July 2008 - jboss-dev-forums - Jboss List Archives

[Design of POJO Server] - Re: Scoping of war/jar file embedded in sar service archives

by adinn

Ok, Scott, I found the cause of the problem -- there were a couple of small but significant details in the sar setup which I failed to include in the problem report. Firstly, the jars are actually layered as 8 war-specific jars which employ code from 5 generic code jars. Sequencer, the class with the static data, was in one of the generic jars (ws-c.jar). Secondly, the war files and war-specific jar files employ Classpath entries in the manifest to refer to the jars they rely on. So, for example, ws-c11.war refers to ws-c11.jar which refers to ws-c.jar (contains Sequencer), while ws-t11.war refers to ws-t11.jar and ws-c11.jar and ws-t11.jar refers to ws-t.jar and ws-c.jar. I worked out by trial and error that removing the classpath entries in the war file manifest solves the class duplication problem. I assume this is because the war class loader grabs its own copy of classes in the the jars (recursively) mentioned in the classpath rather than delegating to the sar loader. I did not actually need to remove the jar classpath entries to fix the problem and assume this would not be strictly necessary since the entries only reference jars contained within the same sar i.e. they do not intercede between the loader of the referencing class and some other loader. Is that a valid assumption or was I just lucky? I'll still remove the classpath entries anyway -- the app works fine without them. I don't know whether this needs to be documented as a bug, change of behaviour or just useful thing someone might want to know some day. Anyway, it probably ought to be noted somewhere where AS users or the support team can find it (well, it is noted here at least :-) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163813#4163813 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163813

18 years

1
0
0 / 0

[Design of Clustering on JBoss (Clusters/JBoss)] - Re: JBAS-5647 HAPartition Sync Call with RspFilter

by galder.zamarreno＠jboss.com

"bstansberry(a)jboss.com" wrote : String manipulation? I was thinking in terms of some kind of map lookup. | | Ah, you're creating new ClusterNodeImpls. That looked the simplest option but I must admit that I did think on whether Address to ClusterNode mappings are/should be stored in a map to avoid creation of ClusterNodeImpls for each RPC call. I'll double check to see whether there's something in place for this. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163777#4163777 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163777

18 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Some things I don't understand in the transport

by jmesnil

"timfox" wrote : 1) Why do we need to create multiple RemotingSession objects per RemotingConnector? Why not just use the same RemotingSession object, then RemotingConnector and RemotingSession could be combined in a single object. | The RemotingSessions object created for the same RemotingSession wraps the same MINA IoSession internally. anonymous wrote : | 2) Why do we create a new RemotingConnection object each time we create a jms connection? why not just use the same one per server? the RemotingConnection object could be stored in the connector registry. Moreover each RemotingConnection object currently starts its own pinger, this means we have one pinger per jms connection even if they are working over the same underlying tcp connection, which is unnecessary. | yes, this code is more convoluted than necessary. Anything refactoring to simplify it would be a good thing. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163772#4163772 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163772

18 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Some things I don't understand in the transport

by ataylor

you're correct on both counts, in the end they use the same connector so it would make sense to change it, this would actually simplify things. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163768#4163768 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163768

18 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Wiki pages for our design topics

by timfox

Yes, that's difficult. I'm going to concentrate on getting the simples case working first. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163764#4163764 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163764

18 years

1
0
0 / 0

[Design of Clustering on JBoss (Clusters/JBoss)] - Re: JBAS-5647 HAPartition Sync Call with RspFilter

by bstansberry＠jboss.com

"galder.zamarreno(a)jboss.com" wrote : I see that DNS lookups are now avoided thanks to http://jira.jboss.com/jira/browse/JBAS-3794. Clever. Clever but poorly done. :-) ClusterNodeImpl.getFastHostName was incorrectly implemented; I just checked in the fix. Also. 1.1.0.CR2 of the ha-xxx libs has been released. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163744#4163744 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163744

18 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Wiki pages for our design topics

by clebert.suconic＠jboss.com

Some thoughts I had: - What should happen if you bring a "backup node" up, after the live node is running? - on that case how do we copy the state to the new node as new requests are coming? If we support that, is it possible to do that without requiring a valve and lock the server for some time? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163727#4163727 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163727

18 years

1
0
0 / 0

[Design the new POJO MicroContainer] - Configure javadoc plugin with APIviz in jboss-cl

by scott.stark＠jboss.org

I added a javadoc configuration to jboss-cl/build/pom.xml to allow the use of the APIviz doclet to generate basic uml images as part of the javadoc package and class overviews. To generate the diagrams you have to have the Graphviz dot utility program in your system path. If its not just plain javadoc is generated. http://www.graphviz.org/About.php View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163714#4163714 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163714

18 years

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Some things I don't understand in the transport

by timfox

Looking through the code in detail as I look at the packet replication stuff, some questions spring to mind: 1) Why do we need to create multiple RemotingSession objects per RemotingConnector? Why not just use the same RemotingSession object, then RemotingConnector and RemotingSession could be combined in a single object. 2) Why do we create a new RemotingConnection object each time we create a jms connection? why not just use the same one per server? the RemotingConnection object could be stored in the connector registry. Moreover each RemotingConnection object currently starts its own pinger, this means we have one pinger per jms connection even if they are working over the same underlying tcp connection, which is unnecessary. Seems to me our cardinalities are a bit messed up. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163695#4163695 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163695

18 years

1
0
0 / 0

[Design of Clustering on JBoss (Clusters/JBoss)] - Re: JBAS-5647 HAPartition Sync Call with RspFilter

by bstansberry＠jboss.com

String manipulation? I was thinking in terms of some kind of map lookup. Ah, you're creating new ClusterNodeImpls. If ClusteredNodeImpl.jgId were made volatile and lazy initialized in getJGName() the constructor would be pretty fast. The getJGName() method isn't even in the ClusterNode interface; I just left the method in the impl when I created the interface just in case someone was using it. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4163684#4163684 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4163684

18 years

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 July 2008