[JBoss-user] [JNDI/Naming/Network] - Re: Error on lookup From a clustered SLSB to another cluster
I'm not sure how you would get high availability between two ejb deployed on distintc clusters making the lookup using JNDI. the point of HAJNDI is that although you initially contact one box in the cluster (any box, doesn't matter), you'll get back info on the rest of the cluster so that requests can be transparently load balanced as well as transparently failed-over (this is all handled under the hood by the HA invoker). If you just use JNDI, you are only ever talking to one box in the cluster, so you do not get high availability. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3950772#3950772 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3950772 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: talking to remote jboss instance from within jboss
i'm curious as to why this is considered acceptable behavior? As far as my implementation is concerned, i've used a differently configured initial context to get a handle to a remote ejb, but that handle violates the context and decides to make the call locally? the optimization is indeed a good thing when invoking ejbs within the local cluster, but in the given example completely violates the configured context. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946928#3946928 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946928 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: talking to remote jboss instance from within jboss
After looking at the code a bit, it seems like it would be fairly straightforward to fix this problem. The InvokerInterceptor already checks whether the invoker is an instance of InvokerProxyHA. A method could be added to this interface to get the partition name of the invoker, which could then be compared against the local partition name. this information could be used to determine whether the clustered call can be made to the local server or not. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946941#3946941 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946941 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: talking to remote jboss instance from within jboss
One last note, i realize that the requirement to separate the partition names is not a bug. i was just noting that nowhere in the wiki/documentation did i find information that gave reasons why you would have to have this value distinct between servers (because in many situtations it does not need to be distinct if the mcast_addr and mcast_ports are distinct). View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946942#3946942 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946942 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [JNDI/Naming/Network] - Re: Error on lookup From a clustered SLSB to another cluster
I had similar problems. Most of the issues were resolved by using distinct partition names between the two servers. See thread here and linked bug: http://www.jboss.org/index.html?module=bbop=viewtopict=83207 View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946944#3946944 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946944 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [JNDI/Naming/Network] - Re: Error on lookup From a clustered SLSB to another cluster
I had similar problems. Most of the issues were resolved by using distinct partition names between the two servers. See thread here and linked bug: http://www.jboss.org/index.html?module=bbop=viewtopict=83207 View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946945#3946945 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946945 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: talking to remote jboss instance from within jboss
Just to let anyone know who happens upon this topic, i have opened a bug related to this situation here: http://jira.jboss.org/jira/browse/JBAS-3241 View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946697#3946697 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946697 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: talking to remote jboss instance from within jboss
Clickable http://jira.jboss.org/jira/browse/JBAS-3241 View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946698#3946698 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946698 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: Talking to two separate jboss instances at the same time
So, i found a solution to this problem while tracking down related issues with cross-jboss communication. It boils down to the fact that if you need communicate between distinct jboss clusters, they *must* have distinct partition names. we were using only the mcast_addr and mcast_port to keep the clusters separate. this works fine in isolation, but once communication starts to bridge that gap, things start to get whacky if the clusters have the same partition name. you can find more details here: http://jira.jboss.org/jira/browse/JBAS-3241 View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3946700#3946700 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3946700 --- All the advantages of Linux Managed Hosting--Without the Cost and Risk! Fully trained technicians. The highest number of Red Hat certifications in the hosting industry. Fanatical Support. Click to learn more http://sel.as-us.falkag.net/sel?cmd=lnkkid=107521bid=248729dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [JNDI/Naming/Network] - Re: Connecting from one server to other
Hmm, I'm starting to think this is a bug in 4.0.3SP1. I'm having the exact same issue, as i posted in a separate forum: http://www.jboss.org/index.html?module=bbop=viewtopict=83207 View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3944957#3944957 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3944957 --- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Talking to two separate jboss instances at the same time
Hey, I'm trying to write a client which will communicate with two separate remote jboss instances (version 4.0.3SP1), call them JBSrc and JBDst (both are actually clusters). The end goal is to copy data from JBSrc into JBDst. There are 3 scenarios i can think of for doing this: 1. Get remote interface from JBSrc (call it Exporter) and hand that interface directly to the import method in JBDst, and allow JBDst to suck the data directly from JBSrc. The data path is JBSrc - JBDst, which is the ideal scenario 2. Get Exporter interface from JBSrc and create Exporter proxy in client which is handed to JBDst. The data path is then JBSrc - client - JBDst, which is okay, but not the ideal. 3. Get the Exporter interface from JBSrc, suck all the data to local disk, then create Exporter proxy which gets data from the local dist. The data path is then 2-part: JBSrc - client/disk, then client/disk - JBDst. So, I've started experimenting with scenarios 1/2, but i can't for the life of me manage to get the client to talk to two different remote jboss instances at the same time. I want to avoid 3 because the data can be very large and it would suck to have to cache it all locally first. The stuff I've done so far: - for the security config, i'm using the ClientLoginModule with multi-threaded=true - I set up two separate Properties configurations (one on each thread) with the provider url, and the init context factory set to org.jboss.naming:org.jnp.interfaces (i've also tried org.jboss.naming.NamingContextFactory which doesn't seem to help) What seems to happen is that thread1 will login to JBSrc and talk to it successfully, then thread2 will login to JBDst and talk to it successfully. However, when thread1 tries to talk to JBSrc again, it actually tries to talk to JBDst. It seems like some global state is being set as to the current jboss instance, and all threads have to talk to that instance...? Has anyone else had to manage simultaneous communication to separate instances, and if so, am i missing something? View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3944536#3944536 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3944536 --- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - talking to remote jboss instance from within jboss
Hey,
We are trying to get an EJB on one jboss instance to make a remote method call
to an EJB on another jboss instance (version 4.0.3SP1). In both cases, the
EJBs are clustered. The problem, however, is that all attempts to contact the
remote EJB result in method calls to the local jboss instance. Initially,
the caller had both applications running, so a successful call was made to
the local application (not really successful because we are trying to call the
remote service!). When we removed the local instance of the other application,
then we would just get service not found exceptions. The code snippet
involved is:
| public MetadataRepository getMetadataRepository(String otherUrl) throws
CreateException, RemoteException, NamingException {
| Properties env = new Properties();
| env.put(Context.INITIAL_CONTEXT_FACTORY,
org.jnp.interfaces.NamingContextFactory);
| env.put(Context.URL_PKG_PREFIXES,
org.jboss.naming:org.jnp.interfaces);
| env.put(Context.PROVIDER_URL, otherUrl);
| Context ctx = new InitialContext(env);
| Object ref =
ctx.lookup(com.hmsonline.scorpio.client.MetadataRepositoryHome.JNDI_NAME);
| return
((com.hmsonline.scorpio.client.MetadataRepositoryHome)PortableRemoteObject.narrow(ref,
com.hmsonline.scorpio.client.MetadataRepositoryHome.class)).create();
| }
|
I've poked around jira and the wiki and seen some references to past bugs which
talk about optimizations of clustered method calls which end up making local
calls, however all referenced bugs seem to be fixed (e.g. JBAS-1442). Any
ideas on what i might be doing wrong?
View the original post :
http://www.jboss.com/index.html?module=bbop=viewtopicp=3944541#3944541
Reply to the post :
http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3944541
---
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642
___
JBoss-user mailing list
JBoss-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: Calling a remote ejb interface in app1.ear from a servle
you did not specify if you were calling the local or the remote interface of the EJB. That is an important distinction. If you are calling the local interface, you pretty much have to turn off isolation or stick the everything in the same ear (same diff). If you are calling the remote interface, then this can be made to work (we are doing this). the trick is that jboss ships with default settings that optimize local calls to remote interfaces (this actually violates the j2ee correctness). while this makes things slightly faster, this makes using isolation impossible in the scenario you describe. In the deploy/ear-deployer.xml, you must have these settings: attribute name=Isolatedtrue/attribute | attribute name=CallByValuetrue/attribute Hope that helps. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3944546#3944546 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3944546 --- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: Calling a remote ejb interface in app1.ear from a servle
Oops, guess i missed the fact that you do say it is the remote interface in the subject. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3944553#3944553 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3944553 --- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: Calling a remote ejb interface in app1.ear from a servle
well, to reference your previous post, isolation is not a form of security. all it does is allow different versions of the same class to exist in different ears. it does nothing to control who can call it. to answer your recent question: ear2 should treat ear1 the same as if ear1 was deployed on a separate app server, i.e. ear2 should contain ear1's interfaces. ear1 should be deployed as normal. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3944624#3944624 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3944624 --- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [EJB/JBoss] - Re: Calling a remote ejb interface in app1.ear from a servle
The security comment was in reference to your comment of May 12. anonymous wrote : After all, if my ejbs are secured, who cares what other ears can attempt to call them. But that's neither here nor there. As for the ClassCastException, i was referring to that problem with my comments about the ear-deployer.xml settings. We are doing exactly what i described (the remote interfaces exist in both ears on the same box), and it is working just fine. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3944675#3944675 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3944675 --- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnkkid=120709bid=263057dat=121642 ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [Messaging, JMS JBossMQ] - Re: xa transaction failure not handled correctly when lots o
just wanted to close out this thread if anyone else happens upon it. i finally found the cause of my spontaneous rollbacks, and it was a jboss issue. it turns out that the XAResource transaction timeouts are controlled independently of the TransactionManager transaction timeouts. in the 4.0.3 version of jboss, a config param was added to the datasource config files xa-resource-timeout, which enables one to set the XAResource transaction timeout (in seconds). by sufficiently lengthening that timeout (default is 5 minutes), the oracle datasource no longer rollsback mid-commit. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3913034#3913034 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3913034 --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [Messaging, JMS JBossMQ] - Re: xa transaction failure not handled correctly when lots o
actually, it *was* partially a jbossmq bug. i'd say that the messages getting committed on rollback was most certainly a jbossmq problem. i am indeed pursuing the problem from the oracle side of things. apparently you have some crystal ball and can tell that the jboss transaction manager is communicating correctly with oracle and the only possible cause of the problem is oracle! i, unfortunately, can't verify that jboss is not at fault, so i'll continue to pursue the problem from both ends until i get a fix. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3911947#3911947 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3911947 --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [Messaging, JMS JBossMQ] - Re: xa transaction failure not handled correctly when lots o
is it possible that i've already read that page and followed the basic instructions to configure oracle for xa? if you read my original email, xa on oracle works fine until the number of messages in the transaction gets large (as in the example code). you are correct that the answer may be oracle related, but i honestly can't point the finger at oracle or jboss at this point. that's why i was asking the question, in case someone has seen this behavior and figured out how to solve it. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3911447#3911447 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3911447 --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [Messaging, JMS JBossMQ] - Re: xa transaction failure not handled correctly when lots o
you're probably right that this is no longer a jbossmq problem (as i've now gotten that question answered). which forum would you recommend that i move my question to? there isn't a forum specifically for transaction related issues. the closest might be the ejb forum? View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3911457#3911457 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3911457 --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [Messaging, JMS JBossMQ] - Re: xa transaction failure not handled correctly when lots o
to follow up on this, i managed to get my test code running on 4.0.3SP1, and the rollback problem has indeed been fixed, but there is still the problem of why the rollback occurs in the first place. so, one of the original questions remains: has anyone else seen this behavior while using jboss xa transactions with messaging, possibly also with an oracle database on the backend. View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3911287#3911287 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3911287 --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
[JBoss-user] [Messaging, JMS JBossMQ] - xa transaction failure not handled correctly when lots of me
well, there really seem to be two issues. first, sending lots of messages
within an xa transaction fails when lots ~15000. second, when the
transaction does fail, the messages that were sent still get sent, event though
the rest of the transaction rolls back. i'm using oracle on the backend, and
the question i have here is: has anyone else seen this behavior (and fixed it
through some configuration tweak), or is this a legitimate jboss bug (in which
case i will file a jira entry)? This is being run on jboss 4.0.1.
the test code that i am using at this point is included below. I am not using
CMT because the actual code doesn't run within an EJB (i just stuck it in one
for simplicity of testing).
public void runTest(int numMessagesToSend) throws Exception
{
try {
Transaction tx = null;
LOG.info(Beginning transaction);
getTxManager().begin();
tx = getTxManager().getTransaction();
tx.registerSynchronization(new TxDebugger(, tx.toString()));
XASession session = null;
MessageProducer sender = null;
try {
LOG.info(Creating Session);
session = getHAXAConnection().createXASession();
tx.enlistResource(session.getXAResource());
sender = session.createProducer(getWorkQueue());
LOG.info(Sending + numMessagesToSend + messages);
for(int i = 0; i numMessagesToSend; ++i) {
ObjectMessage msg = session.createObjectMessage();
msg.setObject(new TestXAPayload());
sender.send(msg, DeliveryMode.PERSISTENT, 4,
Message.DEFAULT_TIME_TO_LIVE);
}
LOG.info(Done sending messages);
} finally {
JMSUtil.close(sender, session);
}
jdbcSave(numMessagesToSend);
// commit tx
LOG.info(Committing transaction);
getTxManager().commit();
} finally {
drainQueue();
}
LOG.info(Finished);
}
private void jdbcSave(int numMessagesToSend) throws Exception
{
LOG.info(Saving TestXAState (jdbc));
DataSource ds = (DataSource)new
InitialContext().lookup(java:/DrenchTestDS);
java.sql.Connection conn = null;
java.sql.Statement stmt = null;
try {
conn = ds.getConnection();
stmt = conn.createStatement();
stmt.executeUpdate(INSERT INTO TESTXASTATE(ID, NUMMESSAGESSENT) VALUES
('1', ' + numMessagesToSend + '));
} finally {
DbUtil.close(stmt, conn);
}
}
private void drainQueue() throws Exception
{
LOG.info(Draining messages);
MessageConsumer consumer = null;
Session session = null;
int numDrained = 0;
try {
session = getHAConnection().createSession(false,
Session.AUTO_ACKNOWLEDGE);
consumer = session.createConsumer(getWorkQueue());
Message msg = null;
while((msg = consumer.receive(5000)) != null) {
++numDrained;
}
} finally {
JMSUtil.close(consumer, session);
}
LOG.info(Drained + numDrained + messages);
}
The output of this when run is as follows (notice that the transaction claims
to have rolledback--and indeed, the separate jdbc update does not happen--but
the messages get sent anyway):
2005-12-06 17:15:08,132 INFO [com.hmsonline.drench.testApp.ejb.TestXABean]
Beginning transaction
2005-12-06 17:15:08,137 INFO [com.hmsonline.drench.testApp.ejb.TestXABean]
Creating Session
2005-12-06 17:15:08,365 INFO [com.hmsonline.drench.testApp.ejb.TestXABean]
Sending 32000 messages
2005-12-06 17:15:12,989 INFO [com.hmsonline.drench.testApp.ejb.TestXABean]
Done sending messages
2005-12-06 17:15:13,009 INFO [com.hmsonline.drench.testApp.ejb.TestXABean]
Saving TestXAState (jdbc)
2005-12-06 17:15:14,859 INFO [com.hmsonline.drench.testApp.ejb.TestXABean]
Committing transaction
2005-12-06 17:15:14,860 INFO [com.hmsonline.drench.testApp.ejb.TestXABean]
TxDebugger beforeCompletion, id , txStr TransactionImpl:XidImpl[FormatId=257,
GlobalId=james.ws.hmsonline.com/13, BranchQual=, localId=13]
2005-12-06 17:18:39,816 WARN [org.jboss.tm.TransactionImpl] XAException:
tx=TransactionImpl:XidImpl[FormatId=257, GlobalId=james.ws.hmsonline.com/13,
BranchQual=, localId=13] errorCode=XAER_NOTA
javax.transaction.xa.XAException
at com.inet.ora.p.prepare(Unknown Source)
at
org.jboss.resource.adapter.jdbc.xa.XAManagedConnection.prepare(XAManagedConnection.java:166)
at
org.jboss.tm.TransactionImpl$Resource.prepare(TransactionImpl.java:2100)
at
org.jboss.tm.TransactionImpl.prepareResources(TransactionImpl.java:1552)
at org.jboss.tm.TransactionImpl.commit(TransactionImpl.java:319)
at org.jboss.tm.TxManager.commit(TxManager.java:200)
at
com.hmsonline.drench.testApp.ejb.TestXABean.runTest(TestXABean.java:155)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at
[JBoss-user] [Messaging, JMS JBossMQ] - Re: xa transaction failure not handled correctly when lots o
i did read the howto on posting questions and i'm not sure where i missed the mark. i don't see how the question is redundant. you were right that i didn't read the release notes for future versions, however i did do extensive googling and other jboss forum searching beforehand to no avail (maybe i picked the wrong terms?). anyway, after going back through the release notes for the 4.0.x versions and looking at all the jms and transaction manager bugs in jira, i think i have half of an answer. i located JBAS-1589 which accounts for why the messages may be commited on rollback, but i see no answers as to why rollback may have started in the first place. did i miss some other issue about transactions failing under load? View the original post : http://www.jboss.com/index.html?module=bbop=viewtopicp=3911058#3911058 Reply to the post : http://www.jboss.com/index.html?module=bbop=postingmode=replyp=3911058 --- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637alloc_id=16865op=click ___ JBoss-user mailing list JBoss-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jboss-user
