understanding, there should be a CLI 'client' tool somewhere, but I couldn't find it. - Then, a bunch of parameters are listed, but I think it's not relevant at that moment. The fisrt thing I'd like to see as a jboss-profiler newbie is how do I get it up and running, not the whole set o config params, that could be available for references only. - So, I found a section called 'Deployment in JBoss Application Server'. That's what I was looking for. I think that should be on the very beginning of the document. - "Edit jboss-profiler.properties in jbossas/bin to include the classes to be profiled" Should I read all the possible config parameters to get the point? Why not just show an example right here (includes=...) and then make a reference to the complete documentation? - "Edit run.conf (Unix) or run.bat (Windows) in jbossas/bin to include JBoss Profiler in JAVA_OPTS" Ok, but HOW?! Again, I must read all the doc to get the right configuration. - "Boot application server" I did, and I got a 'java.lang.NoClassDefFoundError: javassist/ClassPool' because and didn't read further (yes, I'm using JB 4.2.2). - "java -jar jboss-profiler-client.jar getSnapshot 1" And then, what? There is no Web GUI anymore? Now, it's all about TXT files? Just make it clear. I'm not criticizing the great job you've done. It's way better than previous versions. The thing is I want to contribute to make jboss-profiler more popular than ever. []s View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214514#4214514 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214514

17 years, 1 month

1
0
0 / 0

[Design of OSGi Integration] - Re: JBOSGI-38 and bundle deployment order

by thomas.diesler＠jboss.com

There can be an unspecified amount of time between bundle deployments. | Bundle A depends on bundle B, both share bundle X. | | A ---> B | A ---> X <--- B | and a deployment order of | deploy("jbosgi38-bundleA.jar"); | deploy("jbosgi38-bundleB.jar"); | deploy("jbosgi38-bundleX.jar"); | Bundle A & B can only be installed but not started. The deployment should not fail however. When X gets deployed all three bundles should get started. The ResolvedBundleDependencyItem would probably not fix that, because deployment of A would already fail - if I understand this correctly. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214501#4214501 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214501

17 years, 1 month

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: ByteBuffer.allocateDirect ridiculously slow

by clebert.suconic＠jboss.com

"timfox" wrote : It's a well known fact that direct bytebuffer allocation is much slower than non direct byte buffers. | | If you think about it when you allocate a non direct byte buffer then it basically just needs to dereference a pointer to some memory on the Java heap, which is very quick. But for a direct buffer, it has to malloc real memory from the OS, and do a bunch of other house keeping. | | So yes, if you're using direct byte buffers it pays to re-use them. | | You should speak to Trustin about this. I believe Netty uses non direct buffers for exactly this reason. I need a direct buffer on AIO, because of memory alignments and JNI invocations. But there shouldn' t be a reason for that. If I called malloc/posix_mem_align (which is the same as malloc, but aligned) and created the direct myself using the memory position, and a JNI method, it would be way much faster (100x as my test showed). View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214497#4214497 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214497

17 years, 1 month

1
0
0 / 0

[Design of JBoss Web Services] - JBossWS CZJUG Presentation

by richard.opalka＠jboss.com

Downloads available at: http://mediacast.sun.com/users/jakub.podlesak/media/CZJUG2009.01-jboss-ws... Enjoy ;) View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214496#4214496 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214496

17 years, 1 month

1
0
0 / 0

[Design of POJO Server] - Re: Clustered DeploymentRepository for 5.1

by emuckenhuber

"scott.stark(a)jboss.org" wrote : | The test.xml example is a little too file based to be a general metadata view unless the source can be taken as just a URI to be resolved by something. | The meta data does not dictate a uri as source. The only thing required is a profileFactory, which can handle a specific implementation of ProfileMetaData to create a ProfileImpl. Resolving should also be part of this implementation. As this is something which does not necessarily apply to all profiles in the same way. This test is just a quick prototype and is not finalized at all. Basically i just was testing the XmlAnyElement - so that you can plugin your own meta data implementation. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4214490#4214490 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4214490

17 years, 1 month

1
0
0 / 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jboss-dev-forums March 2009