read/write
lock strategy is used.
Is this correct?
On Wed, 2003-12-10 at 19:05, Bill Burke wrote:
> Cool. So you actually found the EntityLockMonitor useful?
>
> Bill
>
> Pete Beck wrote:
>
> > Hi everyone.
> >
> > For the benefit of mankind, I've just w
uture deadlockees some time.
Kind regards,
Pete Beck
--
Peter Beck BEng (hons) MIEE GIinstIT - Managing Director, Electrostrata Ltd.
http://www.electrostrata.com --+-+-- Experts in e-business and e-commerce
---
This SF.net email is sponsor
Aha! :-)
This would explain why I am getting locking even when I'm not expecting
it.
Thanks guys
On Tue, 2003-11-25 at 17:37, Alexey Loubyansky wrote:
> The read-only is applied to the specific bean/method as defined in the
> DDs. No inheritance rules.
--
Peter Beck BEng (hons) MIEE GIinstIT -
On Tue, 2003-11-25 at 11:26, Kevin Conner wrote:
> We stuck with the pessimistic locking and enabled read-only methods.
> The CMP engine grabs the lock for the duration of the method but
> releases it afterwards.
>
> Kev
>
> BTW We are on a modified version of 3.0.2 but I think it still
ance per transaction.
Therefore is there any way to use optimistic locking with commit option
A?
Thanks,
Pete Beck
--
Peter Beck BEng (hons) MIEE GIinstIT - Managing Director, Electrostrata Ltd.
http://www.electrostrata.com --+-+-- Experts in e-bus
On Fri, 2003-10-17 at 01:16, Ken Yee wrote:
> Thanks for the informative post on your experience w/ Hibernate :-)
> Is there any part of your web site that you can see it in action?
> Looks like you actually have to log in to see anything.
The fact you didn't notice it, shows how quick it works. :
I read with some excitement about the marriage between JBoss and the
Hibernate project.
This seems like a good move for both projects, particularly with regards
to JDO compliance.
By sheer coincidence, I have just completed a project (http://i3t.org)
which actually makes use of a JBoss/Hibernate
rests of all of you.
I apologise for the length of the email (if anyone has read this far),
but this is a deadly serious issue. The conclusions that the decision
makers come to at this point will determine the future viability of
open-source enterprise projects and maybe even open-source itself.
Kind
I'm just guessing, but perhaps throwing the extra memory and threads
at the VM made it run faster.
If ther VM runs faster, then the transactions will be shorter which
means there will be less lock contention.
On Tue, 2003-07-29 at 15:51, Christofer Dutz wrote:
> PS: Has anyone got an idea, why t
Has anyone integrated Apache Lucene into JBoss?
What would be the best way to do it?
Is there JCA adaptor?
Thanks,
Pete
--
Peter Beck BEng (hons) MIEE - Managing Director, Electrostrata Ltd.
http://www.electrostrata.com --+-+-- Experts in e-business and e-commerce
-
"Enhancement at deployment time"
I have one word: wow!
Yet again JBoss kicks some backsides with its deployment process. This
looks like some great work guys; I can't wait for inheritance and more
JDOQL support. :-)
On Wed, 2003-07-02 at 18:58, Alexey Loubyansky wrote:
> http://jboss.org/index
I think the confusion arises where people people assume EJB's were
designed for object persistance.
They weren't, and in fact the persistance stuff was probably added more
as an afterthought than anything.
The technology emerged at the time where distributed object frameworks
like CORBA and DCOM
Try setting:
true
On Tue, 2003-07-01 at 10:55, Alexey Loubyansky wrote:
> Hello Janardhan,
>
> JAWS was replaced with JBossCMP a long time ago and is not supported
> anymore.
> Sorry, I really don't know what is wrong with your finder.
>
> alex
>
> Tuesday, July 01, 2003, 12:03:42 PM, Janard
ed.
> I will keep this page up to date.
> And, remember, it is the first iteration.
>
> alex
>
> Monday, June 30, 2003, 9:51:36 PM, Pete Beck wrote:
>
> PB> Hi guys,
> PB> Will the docs identify what *isn't* implemented yet? I think Alex
> PB> mentio
Hi guys,
Will the docs identify what *isn't* implemented yet? I think Alex
mentioned that it doesn't support inheritance yet for example.
On Mon, 2003-06-30 at 19:15, Bill Burke wrote:
> It will be officially released tomorrow or Wednesday. Alex should be
> putting up docs momentarily. It won't
It's also worth pointing out the usual disclaimer that invoking
executables is generally a non-portable activity.
Unless you tread carefully, you may eliminate one of the main benefits
of using Java in the first place.
On Mon, 2003-06-30 at 14:09, David Ward wrote:
> Manoj,
>
> java.lang.Runtime
I did something like this too.
I can't remember how I did it, but you can also put your app in a
separate .war file and use the sitemap rules in the cocoon config file
to forward requests to the correct context path.
The advantage of this is that, you don't have to repackage the
cocoon.war everyt
Hi Mark,
If you're using inheritance a lot in your design, you might be
interested in JDO which seems to have better support for it than CMP
2.x.
Alex Loubyansky has made some headway on a JBoss JDO project which uses
the new AOP features in JBoss 4.0. For me this is a very interesting
development
Recently I found a thread on a forum that discussed the virtues of JDO
over Entity Beans.
There was a very good argument that in most cases Entity Beans aren't
needed.
The main things that entity beans provide over JDO are declarative
security, remote access and automatic transaction demarcation.
On Wed, 2003-04-02 at 01:49, [EMAIL PROTECTED] wrote:
> I'm doing this work on my limited spare time. If I had the time to
> contribute, I would. As it stands, I don't see JBoss as being a
> production ready tool.
If nothing else, you're wrong about this.
We're successfully using it in production
On Tue, 2003-03-18 at 23:09, Philippe de M. Sevestre wrote:
> How often does the data from which you populate your pull-down changes ?
>
> Usually this kind of information is relatively static so you should
> consider caching the resulting bean collection in a session attribute (if
> user specific
Hi,
My application frequently needs to grab a large number of entities from
the database to populate drop down lists.
The current method I use, is to use a CMP finder which returns a bunch
of entities. I then iterate through the list grabbing the values I need
to populate the list.
But, it is to
Maybe I've misunderstood your post, but sounds like you want the
Value-Object pattern.
Have a look at the J2EE design patterns on Sun's website.
On Wed, 2003-03-05 at 17:23, costin wrote:
> Hi!
>
> Never tried this again and haven`t seen any info. I would appreciate
> any info on this one.
On Wed, 2003-03-05 at 05:14, Russell Black wrote:
> My client is concerned about JBoss's scalability. Just wondering if
> any of you have used JBoss in a heavily-used production environment,
> e.g. lots of requests, and a large database. How well does it hold
> up?
>
> Thanks,
> Russell
>
> P
Or even better, keep the old application deployed while the new
application is being deployed.
On Fri, 2003-02-28 at 13:02, David Corbin wrote:
> The one thing I'd like to see jboss provide. When I an application is
> redeployed, I'd like to see JBoss accept requests for the context of the
> a
Yes awstats.
It works directly with the log files produced by Jetty in the JBoss
distribution.
On Thu, 2003-02-27 at 13:39, Stefan Groschupf wrote:
> Hi,
>
> someone had setup jetty to produce logfiles that are processable by
> webalizer?
> Or configure webalizer to read jetty logfile?
>
> http:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> ___
> JBoss-user mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/jboss-user
--
Pete Beck <[EMAIL PROTECTED]>
Electrostrata Ltd.
--
On Tue, 2003-02-25 at 13:44, David Klimek wrote:
> Have you got any experience with JVMPI slowdown, is that true?
Yes.
--
Peter Beck BEng (hons) - Managing Director, Electrostrata Ltd.
http://www.electrostrata.com --+-+-- Experts in e-business and e-commerce
--
On Wed, 2003-02-19 at 15:07, Bill Burke wrote:
> J2EE is our bread and butter. We will still be improving our J2EE
> implementation and strictly following the J2EE specifications. If you don't
> want to use our new AOP services, then don't use them.
The use of AOP sounds very interesting. Do yo
On Tue, 2003-02-18 at 22:31, Bill Burke wrote:
> So the goals of 4.0 are as follows:
...
> 2. Bring J2EE services to plain old java classes through our AOP framework
I've seen this mentioned elsewhere, but I don't quite understand it.
Are you saying that for example, you would be able to take an
On Thu, 2003-02-13 at 15:01, Michael Bartmann wrote:
> I think the deferred update might also be relevant for performane.
>
> Latest 3.2 has "Container Managed Audit Fields".
> E.g. it can automatically update a "changeUser" and "changeTime" field.
> (See chanage note by Adrian Brocke)
>
> This i
On Wed, 2003-01-29 at 10:26, Sacha Labourey wrote:
> Well, that would involve:
> - loading all potential beans in memory first
> - having an in-memory query resolver (able to resolve SQL queries
> in-memory)
>
> You want to do it? ;)
I think I would *love* this feature, but like me, I don't thi
Actually this sounds like exactly what I was after.
Thanks.
On Fri, 2003-01-24 at 16:02, Anders Engström wrote:
> Correct me if I'm dead wrong - but this sounds like DAO.
>
> http://java.sun.com/blueprints/patterns/DAO.html
>
> //Anders
--
Peter Beck BEng (hons) - Managing Director, Electros
Recently we had an interesting thread about the realtive qualities of
Hibernate and CMP.
One of the points that I picked up from the discussion, is that if you
are only using CMP entities for persistent O/R based objects, then
lightweight systems like Hibernate may provide a better and faster
alte
On Sat, 2003-01-18 at 16:25, Jules Gosnell wrote:
> I haven't tried it, but if you have a Jasper that implements JSP2.0, try
> replacing the jasper jars in jbossweb.sar with it and see if it works...
>
> If not, they have changed the external API and we may have to make
> adjustments. I don't s
Anyone know what sort of timeframe we might be able to use JSP 2.0 in
JBoss?
--
Peter Beck BEng (hons) - Managing Director, Electrostrata Ltd.
http://www.electrostrata.com --+-+-- Experts in e-business and e-commerce
---
This SF.NET email
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0026en
> ___
> JBoss-user mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/jboss-user
--
Pete Beck <[EMAIL PROTECTED]>
Electrostrata Ltd.
On Tue, 2003-01-14 at 11:30, Sacha Labourey wrote:
> It is is 3.0.x as well.
>
> Cheers,
>
>
> sacha
That's perfect. I'll take a look in the DTD.
Thanks guys!
Pete
--
Peter Beck BEng (hons) - Managing Director, Electrostrata Ltd.
http://www.electrostrata.com --+-+-- Experts in e-busine
On Tue, 2003-01-14 at 13:22, Muntean Horia wrote:
> I think you can declare "read-only" descriptors at method-level on
> jboss.xml. But AFAIK this feature is available only in JBoss-3.2.0.
> You can use this approach with QueuedPessimisticLock manager and commit
> option A and your EJB won't be
Dear JBoss gurus,
I've noticed that JBoss aggressively locks beans in a pessimistic
locking strategy.
I've read that optimistic locking is on the way, which sounds like a
good thing.
The problem I've found with the current locking strategy is that it is
very easy to get deadlocks as the load on t
On Tue, 2003-01-14 at 04:32, Dain Sundstrom wrote:
> Postgres can't store millisecond accurate time until 7.0. Before that
> it truncates it to a decasecond if I remember.
>
> -dain
Besides which, you're crazy if your not using a more recent version of
Postgres as since 7.1 it has improved dram
On Mon, 2003-01-13 at 08:41, Camus Chan wrote:
> Hi,
>
> I am using JBoss3.0.3 + Tomcat 4.1.12 + Apache 2.0.43, with oracle8i and nls_lang
>set to ZHT16BIG5. I face the character encoding problem and please help:
There are several things to check:
1) The character set your Oracle database is bu
On Fri, 2003-01-10 at 22:44, Schnitzer, Jeff wrote:
> > Which is probably why no-one has developed a commercial solution for
> the
> > problem.
>
> Eh? What was Gemstone/J?
I have no idea. It is not a product that I am familiar with. But if it
is as you suggest, a native CMP store such as the
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
That sounds very high, my messages are handled almost instantly. Do you
have a lot of load on your server?
On Thu, 2003-01-09 at 14:09, Gabriel Esteves Marques Pinto wrote:
> Hello
>
> Can anyone help me with this?
>
> How to reduce the delay from MDBs ?
> It takes almost 3 minutes to read the
On Thu, 2003-01-09 at 14:02, Matthew Baird wrote:
> so you are promoting using what?
No promotion I am afraid, merely an observation.
Comparing CMP with persistent object frameworks is not really a fair
comparison.
Comparisons are odourous as Shakespeare once wrote. :-)
In my mind, it is like c
On Wed, 2003-01-08 at 22:45, Matthew Baird wrote:
> I disagree with your statement of
> "Let's face it, O/R mappings will never be very efficient. "
>
> Both OJB (to which I contribute) and Hibernate (which I watch closely)
> approximate raw JDBC performance in most cases, and can exceed it in
On Fri, 2003-01-03 at 19:37, Luttrell, Peter wrote:
> Actually i'm considering writing a little jmx service that manages the
> filesystem store. my ejbs would store the key that the jmx service
> requires. the service would then enforce/handle all such rules.
>
> I just read this however, which a
A common usage for CMP entity beans seems to be for persistent objects
using an O/R mapping.
Although entity beans can generally act that way, that is by no means
all they amount to.
In fact Sun have a more light-weight spec for this purpose (JDO). I
can't help feeling that hybernate is nearer t
After over a year in development our JBoss/Jetty based site is now live.
Check out http://www.factory121.com
I would like to take this opportunity to express my gratitute and
appreciation to all those who have contributed to the JBoss and Jetty
projects. Without their excellent work, I am sure o
ong in my reasoning, but the original message
> says explicitly "ONLY dynamic applications". So, you say apache should be
> used to serve images/html... as far as I know, that is static content.
> So, if you have NO static content, what will apache improve?
>
> Regards
Would this allow JBossQL be used in an ad-hoc manner? So you could use
it in a very similar way to JDBC, but in the EJB domain rather than the
database domain?
A simple solution might be to have a JBossQL -> Native SQL convertor, so
you can use the standard JDBC functionality provided by your dat
Even better would be if Sun added something to the spec.
It seems to me that this is a fundamental limitation of CMP 2. No
matter how fast the container, it is difficult to see how it would ever
be able to do complex reports using CMP without such a feature.
Some reports may only return a few reco
On Tue, 2002-10-29 at 17:28, Bill Burke wrote:
> JBoss is being used in production everywhere. I've been at 6 sites myself
> over the past year. IMHO and experience, entity beans are not the right
> choice if you're doing complex reporting. A handmade query cache with
> direct JDBC will always b
I'm not entirely sure how this would be handled in JBoss, but in
principle you could send out a bunch of JMS messages to a topic or
queue.
The EJB container then launches threads to process the messages through
MDB instances. Therefore potentially, your code will be executed in
parallel.
Of cour
On Wed, 2002-10-16 at 20:30, Christopher Blunck wrote:
> With EJB, the container manages transactions, concurrency, and
> distributability so you don't have to.
Of course EJB can do much more than this however. You also have
automatically persisted objects (CMP), message based functionality (
There's a good reason why the spec doesn't recommend using threads
within beans.
Apart from the transaction problems etc. it means the EJB container
loses control of the resources. In other words, the threads aren't
pooled by the container, and the container cannot safely pop the EJB in
and out o
I think there is a very simple solution to this which is much safer.
Instead of triggering threads off in your class, move the thread body
into a MDB and then simply send several messages from your session bean.
The MDB will enable your code to run asynchronously which is probably
the effect you
Sorry to drag this thread up again from the bowels of cyberspace, but I
think the idea is very promising. If this project seems like it will be
done well, I should be able to donate some web-space to the cause.
There are a number of issues that must be sorted out first and I am sure
people will
This also works with V3.
I never managed to get Coyote working for SSL though
On Mon, 2002-09-30 at 23:29, James Cooley wrote:
> Hi David,
>
> It's pretty simple to get the Coyote connector working with JBoss 2.4.7
> and Tomcat 4.0.4. Just replace
>
> org.apache.catalina.connector.http10.
On Mon, 2002-09-30 at 14:54, Juha-P Lindfors wrote:
> Now you could argue that web services gives as better-defined or more
> standardized way of doing this. However, it is not a new approach to
> distributed computing.
That is because you are only viewing it from a technical standpoint. A
new a
On Mon, 2002-09-30 at 12:08, Juha-P Lindfors wrote:
> On 30 Sep 2002, Pete Beck wrote:
>
> > On Fri, 2002-09-27 at 19:15, Bill Burke wrote:
> > > What I've been saying all along...
> > >
> > > "People have been building Web services under diffe
I wrote a post on this some time back, but alas didn't get any
responses.
I think having an in-memory implementation of EJB-QL could provide a
huge optimisation to JBoss and would also make it possible to use
non-JDBC datastores.
As Peter notes, it wouldn't make sense for all entities, but of th
On Fri, 2002-09-27 at 19:15, Bill Burke wrote:
> What I've been saying all along...
>
> "People have been building Web services under different names for 20 or 30
> years," he explains. "We've been building distributed systems for years out
> using CORBA and RMI and all of that."
Except of cours
This sounds like a good idea except for one immediate problem:
integration.
Any components would have to be developed as a suite (including the data
model) as it is not usually possible to integrate Order A with Product B
if they are not compatible. Of course you could use the facade design
patt
Try switching to JDK 1.4.1.
According to the docs there was a "leak" in the javac compiler when
invoked from code. This is likely to manifest itself when deploying an
application.
Apparently the problem is fixed in 1.4.1.
On Thu, 2002-09-26 at 13:49, Weiqi Gao wrote:
> Hi,
>
> I'm trying to t
I think I may have discovered a nasty bug in Jetty's cookie based
session management.
I'm using version JBoss 3.0.2 and Mozilla 1.0 and here's the problem I'm
encountering.
When jetty creates a cookie for a session (named jsessionid) it doesn't
set the domain that it applies to. This results in
Are your client applications invoking the remove() method after they
have finished with their beans?
I think that there may also be memory loss if you are redeploying your
application a lot. This wouldn't be a JBoss issue, but according to the
JDK1.4.1 docs, the java compiler used to hang on to
I once read somewhere that JBoss implements CMP by translating EJB-QL
(or the older JAWS style expressions) into SQL.
This means that whenever a finder is invoked, Jboss must query the
database.
If you are in the situation where you can use commit option A, this is a
bit of a bottleneck.
The t
Wouldn't having multiple threads in a transaction break the ACID
principle anyway?
Particular the consistency and isolation parts.
Imagine the scenario:
Thread A reads object 1.
Thread B reads object 1.
Thread B updates object 1.
Thread A adds 1 to the value of object 1.
In this transaction the
On Fri, 2002-09-20 at 13:20, David Jencks wrote:
> AFAIK Oracle xa is the only db that disagrees with our code that sets tx
> isolation.
No surprises there. :-)
--
Peter Beck BEng (hons) - Managing Director, Electrostrata Ltd.
http://www.electrostrata.com --+-+-+-+-- Experts in e-business an
Dear JBoss users,
One of the challenges I have found with JBoss is migrating to new
versions.
Going from V2.4 to V3 was a lot of fun. However I have noticed that
even between minor versions, things change which can make upgrading
difficult. On a couple of occasions, I've even had to go into the
Short answer: Use the timer service in JBoss
Long answer:
Polling is not an ideal solution usually, but sometimes it is the only
solution.
A better solution is to use messaging if you can, as then your app will
only run when it needs to.
However in the draft EJB 2.1 spec, Sun have added a new ti
On that basis, wouldn't it be possible to configure two datasources, and
just use different datasources for different beans?
On Thu, 2002-09-19 at 18:26, Dain Sundstrom wrote:
> Is this even possible? I don't think it is. Isolation level is an
> attribute of the connections and not the transa
I think this ability was in the specification at some point, but it was
removed for security considerations.
If you think about it, it just wouldn't be secure to allow a given
servlet access to other http sessions. Unfortunately this is not
helpful if you have a legitimate reason for needing to
75 matches
Mail list logo
