Hi Heiko,
you are exactly right, modifying xdoclet was the solutiuon.
(some details on how I did it are in my reply to Scotts mail)
Regards,
Michael
Rupp, Heiko wrote:
Hi,
I did not completely understand, what you are trying to
achieve. If you have a standard MBean, you can just add
an xmbean-dd
lete xml sequence
behind &defops; and friends.
Regards,
Michael Bartmann
Scott M Stark wrote:
The resource type for an xmbean does not affect its management
interface.
I'm not familar with the xmbean xdoclet implementation, so you'll have
to drill into that. The jboss-service.xml des
ld be done.
btw: Our 24x7 MES systems with hot deployment of jms
protocol stacks and corresponding mdbs run stable now for
one year with impressive uptimes. the mallorca vision
became reality.
Regards,
Michael Bartmann, 4Production AG
Marc Fleury wrote:
Michael bartman from 4production?
Huh...
(
in the
xxx-service.xml. I found no such attribute.
Regards,
Michael Bartmann
---
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February
ecify that an mbean depends on a jar
(which I remember (again I might be absolutely wrong) once worked this way).
Or is there another way to define the dependency?
Regards,
Michael Bartmann
---
The SF.Net email is sponsored by EclipseCon 2004
Pre
d under CMP2.0
w/o help of the container, because you cannot overwrite the
abstract setters.
Regards,
Michael Bartmann
Meyer-Willner, Bernhard wrote:
Thanks for your info. However, I already knew this. Unfortunately we can't
use this feature because our DBA dislikes it and certain triggers
"myQueue" seems to be your queues name.
"queue/myQueue" is a possible JNDI name (depends on your config)
"jboss.mq.destination:service=Queue,name=myQueue" is a possible ObjectName. (depends on your config)
Regards,
Michael Bartmann
G.L. Grobe wrote:
I'm gett
r
postprocessing.
So are we speaking only of a transformation engine
mapping eql/jbossql to db specific sql, as would suffice
in the case of UPDATES, where there is no need to
postprocess anyway?
Regards,
Michael Bartmann
Dain Sundstrom wrote:
Pete,
To better support complex reporting we just
p didn't charge the bug fix itself; they more or less
charge for your problems specific work (isolating the bug from application
or configuration flaws) and for backporting to older releases, if you
depend on them.
Regards,
Michael
Dr.-Ing. Michael Bartmann
Leitun
eed it into a jdbc statement.
I am aware that "update" has no meaning for eql finders, but I hope you get the idea.
Regards,
Michael
Dain Sundstrom wrote:
Now I am completely confused. What exactly do you want that you can't
do today?
-dain
Michael Bartmann wrote:
Dain,
I'
ll of you data to fit into memory and actually be in
memory. If you want to write it, that would be cool, but I think
there are way more important things to make CMP fast.
-dain
Michael Bartmann wrote:
Comments inline,
Regards,
Michael
saroj kumar wrote:
Running EJB-QL without Container
mixture of entities and plain
sql in the
application, so thing can only get better.
This becomes crucial if the changes to DB are quite frequent.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:jboss-user-admin@;lists.sourceforge.net] On Behalf Of Michael
Bartmann
Sent: Wednesday
Absolutely!
I don't have to add much to this, but I think that this request
(non-container ejb-ql) is so important (and a good "workaround",
too), that I simply couldn't resist to reply.
Anyway, would this be difficult to achieve?
Regards,
Michael Bartmann
Pete Beck wrot
Hi everybody,
I remember having heard of an ant task which
talks directly to the jboss deployer mbean, so
you can get along w/o filesystem copies.
Or does it only provide an url to the deployer?...
Can anybody provide me whith a hint where to
look for the thing?
Regards,
Michael Bartmann
Hi,
it is essential that the JAVA_HOME points to the same jdk which has its
bin directory first in the PATH.
Just an idea..
Michael
[EMAIL PROTECTED] wrote:
Marius Kotsbak wrote:
Maybe you have to set JAVA_HOME to the rootdir of your JDK, or put
java-binary in the path.
Thanks Marius for th
I doubt that this is the problem.
I often run it as an ordinary user
under Linux, which has lower ports
reserver for root, too.
Just my 2c,
Michael Bartmann
JD Brennan wrote:
*It's probably trying to listen on a low numbered port. I think*
*all ports below 1024 require root priv on
I had the same nasty effect when running on a SuSE 8.1 Linux
box using NIS, which did not resolve localhost to 127.0.0.1.
Just my 2c,
Michael Bartmann
Thorbjørn Ravn Andersen wrote:
Luttrell, Peter wrote:
Has anyone had success running JBoss on Solaris?
Yes. We run on a Solaris 7 machine
Jim Crossley wrote:
Upon re-reading this, I got confused...
On Thu, 2002-10-24 at 15:40, Michael Bartmann wrote:
{
// Entity bean = home.create(...)
// bean.setX(...)
this starts a tx1 and locks the instance bean.
Yes, but wouldn't the lock be released as soon as setX returns???
markup.
Change "RequiresNew" to "Requires", so the same tx will get used
throughout your call.
Thanks,
Jim
On Thu, 2002-10-24 at 15:40, Michael Bartmann wrote:
Explanation attempt inline.
Regards,
Michael Bartmann
Jim Crossley wrote:
I have an MDB with the following
Explanation attempt inline.
Regards,
Michael Bartmann
Jim Crossley wrote:
I have an MDB with the following pseudocode in its onMessage method:
public void onMessage(Message msg)
{
// Entity bean = home.create(...)
// bean.setX(...)
this starts a tx1 and locks the instance bean
Hi,
the recent change in CVS structure means you cannot use the
same module (was: jboss-all) for any branch/revision.
If you checkout e.g. some revision of Branch_3_2 you must use
the module jboss-3.2 (as some relevant cvs metadata cannot
be versioned).
What your question refers to is the differen
I got exactly these errors when forgetting to select the correct branch.
the module alone isn't enough it sees :-)
cvs checkout -r Branch_3_2 jboss-3.2
worked for me.
[EMAIL PROTECTED] wrote:
If I check out jboss-head with cvs I can enter the build directory and
run the build.sh command and th
One design is to use mbeans which trigger message driven beans through jms
Regards,
Michael
saroj kumar wrote:
>Hi All,
>
>This problem is resolved now.
>
>We had a while loop which was checking for some boolean value. This
>boolean
>Was updated by worker threads.
>
>This update is done in a sy
e have not eliminated problems on our side, but as the transaction
markup should dictate
(all of it in "REQUIRED" methods) everything should run under one
transaction.
We have Adrian (who is our official supporter) told about our problems.
Just my 2c,
Michael Bartmann
Sacha Labourey wrote:
or even to 2.4.x where we had the same problem...
Michael
Michael Bartmann wrote:
> Adrian found something wrt 3.2 this very morning.
> His solution might be applicable to 3.0?
>
> I just want to avoid double work, thanks,
>
> Michael
>
> Bill Burke wrote:
>
&
Adrian found something wrt 3.2 this very morning.
His solution might be applicable to 3.0?
I just want to avoid double work, thanks,
Michael
Bill Burke wrote:
>Can you log a bug on this, copy this email, and assign to me?
>
>patriot1burke
>
>Thanks,
>
>Bill
>
>
>
>>-Original Message-
Message-
>From: [EMAIL PROTECTED]
>[mailto:[EMAIL PROTECTED]]On Behalf Of Michael
>Bartmann
>Sent: Wednesday, October 09, 2002 4:59 AM
>To: [EMAIL PROTECTED]
>Subject: Re: [JBoss-user] Memory Leaks
>
>
>We had/have big problems with 3.2, some of which should b
now these issues went
away. But of course you should consider this as a bug on our side, not
on jboss's side.
Regards,
Michael Bartmann
[EMAIL PROTECTED] wrote:
>We also experience the same problems using JB 3.0.2 and JDK 1.4.0 from sun.
>
>On Wed, Oct 09, 2002 at 09:31:49AM +
Hi Thorbjørn,
what works for me is: http://www.mail-archive.com/lists.html.
Regards,
Michael
Thorbjørn Ravn Andersen wrote:
> I have a rather tricky problem with jboss-web.xml where a
> does not show up in JNDI for one webapp, although the identical file
> shows up in another webapp in the s
Hi Alexey,
we today encountered this problem (for the first time
as far as I know) after a fresh update on Branch_3_2.
Only to let you know, we had no time to dig into it...
Regards,
Michael Bartmann
Alexey Yudichev wrote:
> From that I cannot see much complains about the above except
[] which are anytime thereafter efficiently transfered to the client
through an incremental bulk-loading mechanism.
I have a client code fragment which replicates the server data to a
javax.swing.table.TableModel. Does anybody know a good _free_ graphing
package I could plug this into?
Mich
es?
What other caches do I have to flush...
Enjoy,
Michael Bartmann
me with a IOException (file not found).
Now what is "the right way" of doing this (preferably working under 3.2 and 4.0)?
Or have I found a bug in 4.0?
(I assume you don't want to ask me to put the myresource.xml in the root of the
yyy.ear?
This would hurt my build process in an ug
ployed anything during the last 200MB of
memory growth.
But on some other occasion I've tested redeployment and it's memory effect:
after 50 redeploys of one of our ears with enclosed JMS Provider and
container config
there were only some 5 MB not reclaimed when doing
uot; threadgroup
There are 21 unnamed threads in the "ASF Session Pool"
The "real" usage is approx. 5 (!) clients connected (thick Swing
clients), and
4 MDBs working on their own ear-scoped JMS Provider.
Does anybody see
ta) ?
- is there some way to influence (flush) caches through the jmx-console?
Regards,
Michael Bartmann
PS.: It should become obvious why I kicked of the "health monitoring"
discussion ;-)
---
This sf.net email is sponsored by:Think
fig shows that the
_XAConnectionFactory_ of both services get registered under the same name.
The MDB Container gets its Queues through a JNDI-lookup of these Factories.
I think that you must name them differently.
Michael Bartmann
--
other? The ProviderLoader
obtains a Connection after looking up these Factories, so how could this work?
Just my 2c
Michael Bartmann
Barlow, Dustin wrote:
> I didn't, but instead used the default one that was already setup in the
> standardjboss.xml as I didn't see any difference b
Michael Bartmann answered to quick when he wrote:
> Did you also copy the MBean with the ServerSessionPool from my
> ...-service.xml?
>
Oops, I did not see your comment below the MBean config :-)
I think that a dedicated ServerSessionPool is necessary.
At least I had it that way fo
Did you also copy the MBean with the ServerSessionPool from my ...-service.xml?
Michael Bartmann
Barlow, Dustin wrote:
> Ok, I did what you suggested, and also looked at Michael Bartmann's
> configuration he was kind enough to forward to me, but it's still now
> working. Jus
in a
way,
that the TCL is the right one by making things protected?
Or maybe I don't think about it in the right way?
Michael Bartmann
---
This sf.net email is sponsored by:ThinkGeek
Welcome to g
I'll try to RTFM, but then again...
...surely you can tell me...
Thanks,
Michael Bartmann
---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
there, or shouldn't this be done
when I call MBeanServer.registerMBean.
Hmmmm...
Regards,
Michael Bartmann
---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_
Hi Adrian,
Your workaround sounds ok, but I have some 20 ear-scoped applications, with
the same single offending method in each, so I decided to stay with my
serialization
"below the API" with a byte[] arg. (I can do this by myself, but the
management
GUI is written by Mehrdad (remember him?),
k?
Thanks,
Michael Bartmann
PS.:
The whole thing of course worked when ClassB is only the return type of
a method,
because the server only has to serialize an already loaded class.
---
This sf.net email is sponsored by: OSDN - Tired of tha
I should have mentioned that all this is
on Branch_3_2 (last time updated: friday).
Michael Bartmann
---
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone? Get a new here for FREE!
https://www.inphonic.com/r.asp?r
I should have been more specific in one point:
The container classloaders which are null are MDB containers.
(The dependent ears in my case only have MDBs; i don't know
if this problem is exactly MDB-specific)
Thanks,
Michael Bar
But when afterwards the create-method is called these classloaders are
still null and
the redeployment fails whith a NPE when the classloader is accessed.
All this happens in a quite complex application; perhaps the description
of the problem
is clear enough, so I don't have to setup
Last month one of our customers called in and complained about
all his JBoss-clients not working. We dialed in to his JBoss-console
and saw that his JBoss had unexpectedly shut down.
The server.log looked exactly the same way as if someone had
pressed Crtl-C on the console.
We a
I once had this effect when not the following setting before calling
run.bat / run.sh:
set JAVA_HOME=...
set PATH= (with JAVA_HOME/bin in it, before any other JDK)
Enjoy,
Michael
---
This sf.net email is sponsored by: OSDN - Tired of th
I have a MBean registering serveral sub-services as MBeans.
These sub-services are registered programmatically, not through normal deployment.
They are normally also unregisterd programmatically.
If this fails, I like to unregister them through the console, or
they stay forever...
So I use the con
the system having the above mentioned problems uses a
version from before your fix, my problem is not double deployment
but seems to be deployment of classes in the same jar by different
loaders of the same repository.
But I will surely try to test an up to date version on monday.
Greeti
David Jencks wrote:
> On 2002.08.31 08:12:17 -0400 Michael Bartmann wrote:
>
>>David Jencks wrote:
>>
>>
>>>The message warns you that if your transaction fails your data may be
>>
>>left
>>
>>>in an inconsistent state. You have more t
want to add this:
Each class of the package for which package private access failed
was in the same single jar.
Michael
---
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone? Get a new here for FREE!
https://www.in
Scott M Stark wrote:
> Yes, the only way you can guarentee package private access is to
> include the classes from the package in the same jar. See bug #602828.
>
They *were* in one jar, but this jar was references from more than
one MANIFESTs -> booom.
I think the problem is that not the per-ja
jboss-development.
thank you,
michael
Dr.-Ing. Michael Bartmann
Leitung Produktentwicklung
*
4Production AG
Adenauerstr. 20
D-52146 Würselen
*
Telefon +49 (24 05) 41 35 26
Telefax +49 (24 05) 41
Scott M Stark wrote:
> Its not an issue of visibility, its that two classes from the same package
> can be loaded by different class loaders and package private access
> fails. I have seen this and the easiest way to cause it is to split the package
> classes across two jars. An ear should use a c
David Jencks wrote:
> This is on a jboss 3 version, right?
This is on a two weeks old version of Branch_3_2 (after
the arrayclass-fix)
>
> I'm very surprised. Everything in your .ear should be visible to
> everything else: it should all be loaded by classloaders in one loader
> repository.
Ever
at this point since in order to recover from
> such failures you would need to run tyrex as your tm and use a jms provider
> that supported recovery which I believe jbossmq does not.
I use my own JMS provider, which supports XA transactions
but doesn't handle recovery either ;-(
>
>
[EMAIL PROTECTED] wrote:
>>When I got back form XA-tx-JMS to local-tx-JMS these messages went away.
>
> But what is the consequece of doing that? I think you then miss the transactions
>involving both JMS and DB. A better solution is to use a XA-enabled datasoure.
Ok, I should have made this c
Hi,
yesterday I had a (temporary) problem with the EAR-scoping:
involved were
- some common-xxx.jars with common code,
- some ejb-xxx.jars all using this common code,
- all of this in one big ear with ear-scoping.
Fine so far, all of this had run before without ear-scoping.
But my application.x
G.L. Grobe wrote:
> Anyone know what this warning means? It doesn't seem to be affecting my
> app and the update actually works. This update is done within a MDB.
>
> 23:09:29,546 WARN
> [LocalTxConnectionManager$LocalConnectionEventListener] prepare called
> on a local tx. You are not gett
production environment.
We are about to integrate hot-redeployment of EAR-scoped
modules combining EJBs and JMS MBeans without service
interruption and have first promising results.
Enjoy,
Michael
Dr.-Ing. Michael Bartmann
Leitung Produktentwicklung
xxxxxxxx
>- Original Message -
>From: "Michael Bartmann" <[EMAIL PROTECTED]>
>To: <[EMAIL PROTECTED]>
>Sent: Wednesday, August 21, 2002 8:05 AM
>Subject: Re: [JBoss-user] Run-As Problems?
>
>
>
>
>>Hi,
>>
>>further stu
hardcoded
username for he MDB. *This works*. But I'd like to be "dynamic" and use
the xml metadata to specify the run-as-role through the .xml-files.
Michael Bartmann wrote:
> Hi everyone,
> are there any known issues with the run-as mechanism under Branch_3_2?
> We use this
Hi everyone,
are there any known issues with the run-as mechanism under Branch_3_2?
We use this for the onMessage-method and tried to put the
RunAsSecurityInterceptor in the interceptor chain (of the MDB and the
SessionBean),
but to no avail.
Enjoy,
Michael
Hi,
I am developing an application which uses Borland-components on the
server-side.
(EJB Dataset Sessions)
This worked fine so far (under JBoss 2.4.x), but now (under 3.x) the
Classloader
complains. Borland uses Class.forName() where I think
Thread.currentThread().getContextClassLoader() would
Andreas Kuckartz wrote:
> building jboss from cvsI also have trouble to build JBoss using the latest
> source code version on a Windows 2000 machine.
>
Co-workers of me who also use Windows 2000 and Java have told me of some
obscure dll-conflict; they had to remove a certain dll but I dont recal
Uroš Jurglič wrote:
> I have all environment variables set and I use suns jdk1.3.1_03.
> Now that you've told me, I've used build.bat, but again not
> successfully. Build batch returns in no-time and prints nothing. Not
> very usefull for finding out what's wrong though.
I never had the build p
Uroš Jurglič wrote:
> I downloaded all the modules from cvs, but am having troubles building
> any of them. I get the following error:
> BUILD FAILED
> Error reading project file: unknown protocol: resource
>
> Are there any general docs about building jboss?
>
>
> Tnx.
>
Most of the time I h
70 matches
Mail list logo
