I'm using camel 2.11-SNAPSHOT in an ear which is running on JBoss AS 7.1.
When deploying the app and while the routes are being constucted (in a #Singleton #Startup bean, using an #Injected CdiCamelContext) i get a lot of warnings (ca. 30) like this in my server log:
2013-03-20 16:40:55,153 WARNING [org.apache.deltaspike.core.api.provider.BeanManagerProvider] (MSC service thread 1-2) When using the BeanManager to retrieve Beans before the Container is started, non-portable behaviour results!
and after the the context has been started i get this WARNINGs:
2013-03-20 16:40:56,339 WARNING [org.apache.deltaspike.core.api.provider.BeanManagerProvider] (Camel (camel-2) thread #1 - file:///tmp/exchange/tmobile/in) When using the BeanManager to retrieve Beans before the Container is started, non-portable behaviour results!
What does it mean? I couldn't find anything useful on google. Is it a bug? Did i configure camel wrong?
It is simply a warning from Apache Deltaspike that the bean manager was accessed before the container was booted. It seems Camel's integration with CDI uses some non-portable features, but assuming that they know what they're doing, it could be nothing.
Related
I am seeing this warning in Wildfly 16:
17:28:43,353 WARN [org.jboss.weld.Bootstrap] (MSC service thread 1-4) WELD-000146: BeforeBeanDiscovery.addAnnotatedType(AnnotatedType<?>) used for class com.sun.faces.flow.FlowDiscoveryCDIHelper is deprecated from CDI 1.1!
I see the following post here: https://developer.jboss.org/thread/280071
is there some way on how to get rid of this warning, using the Spring Framework 3.x.
best
I have this situation:
a JBOSS instance
application client.war
application server.war
a jboss module, properly installed, containing only the interfaces.
The server.war application implements the jboss module interfaces, and publishes these implementations with a JNDI bind. The client.war application with a lookup uses implementations server.war.
A runtime client.war can call the implementation exposed by server.war, but as soon as I try to start a transaction hibernate I get the following error:
ERROR [stderr] java.lang.IllegalStateException: JBAS016071: Singleton
not set for ModuleClassLoader for Module "client.war:main" from
Service Module Loader. This means that you are trying to access a weld
deployment with a Thread Context ClassLoader that is not associated
with the deployment.
There I bumped my head for days, but I can not understand what the problem is. Someone can help me?
Set the class loader on the child thread to be the same as the parent.
Get parent class loader:
ClassLoader cl = Thread.currentThread().getContextClassLoader();
Set child class loader :
ClassLoader cl = Thread.currentThread().setContextClassLoader(cl);
When the child thread is done, make sure to unset the class loader to null, to avoid leaks in case of thread pools.
Although CDI will work in the child thread other things such as remote EJB invocation and JNDI lookups will not.
A much better approch wuld be to use an async EJB invocations You can just create an EJB that looks something like:
#Singleton
public class AsyncBean {
#Asynchronous
public void performTask(int a, int b) {
// the client doesn't care what happens here
}
This would mean that your async task will have the TCCL set correctly, JNDI will work etc (basically it is a full EE invocation).
You can configure the thread pool used for async invocations in standalone.xml, but it will be used for all #Asynchronous methods in the application.
Root Cause
When an application launches its own threads, the new threads use a classloader which is different than the classloader of the originating thread, therefore injection is failing.
Reference
https://access.redhat.com/solutions/257663
I'm trying to deploy a very simple application on Openshift. It's an EAR project with a single WAR and EJB module. Inside the WAR there's a REST service that calls an EJB defined in EJB module. Locally and on Openshift I'm using Wildfly 9.0.0 CR2 and PostgreSQL 9.2. When deploying locally everything works fine. When the same code is deployed on Openshift I'm getting following errors in logs:
2015-06-28 18:23:04,574 WARN [org.jboss.as.clustering.jgroups] (MSC service thread 1-1) WFLYCLJG0006: property bind_addr for protocol org.jgroups.protocols.TCP attempting to override socket binding value 127.12.77.1 : property value 127.12.77.1 will be ignored
2015-06-28 18:23:04,574 WARN [org.jboss.as.clustering.jgroups] (MSC service thread 1-1) WFLYCLJG0006: property bind_port for protocol org.jgroups.protocols.TCP attempting to override socket binding value 7600 : property value 7600 will be ignored
2015-06-28 18:23:06,252 INFO [org.jboss.as.jpa] (ServerService Thread Pool -- 70) WFLYJPA0010: Starting Persistence Unit (phase 2 of 2) Service 'cooking.ear/cooking-ejb-1.0-SNAPSHOT.jar#cookingPU'
2015-06-28 18:23:08,165 ERROR [org.jboss.msc.service.fail] (MSC service thread 1-1) MSC000001: Failed to start service jboss.jgroups.channel.ee: org.jboss.msc.service.StartException in service jboss.jgroups.channel.ee: java.security.PrivilegedActionException: java.net.BindException: [TCP] /127.12.77.1 is not a valid address on any local network interface
at org.wildfly.clustering.jgroups.spi.service.ChannelBuilder.start(ChannelBuilder.java:79)
at org.jboss.msc.service.ServiceControllerImpl$StartTask.startService(ServiceControllerImpl.java:1948)
at org.jboss.msc.service.ServiceControllerImpl$StartTask.run(ServiceControllerImpl.java:1881)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Caused by: java.security.PrivilegedActionException: java.net.BindException: [TCP] /127.12.77.1 is not a valid address on any local network interface
at org.wildfly.security.manager.WildFlySecurityManager.doChecked(WildFlySecurityManager.java:638)
at org.jboss.as.clustering.jgroups.JChannelFactory.createChannel(JChannelFactory.java:99)
at org.wildfly.clustering.jgroups.spi.service.ChannelBuilder.start(ChannelBuilder.java:77)
... 5 more
Caused by: java.net.BindException: [TCP] /127.12.77.1 is not a valid address on any local network interface
at org.jgroups.util.Util.checkIfValidAddress(Util.java:3480)
at org.jgroups.stack.Configurator.ensureValidBindAddresses(Configurator.java:902)
at org.jgroups.stack.Configurator.setupProtocolStack(Configurator.java:118)
at org.jgroups.stack.Configurator.setupProtocolStack(Configurator.java:57)
at org.jgroups.stack.ProtocolStack.setup(ProtocolStack.java:477)
at org.jgroups.JChannel.init(JChannel.java:854)
at org.jgroups.JChannel.<init>(JChannel.java:159)
at org.jboss.as.clustering.jgroups.JChannelFactory$1.run(JChannelFactory.java:96)
at org.jboss.as.clustering.jgroups.JChannelFactory$1.run(JChannelFactory.java:93)
at org.wildfly.security.manager.WildFlySecurityManager.doChecked(WildFlySecurityManager.java:634)
... 7 more
The address mentioned - 127.12.77.1 is $OPENSHIFT_WILDFLY_IP.
I have no idea what is causing this issue. First I thought it's a database connectivity issue because it happens when 2nd phase of starting persistence unit happens. I connected to DB on Openshift and saw that it was created successfully so maybe that's not it, but here's the persistence.xml I'm using:
<?xml version="1.0" encoding="UTF-8"?>
<persistence version="2.0" xmlns="http://java.sun.com/xml/ns/persistence" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://java.sun.com/xml/ns/persistence http://java.sun.com/xml/ns/persistence/persistence_2_0.xsd">
<persistence-unit name="cookingPU">
<provider>org.hibernate.jpa.HibernatePersistenceProvider</provider>
<jta-data-source>java:jboss/datasources/PostgreSQLDS</jta-data-source>
<properties>
<property name="hibernate.dialect" value="org.hibernate.dialect.PostgreSQL9Dialect" />
</properties>
</persistence-unit>
</persistence>
The datasource used is the default one. I didn't change anything in standalone.xml.
Another thing I noticed is that the deploy problem happens when I add any EJB to the project.
This is a simple one I tried to use:
#Stateless
public class AnyEjb {
public String hello() {
return "Hi there!";
}
}
This is defined in EJB module. Then in web module I have this class calling it:
#Path("anything")
#Consumes(MediaType.APPLICATION_JSON)
#Produces(MediaType.APPLICATION_JSON)
public class AnyEndpoint {
#EJB
private AnyEjb anyEjb;
#GET
public String sayHi() {
return anyEjb.hello();
}
}
I'm not sure if and how it can be connected with this BindException.
I've tried running this application locally with both standalone and standalone-full-ha profile and it works in both cases. I just feel it has to be some issue with Openshift configuration but I have no idea where to look anymore. I'm very new to Openshift and Java EE. Please point me in a right direction. Any help will be much appreciated.
Might be https://issues.jboss.org/browse/JGRP-1928. Talk to Rado Husar to see how to resolve this.
It looks like the problem is (as #Bela Ban suggested) connected with the version of JGroups shipped with WildFly 9.0.0CR2. I will be waiting for the fix coming with the Final version of WildFly.
Meanwhile as a workaround to be able to deploy an application on WildFly 9.0.0CR2 on Openshift, I decided to disable clustering capabilities for my server.
In standalone.xml available in .openshift I have removed org.jboss.as.clustering.jgroups module and changed infinispan cache settings from distributed to local.
I have been basing on this solution (for WildFly 8): https://gist.github.com/fjuma/3df7f64fbaebd5506ef5#file-standalone-xml
But I had to modify it so that it works on Wildfly 9. Full standalone.xml that's been working for me is available here for reference http://pastebin.com/aANkPUWk
My application is deployed on JBoss 7.1 (standalone).
I am getting an exception on the following line:
MBeanServerConnection server = MBeanServerLocator.locateJBoss();
The exception is:
JBoss: java.lang.IllegalStateException: No 'jboss' MBeanServer found!
That code above worked fine when the app was deployed on JBoss 5.
From what I was reading online, the code is supposed to work only when it's called from the same JVM in which the MBeanServer was created. Otherwise it's a remote call and I have to use JNDI. But is it not a local call (same JVM - i.e. the JBoss JVM)? How did it work on JBoss 5 then?
How do I make it work on JBoss 7.1 standalone, without changing this specific code?
Here is the solution:
https://community.jboss.org/thread/221708
Quote:
Above problem is dueto locateJboss implementation that is compatible with older version of Jboss. The MBeanServer used by JBoss 7 (by default) is the platform MBeanServer. The class name iscom.sun.jmx.mbeanserver.JmxMBeanServer and the default domain is DefaultDomain. Accordingly, you can simply use:
java.lang.management.ManagementFactory.getPlatformMBeanServer()
Alternatively:
for(MBeanServer server: javax.management.MBeanServerFactory.findMBeanServer(null)) {
if("DefaultDomain".equals(server.getDefaultDomain())) return server;
}
throw new Exception("Failed to locate MBeanServer");
On another note, jboss.system:type=ServerInfo object name doesn't work in AS 7.1 I had to use JVM specific parameters to nail down to MBean attributes. 'java.lang:type=Memory' and attribute as 'HeapMemoryUsage'.
I am new in the world of EJB 3.1 and trying to get some basics with the JBoss Application Server 7.1.
At the moment I am stuck at a - really basic - problem. When a bean on the server wants to use another bean I need to use the InitialContext.lookup() method. If I look in the literature I found calls like:
TheBean = (<Interface>) new InitialContext().lookup("<NameOfTheBean>/local");
But this call doesn’t work for me. Every time I get an error like this:
javax.ejb.EJBException: javax.naming.NameNotFoundException
After hours of looking for a solution I found code examples which use another call, something like this:
TheBean = (<Interface>) new InitialContext().lookup("ejb:/<Package>//<NameOfTheBean>!<Package>.<Interface>");
Well this solution works for me but the question is why? Does somebody know why the first call produces exceptions while the second one works fine?
Thanks a lot!
Why? Different versions of JBoss deploys beans with differently default names in JNDI namespace.
ctx.lookup("BeanName/local")
was right for the JBoss 4.x.x and higher but not for JBoss v7.
You can see in you server.log of JB7 how your beans mapped to JNDI names,
for example (see java:/jboss/exported/... and how it correspond to your second successive call):
13:57:05,550 INFO [org.jboss.as.ejb3.deployment.processors.EjbJndiBindingsDeploymentUnitProcessor] (MSC service thread 1-4) JNDI bindings for session bean named ProductionHistoryBean in deployment unit deployment "navi-ejb3.jar" are as follows:
java:global/navi-ejb3/ProductionHistoryBean!navi.ejb3.production.history
.ProductionHistoryRemote
java:app/navi-ejb3/ProductionHistoryBean!navi.ejb3.production.history.Pr
oductionHistoryRemote
java:module/ProductionHistoryBean!navi.ejb3.production.history.Productio
nHistoryRemote
java:jboss/exported/navi-ejb3/ProductionHistoryBean!navi.ejb3.production
.history.ProductionHistoryRemote
java:global/navi-ejb3/ProductionHistoryBean
java:app/navi-ejb3/ProductionHistoryBean
java:module/ProductionHistoryBean