Look at the "Not Optimized Stateless SessionBean" in the
testsuite/src/resources/web/META-INF/jboss.xml descriptor:
Not Optimized Stateless SessionBean
org.jboss.proxy.ejb.HomeInterceptor
org.jboss.proxy.SecurityInter
Help! this Exception is thrown in the Remove of some
of my entity beans. It WORKS in Jboss 3.0.0 but NOT in
Jboss 3.0.3 and Jboss 3.0.4
java.lang.IllegalArgumentException:
disconnect(ManagedConnection mc: null, Object c:
org.jboss.resource.adapter.jdbc.local.LocalConnection@70ad76)
called with unk
On Thu, 2003-01-09 at 19:32, Dan Christopherson wrote:
> Probably the second biggest thing (after fear of change) that caused
> ODBMS's to be stillborn: "Will tool> work with it? No!?!?! How do you expect to sell it?"
Which is probably why no-one has developed a commercial solution for the
probl
I am experiencing problems running JBoss-3.0.4 using
PostgreSQL-7.3 as the database for CMP. It appears
PostgreSQL-7.3 doesn't understand "show xactisolevel"
command (which is been sent to it by JBoss) unlike
PostgreSQL-7.2.x. I've deployed the latest jdbc driver
for PostgreSQL-7.3 from jdbc.postg
First check if reordering of getConnection() and getUserTransaction() will solve your problem
javax.resource.cci.ConnectionFactory conFac = (javax.resource.cci.ConnectionFactory)
ctx.lookup ("java:comp/env/HPIARM");
cx = conFac.getConnection();
tran = (UserTransaction) context.get
We are using the protegra tool for jbuilder 7 and it works great til we
tried to add a security realm for the ejb access.
Does anyone know what to get a section
in the jboss.xml file with the protegra tool.
We entered our security realm data in the EJB DD Editor of the EJB Designer,
but there i
I am wondering if anyone has got an EJB working on 3.0.4 with
a property which is stored as a serialized java object. Even the EJBs that
worked for me on 3.0.2 do no longer work on 3.0.4 if they have a property which
is of some Serializable type which is converted to BLOB. I am using Sap DB
I totally agree on the JSTL part for the same reason.
The move to PHP is not really convincing, IMHO.
My concern is: Can you imagine, what a smart marketing guy from
Microsoft could make of this?
It seems that the marketing disaster around Petstore and the TSS
benchmark have not been enough.
Ot
Hello Randahl,
there are BLOB tests in the testsuit that pass successfully.
So, I think your problem is a mapping. Have you tried different
mapping?
alex
Friday, January 10, 2003, 11:34:19 AM, you wrote:
RFI> I am wondering if anyone has got an EJB working on 3.0.4 with a property
RFI> which is
> Have you tried to implemented the PHP Accelerator
> (http://phpa.phorum.org/)? It only takes minutes to install and
> configure and the performance gains are impressive.
>
> Bruce
> --
>
> perl -e 'print
> unpack("u30","<0G)U8V4\@4VYY9&5R\"F9E
why not perl?
"use perl. everybody wants to get
Have you tried using a PHP servlet in JBoss/Jetty?
That way we are still using our own server (eating our own dog food
in Julian speak).
It also may be more efficient as it avoids the CGI thang - but
then it could also be worse?
marc fleury wrote:>
-Original Message-
From: [EMAIL PRO
Well, it is interesting to see that only a few guys are complaining about
the new
jboss.org website.
Moving to a php-driven website as a J2EE server developer is somehow
ridiculous.
It just looks like the JBoss group does not have the confidence in their own
product
and the JAVA technology in gener
just metioned this in jboss-dev yesterday ...
bax
> Von: Greg Wilkins <[EMAIL PROTECTED]>
> Organisation: Mort Bay Consulting
> Antworten an: [EMAIL PROTECTED]
> Datum: Fri, 10 Jan 2003 10:29:52 +
> An: [EMAIL PROTECTED]
> Cc: "'JBossGroup'" <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
> Betreff: R
Randahl,
do not, please, post it to the development list.
The testsuit deals with byte[] instead of java.lang.Object, though.
But, nevertheless, try different mappings first. Try also BLOB instead
of JAVA_OBJECT.
alex
Friday, January 10, 2003, 1:18:46 PM, you wrote:
RFI> That sure is a construc
Randahl,
I guess I'm the "blob" guy at the moment.
I went back and looked at your earlier emails.
The problem is that you are using a jdbc-type of JAVA_OBJECT.
This is an incorrect use of JAVA_OBJECT. Change it to LONGVARBINARY and it should work fine.
We really need to go through the current
On a personal note:
(Using jboss/jboss.net) In terms of good design, say I have a legacy
system that exposes its api as web services, and I wish to use this in
several places in a large deployment (including lots of other integrations,
and cmp where appropriate), what is the community take
Thank you so much Stephen – you are
absolutely right and now it works perfect.
I agree that you definitely need to get
rid of this error in the distribution – I probably won’t be the
only one on the planet spending hours trying to figure out what in the world
went wrong from 3.0.2 to 3.
Did I get this right, you have a web-service layer that behaves like a
(data-centric) resource and you like to
use/hide it from/by a higher-level J2EE logic.
Interesting.
David Jenks told something about the new JCA<->JAXM specification drafted
for J2EE1.4 ... Maybe there could
be design stances
We have developed a multitier app (cmdline clients, webclients, ejb business
layer and db layer). Running all on a jboss 3.0.3; the webclients are
developed using struts 1.1; the complete application can be put in one ear
file.
The access to the webbclients is controlled by a form-based authentica
The beans which worked on 3.0.2 and failed
on 3.0.4 also fails on 3.0.5rc1 – I just tested that.
I am forced to go back to 3.0.2 until the
bug related to fields of type java.lang.Object is fixed.
Randahl
-Original Message-
From:
[EMAIL PROTECTED]
[mailto:[EMAIL PRO
That sure is a constructive suggestion. I currently use the default
mapping which stores serialized java.lang.Object instances as LONG
BINARY - is there a relevant alternative?
Again, the mapping worked fine under 3.0.2 - what changed in 3.0.4?
Would somebody please care to comment on that?
Rand
Hi,
I want to move my application from
JBoss-2.4.6_Tomcat-4.0.3 to jboss-3.0.5RC1_tomcat-4.1.18. Jboss 3 is
starting, I was able to access http://192.168.1.110:8080/jmx-console/index.jsp
jmx-console page.
I copied my war file to deploy (G:\jboss-3.0.5RC1_tomcat-4.1.18\server\default\de
Hi,
I want to move my application from
JBoss-2.4.6_Tomcat-4.0.3 to jboss-3.0.5RC1_tomcat-4.1.18. Jboss 3 is
starting, I was able to access http://192.168.1.110:8080/jmx-console/index.jsp
jmx-console page.
I copied my war file to deploy (G:\jboss-3.0.5RC1_tomcat-4.1.18\server\default\dep
On Thu, Jan 09, 2003 at 11:16:38PM -0500, sherlock costello wrote:
> >Does anybody (other than me) find it ironic that the leading open source
> >application server project uses PHP on it's home page instead of JSP?
> Absolutely. I think it's insane.
Not to fuel the flames thus far (it seems li
Hi MS,
I understand with CMP you can have it create a table if it does not already exist. I'm
looking for the case where there is an existing table, with data in it, and need to
add new fields to the table.
thanks,
-D
Hello,
Redeploying the entity bean should take care of creating the table
You are just dreaming ;)
Entity bean will extend your Db vertically (i.e. add new rows) but not
horizontally (add new columns). Nevertheless:
- you could implement a BMP that does that (good luck) but your changes
won't necessarly be transational (depends on your db)
- if you redeploy your bean
reordering did fix the problem, thanks. however, shouldn't my
previous ordering have worked as well? i would expect that in a
CMT case, the transaction is begun and committed outside the calls to
getConnection/close. i would expect this ordering to work in a BMT case as
well.
i've notic
We are in the process of creating our own CMS framework. Actually a direct
port of PostNuke. When that is complete we'll switch back to Java land.
Bill
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Christopher
> Blunck
> Sent: Friday, January 10
SL> - if you redeploy your bean and use the create-table feature of CMP, JBoss
SL> will re-create the table (but not update it => you loose your data).
No, JBoss won't recreate the table if remove-table wasn't set to true.
The table will remain the same. Thus, if entity's structure/definition
cha
Let's try viewing this from a different angle:
How much fun would we have with the .NET
advocates if www.microsoft.com was built
using PHP?
/Goran Andersson
---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + Lin
Hi,
I created a custom implementation of the Schedulable interface.
I tried deploying it as a SAR with the Schedulable implementation included
or a scheduler-service.xml with the implemention in the JBoss lib directory.
scheduler-plugin.jar is in lib, too. Either way I'm getting the stacktrace
be
If is not ask to much, can we keep this as an help
system? If you like you tool (any name you like)
better use it.
Thank you,
John Mello
--- Göran_Andersson <[EMAIL PROTECTED]>
wrote:
> Let's try viewing this from a different angle:
>
> How much fun would we have with the .NET
> advocates if
We have been using a home-made solution where a pre-deployer reads an xml-file from
the jar-file containing any db-script to be executed before the actual deployment of
the ejb. It works ok. BUT... the BeanShell solution looks very interesting...
/Lennart
- Original Message -
From: "Ale
In the early days of Java, I read that a large retailer had about 70 Java
programmers because it was "a business decision". It was the only tool
available that they could use to integrate their systems.
Most likely, Marc's decision to use PHP on the website is a business decision
as well. The
hi,
in a
previous post with subject "[JBoss-user] "can't enlist" error on second pass
through resource adapter" i described a problem i was having with multiple,
non-concurrent invocations of a resource adapter via a stateless session
bean. the responder's suggestion (swapping the order o
Attached file:
On Fri, 2003-01-10 at 19:43, Allan Kamau wrote:
> I am experiencing problems running JBoss-3.0.4 using
> PostgreSQL-7.3 as the database for CMP. It appears
> PostgreSQL-7.3 doesn't understand "show xactisolevel"
> command (which is been sent to it by JBoss) unlike
> PostgreSQL-7.2.x. I've deployed
> From: Pete Beck [mailto:[EMAIL PROTECTED]]
> Sent: Friday, January 10, 2003 12:36 AM
> To: [EMAIL PROTECTED]
> Subject: Re: [JBoss-user] Cmp vs hibernate
>
> On Thu, 2003-01-09 at 19:32, Dan Christopherson wrote:
> > Probably the second biggest thing (after fear of change) that caused
> > ODBMS'
1. Has been added to 3.0.5.
Scott Stark
Chief Technology Officer
JBoss Group, LLC
- Original Message -
From: "Brian Topping" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, January 07, 2003 2:06 AM
Subject: RE: [JBoss-user] [no
It a simple lock inspection monitor as described in the admin/devel docs:
The EntityLockMonitor has no configurable attributes. It does have the following
read-only attributes:
-MedianWaitTime: The median value of all times threads had to wait to acquire a lock.
-AverageContenders: The ratio of th
Its been added for 3.0.5
Scott Stark
Chief Technology Officer
JBoss Group, LLC
- Original Message -
From: "Meyer-Willner, Bernhard" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, January 09, 2003 9:02 AM
Subject: Re: [JBoss-us
That you redeployed the ejb being called without redeploying the calling context
and the class versions are out of date.
Scott Stark
Chief Technology Officer
JBoss Group, LLC
- Original Message -
From: "Matthew Van Horn" <[EMAIL PROTECTED
You use the bundled Tomcat or Jetty containers and use the same security-domain
value in the ejbs as the web app and do nothing else.
Scott Stark
Chief Technology Officer
JBoss Group, LLC
- Original Message -
From: "Scheil, Sven" <[EMAIL
I may be going out on a limb here, but I think Sven's question may be along the same lines of what I have been wondering about, which is:
If you do not lock down the servlets (i.e. no security) but you want to collect username and credential information in the web layer, how do you create and propa
On Saturday 04 January 2003 04:48, Pete Beck wrote:
> I agree with Sun 100% on this. Using the file-system is bad news for
> maintenance, scalability and as the article says security.
> I've seen the chaos that using the filesystem can cause in clustered
> environments and I would say avoid it if
Then you have to behave as any other client
accessing a secured ejb, and
do a JAAS login. There will be no principal in general on an unsecured page
though.
Scott StarkChief Technology
OfficerJBoss Group, LLC
- Original Message -
On Thursday 09 January 2003 21:53, Sundaram Ramasamy wrote:
> Today I visited jboss.org web site, it looks different (with php . I don't
> know much about php.
I think it is one of the most unprofessional and abnoxious sites I've seen in
a long time. I preferred the old site much better. They n
Your target Schedulable is causing an OutOfMemoryError in its ctor I would say.
The msg says it find the ctor, OR create the instance. The create is failing due to
the OutOfMemoryError.
Scott Stark
Chief Technology Officer
JBoss Group, LLC
- O
"I think it is one of the most unprofessional and abnoxious sites I've seen
in a long time."
Unprofessional and abnoxious (guess that's like obnoxious) works for me but
that achievement notwithstanding the project would be better off with a
fully buzzword compliant J2EE application front to back.
I've not read Sun's argument, but I've never understood the rationale for
storing large invariant data in an RDBMS. Or course the RDBMS vendor wants
you to do it - they want the whole world stored under their product. And
the DASD vendors love it. But what does it buy you, the user who implements
50 matches
Mail list logo
