September 2008 - jboss-dev-forums

[Design of AOP on JBoss (Aspects/JBoss)] - Some maven updates

by pgier

I'm planning to continue some work on the AOP mavenization tomorrow. I will be moving the PluggableInstrumentor class to it's own module, and using maven to compile classes and create the jars in the aop module. The testsuite will still be run using ant the same way as before, and you will still be able to build all the modules using the shell script (build.sh) or ant from the main build directory. If this causes problems for anyone, please let me know. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177543#4177543 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177543

17 years, 6 months

1
0
0 / 0

[Design of JBossCache] - Re: memcached client/server

by genman

ConsistentHashing allows you to replicate to as many servers as you want. Imagine a circle with dots representing "hashed server addresses". Then, pick a point, go clockwise/counterclockwise on the circle and find two or more unique instances. | package org.jboss.cache.util; | | import java.util.Collection; | import java.util.Collections; | import java.util.HashSet; | import java.util.Set; | import java.util.SortedMap; | import java.util.TreeMap; | | import org.jgroups.Address; | | /** | * Consistent hashing algorithm; see http://weblogs.java.net/blog/tomwhite/archive/2007/11/consistent_hash.html | * | * @param <T> type of nodes to use; consider {@link Address} | */ | public class ConsistentHash<T> { | | /** | * Default (but overrideable) hash function. | */ | public static class HashFunction { | | /** | * Implementation from ConcurrentHashMap. | * @param o object to cache | * @return hopefully a well distributed hash code result | */ | public int hash(Object o) { | int i = o.hashCode(); | i += ~(i << 9); | i ^= i >>> 14; | i += i << 4; | i ^= i >>> 10; | return i; | } | | } | | private final HashFunction hashFunction; | private final int numberOfReplicas; | private final SortedMap<Integer, T> circle = new TreeMap<Integer, T>(); | | public ConsistentHash(HashFunction hashFunction, int numberOfReplicas, | Collection<T> nodes) { | this.hashFunction = hashFunction; | this.numberOfReplicas = numberOfReplicas; | | for (T node : nodes) { | add(node); | } | } | | /** | * Constructs a new ConsistentHash with default options. | */ | public ConsistentHash() { | this(new HashFunction(), 5, Collections.<T>emptyList()); | } | | /** | * Adds a node. | */ | public void add(T node) { | for (int i = 0; i < numberOfReplicas; i++) { | circle.put(hashFunction.hash(node.toString() + i), node); | } | } | | /** | * Removes a node. | */ | public void remove(T node) { | for (int i = 0; i < numberOfReplicas; i++) { | circle.remove(hashFunction.hash(node.toString() + i)); | } | } | | /** | * Returns N nodes for a key. | */ | public Set<T> get(Object key, int count) { | if (count <= 0) | throw new IllegalArgumentException("count"); | if (key == null) | throw new NullPointerException("key"); | if (circle.isEmpty()) { | return Collections.emptySet(); | } | // TODO bound the "count" to the number of instances | Set<T> nodes = new HashSet<T>(count); | int hash = hashFunction.hash(key); | for (T node : circle.tailMap(hash).values()) { | nodes.add(node); | if (nodes.size() == count) | return nodes; | } | for (T node : circle.values()) { | nodes.add(node); | if (nodes.size() == count) | return nodes; | } | return nodes; | } | | /** | * Returns a debug <code>String</code>. | */ | @Override | public String toString() | { | return super.toString() + | " numberOfReplicas=" + this.numberOfReplicas + | " circle=" + this.circle + | ""; | } | | } | View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177513#4177513 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177513

17 years, 6 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Annoyance when running unit tests in eclipse

by ataylor

Ok, Ive fixed it. tests/config was ahead of src/config in the classpath. When i added the log4j file to tests/config it was picked up first. Ive switched these around. Ive installed an eclipse plugin for intellij so from now on i'll use the same classpath as you eclipse users View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177472#4177472 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177472

17 years, 6 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Eclipse code formatter and code cleanup profiles

by ataylor

Ive added intellij settings as well. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177449#4177449 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177449

17 years, 6 months

1
0
0 / 0

[Design of JBossCache] - Re: memcached client/server

by manik.surtani＠jboss.com

My issue with the metadata approach is that it can lead to unnecessary traffic, and while this traffic is small, could be a scalability problem. I very much prefer the consistent hashing approach if we can make it work for us. Using two hash functions to map an Fqn to two different servers is one approach, but this is hard when you want > 1 backup copy. Regarding memcached in relation to JBC, I actually see JBC as a memcached server rather than a client. E.g., Bela's memcached server code could act as a "front end" to JBC, and instead of putting stuff in a CHM, it could put stuff into a JBC instance. This buys us: 1. The ability to use different types of clients talking to JBC (any memcached client library should work) 2. The ability to be colocated in-VM as well, for a more p2p-like setup To make this work we would need clients to be aware of the consistent hashing algo, so it could direct requests to the relevant server (as an optimisation), and for clients that don't do this (as well as colocated clients) the JBC instance would need a subclass of the ClusteredCacheLoader that has the consistent hashing algo. We then automatically have cache instances acting as L1 caches if they aren't the "host" cache for a particular fqn, and automatically would have L1 cache invalidation (assuming the cache uses INVALIDATION. REPLICATION would be pointless anyway in such a setup). View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177433#4177433 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177433

17 years, 6 months

1
0
0 / 0

[Design of JBoss Deployment Framework] - Re: Datasource deployer does not substitute system propertie

by alex.loubyansky＠jboss.com

The discussion will be continued on the thread associated with the JBAS-4585 http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177422#4177422 View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177423#4177423 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177423

17 years, 6 months

1
0
0 / 0

[Design of JCA on JBoss] - Re: Datasources not parsing jboss.server.data.dir

by alex.loubyansky＠jboss.com

Currently most (if not all) of the string properties use javax.xml.bind.annotation.adapters.CollapsedStringAdapter. As a quick fix I would propose to extend it, e.g. public class ManagedConnectionStringAdapter extends CollapsedStringAdapter | { | @Override | public String marshal(String v) | { | return super.marshal(v); | } | | @Override | public String unmarshal(String v) | { | v = super.unmarshal(v); | return v == null ? null : StringPropertyReplacer.replaceProperties(v); | } | } View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177422#4177422 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177422

17 years, 6 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Notifications over JBM core connections

by timfox

"jmesnil" wrote : "timfox" wrote : | | It makes sense to move the notification handling to the session. | | But it still got the same issue: ServerSession.send() expect a ServerMessage that I need to create. | | | | You could just add a constructor to ServerMessage, like the ClientMessage one: | | | | public ClientMessageImpl(final boolean durable, MessagingBuffer body) | | { | | super((byte) 0, durable, 0, System.currentTimeMillis(), (byte)4, body); | | } | | | | And create a buffer before creating it, like is done in ClientSessionImpl::createClientMessage() but with a server message rather than a client message. | | Then call session::send directly on the server, and that will handle all the assigning of a message id etc. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177420#4177420 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177420

17 years, 6 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Notifications over JBM core connections

by jmesnil

"timfox" wrote : | What if there is no jmx server running? All this stuff should work with no JMX server too. | It does work even when there is no JMX MBeanServer (the ManagementService keeps a registry of managed resources using ObjectNames as their keys) "timfox" wrote : | I think it might be better to introduce a new packet on the wireformat ManagementMessage and send via that, then there's no need for special routing in the post office - it can be intercepted at the session level | "timfox" wrote : | "jmesnil wrote : * the management service creates a NotificationListener for the given ObjectName | | | | This seems JMX dependent to me | Same as above, I use a NotificationBroadcaster to keep track of notifications but it does not depend on any MBeanServer "timfox" wrote : | Are notifications supposed to be persisent (i.e. do they survive after the session that created them has died?) | | If not, then the listener could just be the session, and you could use the standard session send functionality to send the notification I do not intend for the notifications to be persistent. It makes sense to move the notification handling to the session. But it still got the same issue: ServerSession.send() expect a ServerMessage that I need to create. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177415#4177415 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177415

17 years, 6 months

1
0
0 / 0

[Design of EJB 3.0] - Re: JNDI Bindings

by jaikiran

"wolfc" wrote : | You can have remote-binding elements in the descriptor and these can have jndi-name. They're under a session element. | | Is this currently available? The jboss_5_0.dtd http://www.jboss.org/j2ee/dtd/jboss_5_0.dtd does not mention this View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4177410#4177410 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4177410

17 years, 6 months

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 September 2008