diff --git "a/developer_jboss_org/developer_jboss_org_samples.json" "b/developer_jboss_org/developer_jboss_org_samples.json" new file mode 100644--- /dev/null +++ "b/developer_jboss_org/developer_jboss_org_samples.json" @@ -0,0 +1,602 @@ +[ + { + "url": "https://developer.jboss.org/thread/190531", + "domain": "developer.jboss.org", + "file_source": "part-00479-d9b45f53-efdc-4ae8-a3e4-b1d5d9e13869-c000", + "content": "
Hi ,
I am using JBoss Studio 2.1 to develop my new Seam project.I just start copying the steps of the link to understand the working of simple Seam project.
But When I start testing of booking example in my JBoss Studio , it gives me exception written below.
an error occurred when running TestNG tests
Path
Project location :
C:Projects
JBoss Studio :
C:jbdevstudio
JBoss AS 5.1 :
C:-Program Files jboss-5.1.0.GA
One more thing , When I start using default Seam project using JBoss Tools , it deploys successfully.But when I try to do reverse engineering using Generate Entity , POJO and DAO are not created in the default package(i.e. src/hot).It will created in another package and it will not be accessible to class of src/hot package.
Please help me out.
Please tell me about the mistakes I had done in booking example and reverse engineering process.
Thanks
Sandy
Hi all jboss guru
It is very thankful one can share with me your viewpoints on my below question.
l plan to write a cached service locator which provide a standard interface for ejb app lookup ejb home handle / datasource. This service locator can cache ejb home handle / datasource and return the cached instance back to caller. But l am concerned how does the ejb home handle and XADataSourceImpl deal with multiple concurrent calls arriving at the same moment. Are the implementation of these class thread-safe ? Are the multiple concurrent calls handled in serialized / parallel manner ?
thx & regards
fox
ejbs calls are completely thread safe per ejb spec. datasources are thread safe modulo bugs of which there were several in 2.4 before 2.4.8. (Of course, now there may be new ones).
Hey, not anyone? Can't be...
Hi,
Seam cooperates with the EJB3 container, but does not have absolute control. So when you redeploy a Seam application which contains EJB's, halt and restart to get things right.
This is not a bug but a normal restriction.
Leo
Thanks Leo,
I understand. However I ran into this issue many times in production, both after redeploys and full restarts of JBoss. It feels strange to redeploy as many times as it needs to get it working (I need to redeploy the app to fix it, not restart
JBoss as this doesn't seem to help).
This behaviour means that one cannot rely on automated start scripts. After a reboot of the prod server the application might run or it might not. This restriction has been bugging me for almost 3 years when went into production.
Tom
Hi,
JBoss is sometimes too good
for applications and tries to save everything for later restarts. I have problems with wrongly zipped wars. After redeploying the server tries to use earlier deployed stuff from the temp directory (don't ask me why..)
I normally clean up
everything after redeployment with problems; meaning deleting the TMP, DATA and WORK directories. However this only works if everything was correctly persisted in the DB.
Leo
I have seen JBoss to pickup the older version from the tmp folder even after deploying a new war. The safest way would be to remove all files under tmp folder prior before re-starting the server.
I'm working on a web app that allows our users to log in with certificates. This is working fine - the keystore/truststore, login modules, etc all work correctly.
But there are a couple of different certificate authorities in our domain, and our users will often have multiple (theoritically valid) certificates loaded in their browser. To try to ensure future compatability, we had configured our JBoss server to respect all the certificate authorities. This lead to a problem - users will often register one certificate when they create their account, and try to login with another.
Is there a mechanism in JBoss or the Java SSL stack to control which certificates the users can pick when they try to log in with a browser? For instance, could I restrict this list (in some places) to a list of certificates that matched some data in a database.
I could restrict the set of certificate authorities in our system, but I was wondering if there existed any programmatic solution?
Thanks.
I am having a problem starting JAS in debug mode, and don't know what could be causing it since I got the error using 4.2.2GA and then did a clean install of 4.2.3GA and still have the same error.
What happens is when I try to start my server in debug mode through myEclipse, It stops after
16:04:38,213 INFO [Server] Starting JBoss (MX MicroKernel)...
16:04:38,213 INFO [Server] Release ID: JBoss [Trinity] 4.2.3.GA (build: SVNTag=JBoss_4_2_3_GA date=200807181439)
16:04:38,213 INFO [Server] Home Dir: C:\\jboss-4.2.3.GA
16:04:38,213 INFO [Server] Home URL: file:/C:/jboss-4.2.3.GA/
16:04:38,213 INFO [Server] Patch URL: null
16:04:38,213 INFO [Server] Server Name: default
16:04:38,213 INFO [Server] Server Home Dir: C:\\jboss-4.2.3.GA\\server\\default
16:04:38,213 INFO [Server] Server Home URL: file:/C:/jboss-4.2.3.GA/server/default/
16:04:38,213 INFO [Server] Server Log Dir: C:\\jboss-4.2.3.GA\\server\\default\\log
16:04:38,213 INFO [Server] Server Temp Dir: C:\\jboss-4.2.3.GA\\server\\default\\tmp
16:04:38,213 INFO [Server] Root Deployment Filename: jboss-service.xml
16:04:38,464 INFO [ServerInfo] Java version: 1.6.0_03,Sun Microsystems Inc.
16:04:38,464 INFO [ServerInfo] Java VM: Java HotSpot(TM) Client VM 1.6.0_03-b05,Sun Microsystems Inc.
16:04:38,464 INFO [ServerInfo] OS-System: Windows XP 5.1,x86
as if it has hit a breakpoint, and in the development window I get the message 'Source not found for XMLEntityScanner.load(int, boolean) line: not available'
Pressing the continue execution button just brings up more of the same error. If anyone has an idea of what could be causing this, or anything else I can check, please share.
Thanks,
CDiggins
Some of the things you could try:
1) Let JBoss start successfully. Then start your debug mode in Eclipse (for this to happen, you should configure suspend=n in the JAVA_OPTS in run.bat for JBoss).
2) In the \"Breakpoints\" window of your project, if you see any breakpoints already set, then remove them.
3) In the \"Debug...\" window of Eclipse, remove/delete any existing debug configurations. Then create a new configuration.
I have a long running query that is kicked off by clicking a submit button on a page. When the server gets the request it may take 10-30 sec. to generate the request. I would like to have the user see a page that says I am working on there query and to please wait. When I am done generating the output I then send them/give them the result.
Okay now here is the kicker. If I hit the back button I want to go back to the original start page and not a waiting/processing page.
Thanks
Jim
The JBoss SOA early adopter Swedish Railroad has been blogged about here earlier. Matt Asay picks up on this deployment more recently here.
Additionally, I presented Open Source SOA Value and Strategy at Red Hat Summit and SOA World in late June. CIO e-magazine picked up on the Swedish RR example and other current status of open source SOA and future projections here. In particular, we see the convergence of EAI, SOA, BPM, EDA and CEP into a set of integration styles that may be implemented using a common platform - the 2nd generation ESB or SOA Platform. A flexible platform like this is a more cost effective and simpler (reduced to no need for separate architectural bases for EAI, SOA, EDA, BPM, etc...) way to implement an agile IT and enterprise in the future.
I seem to notice a strange behaviour,
As I submit forms to JBoss, I seem to receive them correctly within the servlet.
though as I post a form using following action
servlet/srv?selector=variable?second=blabla
I was attempting to use use the inline arguments to qualify what to do with the form....
the arguments included in the form (submit) are intermixed with the
query section in the http line selector and second are considered as part of the form, eventualy overriding the form content*chose of variables)
THe implementation intermixes Urlencoded variables as the post information. Of caorse this seems to simple to implement wherever the parameter comes from it is used as a parameter, but it allows someone changing or tricking the resultset without leaving some clarity...
thanks for your thoughts
Hi,
Our application does a lot of HTTP communication to other systems. We have found that the IP addresses of the hosts we are communicating with are being cached by JBoss. When the IP is changed, we get errors until JBoss is restarted.
Is there anyway to turn this off? I would rather that we query DNS everytime than have to restart our server in the middle of the day when we hear about a problem.
Thanx.
Norton
It's the JVM that caches DNS entries forever by default - which is an annoying thing for it to do.
You can change the settings in two ways:
1) Set \"sun.net.inetaddr.ttl\" system property
e.g. -Dsun.net.inetaddr.ttl=30 in JAVA_OPTS when starting JBoss to limit the caching to 30 seconds
2) Set \"networkaddress.cache.ttl\" security property (I assume in a policy file) - this does the same thing and isn't a sun specific property so it isn't liable to change.
There's some documentation on this here: http://java.sun.com/j2se/1.4.2/docs/guide/net/properties.html
Hi
I am developing JSF application.create login page using jaas(login.js) and configure login-config.xml ,web.xml
after that deployed jboss server .and run the server ,this login to my application and login page run sucessfully(one day only),i wil login to that application next day cant login that page ,it go to error.xhtml page,but jboss server til running.
show the following url only
http://localhost:8080/xxx/faces/general/j_security_check
Any idea pls help me...
By
Chandrasekaran
replay me..
What kind of error? Do you see any exception stacktrace in the server.log?
Hi
Not comming error in console,but cant login to login page . give username and password to click sumbit button go to error.xhtml page and show url path is below but i give correct username and passowrd..
http://localhost:8080/xxx/faces/general/j_security_check
@i wil start jboss server and my application working praperly in that day only,but i wil login next day cant login page..it go to error page(I am isong Jboss6.0 and jsf2.0)
How to solve this issue
Chandrasekaran.k
Can't really guess based on what you have mentioned so far. You'll have to take a look at the logs (server.log) and see what's wrong or maybe debug the application.
Hi,
When I save my changes in JBPM console there are changes in bare git repository. I can then push that changes into another repository, but autor of commits is always: salaboy <>
(<> means there is no email).
I need to change it. I could do it with some git commands (rebase and commit --amend) but I need to do it for every commit.
Is it possible to change email for salaboy that jBPM would use that email in commits to bare repository of my project ?
Regards
Michał
Is this what you are looking for?
\ngit config --global user.email \"Your_Email_Address@example.com\"\n
I have tired this. I have tried also --local.
Both doesn't work.
jBMP have some other configuration of a user. jBPM commits all changes with an author set to a user that I am logged in to the web jBPM-console.
I work in jBPM-console as a salaboy user and this user is an author of all commits, that happen when I save the changes in jBPM-console.
I don't know where I can change email of salaboy (I'm not sure if it is git or jBPM or JBoss configuration).
In other projects (non-jBPM projects) where I use git, I have configured my name and my email and all commits contains my name and my email in author section.
In general - we need to work out some kind of process for distribution of our bussines processes trough dev/test/prod envrionments. The way to handle deploying and sharing processes between different instances of workbenches. Currently, for example, we don't find eclipse plugin usefull enough to use it. Probably we miss something, but for us it looks like there is a lack of lot of features. So we use modeller included in kie-workbench application. Let's say we have development instance of workbench. W create processes an test it using this instance. Now we need to put it into one of test instances. Should we pull/push/clone git repositories or maybe rather download artifacts and deploy them? The same later in UAT phase and in production deployment. What is the best model to handle distribution of implemented processes? What are the best practicies?
Greets,
Patryk
you can follow this practice:
This will allow you do build projects in single environment and then deploy same artifacts into multiple environments.
In 6.2 there will be option to deploy from build project to given environment via REST but anyway above procedure needs to be followed as well so artifacts can be found when not build on given environment.
I would recommend this approach instead of moving sources through all environments which I believe you won't allow changes any way....
HTH
If I'm creating a CMP many-to-one relationship for an entity bean and that link is optional, if the link is not active must I set a null value into the field representing the many end of the relationhip?
For example, I have an Address entity bean that includes an integer 'RegionID' field. This field value is optional. I want to create a relationship field that allows the Region entity to be accessed from the Address entity. As the address to region many-to-one link is optional must I set the default value of RegionID to be null?
The JBoss default root page probably might need a change to be more dynamic, based on the server profile that's being run. With JBoss-5 GA, there's a \"web\" server under %JBOSS_HOME%/server/web. This server profile does not seem to be deploying the \"web-console\" application. When the root home page is accessed at http://localhost:8080, it has a link to \"JBoss Web Console\" which when clicked leads to http://localhost:8080/web-console and obviously fails with 404 \"The requested resource (/web-console/) is not available.\" error.
Thinking more about this, should the root page be having links to applications deployed within the server? What if the web-console or jmx-console are intentionally not deployed by the user? The root page will still continue to show those links.
How do I obtain a handle for an EJB in EJB 3? And once I have the handle, how do I get the EJB again? I can't just call up an EJBObject's getHandle, right?
I couldn't find anything related to that in the latest early draft. Does anyone know how this will be addressed?
Sorry for the lame title.
I'm hoping someone can work a miracle.
I have 2 domains (one with internet access and one without). I need them both to trust my verisign certificate. The one with internet access is working. The one without is throwing the 'Revocation list is not available' error message.
I can ignore this and access my web page, but I need the trust to work for Adobe LiveCycle.
Versign tells me there is something I can do in JBoss, but either doesn't know how or won't tell me.
Thanks in advance for any help!
God bless,
BroMatt
Can you be more precise the about the enviroment?
If you have two domain, one with internet access and one without, you have to generate two certificates.
But be careful that you have to create two VirtualHost, with different IP, so you can use two domain in mode SSL.
Thanks for your reply!
I'm running Adobe LiveCycle ES 8.2 with JBoss 4.2.0. I have a versign certificate installed on the server and am having no trouble accessing the admin page.
I can ignore the warning and open the page anyway, but Adobe needs to do some things in the background for document security. This is where the trust has to be in place. If there is no trust, the document fails to check the security settings and I get a message that tells me to connect to the network before opening this file.
I hope this helps.
Thanks and God bless,
Matt
I found a work around and am posting for anyone else having this problem.
Internal clients (no internet) fix for Windows devices.
In IE 6.0,
Tools --> Internet Options --> Advanced
Uncheck \"Check for publisher's certificate revocation\"
Uncheck \"Check for server certificate revocation (requires restart)\"
This did the trick for me.
God bless,
Matt
The purpose of this page is to detail how JBoss Cache is used in JBoss AS.
Integration code is in svn here in the org.jboss.ejb3.cache.tree package.
Testsuite is here via the clustered-tests target. The actual test code is in the org.jboss.ejb3.test.clusteredsession package.
A new implementation is in the works which will better define the SPI for the jbc integration.
Session-type use case. Sessions are only meant to be accessed on one node at a time, with that node remaining sticky. Replicating data via JBC is solely to provide HA in case of loss/shutdown of the owning node or some failure in session stickiness. Session stickiness makes this a good candidate for buddy replication.
/sfsb
+++ /uniquenameofdeploymentandbean
++++++ /bucket [0-99]
+++++++++ /sessionid
Session is stored as a single K/V pair, key is string \"bean\", value is the StatefulBeanContext
TODO: convert the key to an enum.
The region for region-based-marshalling and partial state transfer is /sfsb/uniquenameofdeploymentandbean.
The \"bucket [0-99] Fqn level above is a workaround to a FileCacheLoader issue wherein some filesystems limit the max number of child directories under /sfsb/uniquenameofdeploymentandbean to ~32K. The addition of 100 bucket nodes in the tree allows passivation of up to 3.2 million beans.
See the sfsb-cache config in the standard JBoss AS cache manager config file.
Integration Code is in svn here
Testsuite is contained in the same project as the main code.
Extensive documentation is available here
For now, I (Brian) am not going to comment much on this as the docs are good and the code is clear; learn from those while I focus on the others. :-) I'll come back to this later.)
This is on a JBoss AS 5.x wishlist for JBoss Cache 3.x
A Testing JBoss Cache 3 in AS 5 Trunk was created detailing how this can be done, specific to JBoss AS 5.
Are you sure you want to delete this article?
\n\nWe are thinking of taking the plunge (clustering our servers).
Should we wait until the 4.0 version is available. We are currently on
3.2.3 ? It was painful to upgrade from 2.4.4 to 3.0 and basically I want to do it right once.
I'm running 3.2.5 with clustering, and it was relatively painless to move from 3.2.3.
--James
Hi,
I am try to find the way to collect the Heap dump when ever I need. I have tried -xx:+HeapDumpOnOutOfMemoryError, it is creating dump during startup and application is not starting properly. We are using Jboss 4.3 and jre1.6.0_10.
Please help me to get raid of this.
Thanks
Could also attached the Heap Dump. There are other Java memory areas that will cause a OutOfMemory.
Thanks,
Hi
I am using Jboss-2.4.3. I have a session bean which does a lot of processing sometimes taking more than the transaction timeout period of 300 ms. The problem is that whenever a transaction timeout occurs, no notification is given to the client via an exception or something. The print indicating the timeout appears on the console and no other activity takes place leading to the client to hang for ever.
Is there a solution for this? In WEBLOGIC an exception is thrown whenever a time out occurs so that the client is notified of the problem. The client can then take appropriate steps like firing the method again or something.
The print about the time out is shown below :
Transaction XidImpl [FormatId=257, GlobalId=bom2//4, BranchQual=] timed out. status=STATUS_ACTIVE
Can somebody help?
Thanx
Jikky
Please forgive me if this was discussed at length before (and point me to these discussions).
It looks like event latest greatest of Wildfly (10.0.0 CR4) still using 1.x version of log4j. Am I correct? Is there are any plans to move to 2.x any time soon? If not, I am curious what are the reasons why Wildfly stick with older version.
You comments would be highly appreciated.
Cheers,
Andrei
You should be able to use your own version of log4j as part of your application. Please see: How To - WildFly 10 - Project Documentation Editor
If you need/want to include your version of log4j then you need to do the following two steps.
This only works for logging in your deployment. Server logs will continue to use the logging subsystem configuration.
Jay thanks for quick answer, but I am afraid this would not solve the problem completely. Correct me if I am wrong, but this would not allow me to control logging via Wildfly Web Management interface. Am I wrong?
Hi everybody,
I want to deploy a folder containing some files that have to be accessable via http.
Under native tomcat I just put the folder in my deploy-Folder and it works.
In JBoss it doesn't.
Can anyone provide a solution out there?
Greetz
Piecko
You could place them in ROOT.war. Or you could try this:
http://www.jboss.org/community/wiki/ExternalDirectories
For external directories on JBoss AS 5, see:
http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4212358#4212358
We are using MBean as startup class to start some configuration property/xml files. We want read the configuration when jboss(like weblogic) starts and put all the config information into memory so that application can get config data from memory. I created an ear, which has sar and jar(ejb) in it. The jar has some utility class which mbean class is using. But the problem is Jboss while loading ear file. It is loading sar first. Is there any way to load jar first and then sar. sothat sar can access utility class from jar.
Thanks in Advance
Sri
I was not able to find a graceful solution. So I used jboss schedule to start that start up class twice. The first one will fail. However, the second will succeed and it will work just fine. I hope someone find a better solution.
Brian
Hi,
I'm quite new to EJB in general and I have tried to follow the tutorial
in http://www.tusc.com.au/tutorial/html/. But I couldn't get
lomboz work, so I had to use JBoss-IDE. Not that there is anything
wrong with it (Is there?).
So I have been trying to apply it in JBoss-IDE.
It doesn't work the same way as lomboz, so I have had some difficulties
do the examples. The chapter one was quite easy, but now I'm working
with chapter 2. It uses DAO to access the database. I don't really
understant how to implement that in JBoss-IDE. In the turorial it uses
the ejbGenerate.xml to create the class or whatever. So how should
I create the interface/class?
So my main question is: how create a DAO?
Shmuel.
Hello,
Sometimes, when using jBoss more exhaustively (like querying the DB) it \"blows up\", with the error: \"java.lang.OutOfMemoryError\". Is this a common problem? Do I need to enlarge the heap space to overcome this? How do I do that?
I have JBoss-2.2.2, with tomcat-3.2.2, Java 1.3.1_01 and a mySQL DB.
Thanks!
Andre
You probably need to increase the heap size given to your jvm. Here is a snippet of my run.sh:
# increase heap size to 512 meg
MEMORY=-Xmx512m
echo JBOSS_CLASSPATH=$JBOSS_CLASSPATH
java $HOTSPOT $JAXP $DEBUG $MEMORY -classpath $JBOSS_CLASSPATH org.jboss.Main $@
This is a unix or cygwin configuration, the modifications to run.bat for windows are very similar.
Nick.
The memory exception is most likely from a jdk bug, that is if you are using Sun's jdk 1.3.1_01. http://developer.java.sun.com/developer/bugParade/bugs/4504200.html
I had the same out of memory exception problem, after certain loads it was very persistent. (jboss 2.4.8, tomcat 3.2.3).
Upgrading to jdk 1.3.1_04 fixed it.
Pari Shanmugam
I have a package of classes that all share a common base class Model. The Model implements Serializable and houses a few common values important values:
public class Model implements Serializable
{
long id; // primary key
Timestamp version; // optimistic locking aid
boolean deleted; // \"soft\" delete
....
}
The property access methods are in the derived models because they have annotations @Id, @Version and @Column for each of these. In a real POJO those accessors would be in the base Model class.
How can I persist base class properties from within the subclass?
Not tried this myself, but I'm assuming you've seen this?:
http://docs.jboss.org/ejb3/app-server/tutorial/singleinheritance/single.html
Hi,
I'm working on https://jira.jboss.org/jira/browse/JBIDE-4859 which is about provding a better update site layout than what we have today.
I didn't like our current updatesite because it doesn't show very well how big a feature set we actually we have and many might not realize they don't need to install *everything*.
On the other hand I would like there to be an \"easy\" option to install everything that should be able to work on Eclipse JEE bundle.
So I took a look at Eclipse Galileo updatesite and saw they got some pretty good names for most things and decided to use where it made sense and then also keep the JBoss Tools category (now renamed to \"All JBoss Tools\") which is everything except plugins that drags in big dependencies beyond Eclipse JEE.
I would love to hear if this updatesite categorization works for you and if not, what you would prefer instead ?
btw. the first category is not expanded in the screenshot since it would then not fit in one screen.
And please note that pi4soa and jbpm plugins are not shown here purely because of me being lazy - they are supposed to be shown inside both the All JBoss Tools and SOA Tools category.
max.andersen@jboss.com wrote:
On the other hand I would like there to be an \"easy\" option to install everything that should be able to work on Eclipse JEE bundle.
I would propose to provide big feature to install everything like it is done for Eclipse for JEE Developers. The benefits are:
1. One checkbox to install everything
2. One item to remove everything if you need reinstallation.
Here is an example for Eclipse for JEE DEvelopers
Yes, I wanted to add that since beginning of time but there were some build issues around this afaik. If not the case anymore then yes, lets do it.
Having some better aggregated feature sets would work for me - as long as the updatesite can still provide individual feature/plugin downloads.
Hi,
where to put the files my jsf application consist of in the jboss filesystem.
i can run jsp application without problems, do i have to take care of something else with jsf?
I am a beginner, so everything might help.
Thanks and regards
just put all of JSF related .jar file to the web-inf\\lib folder, then you can use JSF in your application
Thanks for your help.
Now it seems to recognize JSF, but it still encounters an error:
org.apache.jasper.JasperException: The absolute uri: http://java.sun.com/jsf/html cannot be resolved in either web.xml or the jar files deployed with this application
i have a folder with following content (Folders with capital letters)
JSFFOLDER
-WEB-INF
--faces-config.xml
--web.xml
--LIB (contains all jars from the reference impl)
-classpath
-project
-index.jsp
I see that in 4.0.1, xalan.jar was moved to /jboss-4.0.1/lib/endorsed. I don't see the connection, but I'm now seeing this error:
16:40:34,684 INFO [Init] Unable to patch xalan function table.
java.lang.NoSuchFieldException: m_functions
at java.lang.Class.getField(Class.java:919)
at org.apache.xml.security.Init.registerHereFunction(Unknown Source)
at org.apache.xml.security.Init.init(Unknown Source)
at org.jboss.jaxr.servlet.SAAJServlet.init(SAAJServlet.java:86)
etc.
Any suggestions?
Using what jdk and how is jboss being started?
I am getting the same using jdk j2sdk1.4.2_6. I am starting JBoss using the JBoss IDE plugin for eclipse. Any ideas?
Thanks,
Neil
You have to use the xerces parser found in the jboss-4.0.1/lib/endorsed directory so you need a -Djava.endorsed.dirs=\"$JBOSS_ENDORSED_DIRS\" option as is done with the run.sh/run.bat scripts.
Alternatively if your not using jaxr you can just remove the all/deploy/ebxmlrr-service.sar directory.
Thanks for the quick response!
how can i configure my jboss 3.2.1/jetty
if somebody type s
http://x.x.x.x/index.html
or
http://x.x.x.x
it becomes redirect to
http://x.x.x.x/myproject/main
http://x.x.x.x -> ip adress of running jboss server
In what xml must i do changes?
Greets Johan
You could create a root context with a page that redirects to your designated area.
You could also have a front-end Apache with URL rewriting to achieve the same result.
However, normally I would guess that you would move your whole application to the root context. See the FAQ for JBoss 3.2.0 changes for information on mapping web applications to the root context.
Problem solved,
I have maked a small war with one page index.html that's run as root context and does a redirect
anyway thanks for the tips
Johan
Hello,
what are the products required to build ESB develop , deploy & implementation?
Are JBoss Fuse and JBoss AS enough to build , develop Enterprise Service Bus?
Thanks,
Krishna
Yes you just need JBoss Fuse.
Thanks for response, fuse projects developed in JBoss Developer ?? or do we have any other Fuse IDE as couple of guys in forums mentioned/used Fuse IDE word
Jboss Developer Studio is the Eclipse tooling from Red Hat / JBoss.
It includes what was formerly known as Fuse IDE - eg the graphical Camel editor.
So from the Developer Studio you have all the tooling in one product.
Thanks , got it.
daySlotForResourceKind is null inside the call getDateNotPassedAlready which afaik is not possible.
Whats going in inside rich:columGroup?
I have had to add a null check in the code to handle this.
\n<rich:dataTable var=\"daySlotForResourceKind\" value=\"#{dayForResourceKind.daySlots}\">\n <f:facet name=\"header\">\n <rich:columnGroup>\n <rich:column>\n <h:outputText id=\"allocation_slot\" value=\"#{messages.allocation_slot}\" />\n </rich:column>\n <rich:column>\n <h:outputText id=\"allocation_status\" value=\"#{messages.allocation_status}\" />\n </rich:column>\n </rich:columnGroup>\n </f:facet>\n\n <rich:columnGroup rendered=\"#{resourceAllocationController.getDateNotPassedAlready(dayForResourceKind, daySlotForResourceKind)}\">\n <rich:column breakBefore=\"true\">\n <h:outputText value=\"#{daySlotForResourceKind.slot.label}\"/>\n </rich:column>\n <rich:column>\n <h:outputText value=\"#{daySlotForResourceKind.title}\"/>\n </rich:column>\n </rich:columnGroup>\n</rich:dataTable>\nI have been looking into this with the debugger and it seems that
getDateNotPassedAlready()
is called before
getDaySlots()
This makes me think that the rich:columnGroup is being elaborated at least ONCE (three times in my case) BEFORE rich:dataTable.
I know you get this strange kind of stuff happening in JSF (it really makes JSF unresponsive as often routines are called multiple times).
I should actually learn more about JSF so I can no process some things when its not the right phase!
Anyhow, I hope this helps you analyse what is going on with rich:columnGroup (and possibly update the docs if its a problem that you cant get round).
I have looked at this again and still cant see why it is happening.
I can confirm it is happening however.
That's because dataTable processes (decodes, validates, renders, etc.) column headers prior to rows that contain data.
Thanks for the reply.
Perhaps this is worth a comment in the doco for others like me who are left wondering about all the NPE's
:)
Hello,
My EAR file contains 2 .ear and 1 .war files.
in the .war, there is a .jar in the WEB-INF/lib, that includes some customer tag library and some utility classes, etc.
In one of the Tag class, I instance the org.apache.struts.util.MessageResources. But JBoss just raised an ClassNotFoundException of that MessageResources.
In fact, I have put the struts.jar in the WEB-INF/lib (in the .war)
Solution:
I put those 3rd-party JAR files under the EAR and add
the following into the application.xml
struts.jar
.... (all othre 3rd-parties jar)
Strange, it works.
Is there any problem in the JBoss class loading ????
I am using JBoss3.2.1 + JDK1.4.2 in Win98
Eric
Sorry, it should be contains 2 .jar and 1 .war.
Hi Bill Burke,
I will like to say you have done a good work on the AOP Framework with all you the JBOSS developers. I have started documenting the code using the current version on cvs tree and I will only be add comments
I was primary doing this for me, to understand the code. I believe this might benefit others as well. If this is okay with you let me know where to email the documentated source, one class at a time. This is the only way I can contribute for now, and I have quite a lot of time to spare.
Regards
Adeyemi E Adegbile
Hi!
I am trying to use URLDirectoryScanner by modifying jboss-service.xml
According to
https://sourceforge.net/tracker/?group_id=22866&atid=376685&func=detail&aid=711341
I cannot use URLDirectoryScanner only.
(I tried, and it really didnot work)
But, when I tried to use both URLDirectoryScanner and URLDeploymentScanner, it throws some error messages as
\" org.jboss.deployment.DeploymentException: Trying to install an alredy registered mbean: jboss.deployment:type=DeploymentScanner,flavor=URL\"
Could you tell me how to use URLDirectoryScanner?
Thank you!
Quite old post, but ...
Have you tried to give the 2nd deployer a unique name? E.g.
name=\"jboss.deployment:type=DeploymentScanner1,flavor=URL\">
Hi,
I am using bundled version of Jboss Portal 2.7.2. From the Viewing the document(pdf or doc, not html. Html works fine.) that has READ permission assigned in CMSAdmin Portlet causes \"Access to this resource is denied\". According to the Jira (JBPORTAL-1378), it should have been fixed already. However, I still have that problem and it is a show stopper for us. Does anyone know what is causing this problem? Thank you advance.
Brian Ko
Hello,
I was unable to reproduce the issue, I've used 2.7.2 budled version, but permissions work fine with both html and pdf. Can you please post the exact steps on how to reproduce the error?
Thanks,
Viliam
Have you tried with non CMS super user? It works fine with CMS super user. Also it works with anonymous ot user permission. It does not work only with role permission. I already found the solution and I will post it.
I found that user is saved in session, but not the roles. Therefore, I changed the viewfile.jsp and pending_items.jsp so that roles are saved in the session.
I also changed the CMSPreviewServlet so that it retrieves the roles from the session and put it in JCRCMS object. Now it is working fine. The changed codes are as follows.
Following are added to the jsp's.
Set roles = new HashSet();
// Get the current authenticated subject through the JACC contract
Subject subject = (Subject)PolicyContext.getContext(\"javax.security.auth.Subject.container\");
if (subject != null)
{
Set tmp = subject.getPrincipals(JACCPortalPrincipal.class);
JACCPortalPrincipal pp = null;
for (Iterator k = tmp.iterator(); k.hasNext();)
{
pp = (JACCPortalPrincipal) k.next();
if (pp != null)
{
break;
}
}
if (pp == null)
{
pp = new JACCPortalPrincipal(subject);
tmp.add(pp);
// Lazy create all the permission containers for the given role names
for (Iterator k = pp.getRoles().iterator(); k.hasNext();)
{
Principal role = (Principal) k.next();
roles.add(role.getName());
}
}
}
request.getSession().setAttribute(\"remoteRoles\", roles);
Following were added to CMSPreviewServlet
Set roles = (Set)request.getSession().getAttribute(\"remoteRoles\");
JCRCMS.setRoles(roles);
Hi, there,
I am trying to monitor several SVN projects. Some svn repostiories' \"svn status data\" could be generated, but some failed.
It is reported \"time out\". (i already changed the \"TransactionTimeout\" to 3000 in the \"jboss-service.xml\" file)
Did you happen to meet the same problems? Would you please give me some suggestions? Thanks in advance.
Below is the error log:
20:19:34,898 INFO [SvnMonitoringPortlet] Viewing...
20:19:35,054 INFO [STDOUT] 20:19:35,053 INFO [MethodResultCacheInterceptor] Cache-miss: reloading \"hu.midori.kosmos.server.svn.SvnServiceImpl.getRepositories.http://user:passwd@sh-svn.sh.xxx.com/ssg_repos/svn_WML/WML/\"...
20:19:35,140 INFO [STDOUT] 20:19:35,140 INFO [SvnServiceImpl] Analyzing log of \"http://user:passwd@sh-svn.sh.xxx.com/ssg_repos/svn_WML/WML\" repository...
20:19:55,506 INFO [STDOUT] 20:19:55,506 INFO [SvnServiceImpl] 485 log-entries received.
20:19:55,507 INFO [STDOUT] 20:19:55,507 ERROR [SvnServiceImpl] Skipping invalid entry (no author)
20:19:56,253 INFO [STDOUT] 20:19:56,253 INFO [SvnServiceImpl] Analyzing content of \"http://user:passwd@sh-svn.sh.xxx.com/ssg_repos/svn_WML/WML\" repository...
21:09:34,934 WARN [TransactionImpl] Transaction TransactionImpl:XidImpl[FormatId=257, GlobalId=silc2003/137, BranchQual=, localId=137] timed out. status=STATUS_ACTIVE
You should increase the timeout period more.
Plz note that even if the request transaction is cancelled by the appserver, it just kills only the HTTP request-response cycle. The background thread that collects the info will not stop, so as soon as it completes the job, you can just do a \"Reload' in your browser and voila!
I am a new user of JBoss2.4.4.I am trying to use Application
authenticated datasource in my Application.
The entry in my ejb-jar.xml file is like this
<resource-ref>
NID database
<res-ref-name>jdbc/NidDB</res-ref-name>
<res-type>javax.sql.DataSource</res-type>
<res-auth>Application</res-auth>
</resource-ref>
the entry in my jboss.xml file is like this
<?xml version=\"1.0\" encoding=\"UTF-8\"?>
<!-- All bean containers use this security manager by default -->
<security-domain>java:/jaas/NID</security-domain>
<enterprise-beans>
<ejb-name>CarrierEntityBean</ejb-name>
<jndi-name>orillion/carrier</jndi-name>
<resource-ref>
<res-ref-name>jdbc/NidDB</res-ref-name>
<resource-name>NID</resource-name>
</resource-ref>
</enterprise-beans>
<resource-managers>
<resource-manager res-class=\"org.jboss.jdbc.XADataSourceLoader\">
<res-name>NID</res-name>
<res-jndi-name>java:/NID</res-jndi-name>
</resource-manager>
</resource-managers>
The corresponding entry for NID Pool in jboss.jcml file is
NID
org.jboss.pool.jdbc.xa.wrapper.XADataSourceImpl
jdbc:oracle:thin:@192.168.0.185:1521:test3
1200000
10
false
false
false
true
120000
1800000
false
false
1.0
0
As i am using the Application authenticated resource in my application i
did not give the JDBCUser and Password attribute while configuring the
NID connection Pool.
But when I am trying to start the Jboss Server ,the Server hangs up
while trying to create the connection pool.
Can you please help me out and say how do i configure and use
Application authenticated datasource in JBoss Application Server.
Thanks
Asit
I'm not sure how well this is supported on 2.4.x. You need to supply a user and password so the XADataSourceLoader can check that it can get a connection when it starts (or you can comment out the line in which it does this).
After that, you can get connections using ds.getConnection(user, pw);. What I don't know is how well connections with different users are distinguished by the pool. There are settings in the pool that tell it something about what to pay attention to, but I'm not sure where they are set or to what.
Hi,,
I am Using Rich :data table and rich : subtable to display my data
In my rich :subtable i am performing an action and to see the updated values in the columns i need to rerender my entire data table
Is there any way so that i can rerender only the particular row of the subtable where an action is performed
Thanks
Mahi
rich:dataTable and rich:subTable have ajaxKeys attribute:
| ajaxKeys | false | false | javax.el.ValueExpression (must evaluate to java.util.Set) | This attribute defines row keys that are updated after an AJAX request |
It should do the trick.
Take a look at this:
http://java.dzone.com/articles/an-introduction-to-jboss-richf?page=0,5
Could u elaborate more on this
I am posting my scenario here ..
I have a rich data table
In that rich data table each row is a sub table
I want to re-render a particular row and a column in the data table based on the action performed in the column of the sub table
Thanks
Mahi
Please, look at this link:
http://ftp1.aua.am/CIS/AVSL/Literature/JSF/eBooks/JbossRichFaces/subTable.html
I think it describes a situation very close to yours.
Is it possible to automatically have a process instance started once it's deployed (aka without human interaction)? I would like to run a process once a day and have it follow a defined workflow. Normally, a cron job or the like would be sufficient, but I want to utilize jBPM functionalities.
Thanks,
Julian
I think it is impossible that the process you have deployed can create a process instance automatically.Maybe the process should receive a message,just like the relation between father process and process-state.It's just my opinion.Hope to debate with you about jBPM.
If you create your own deployer (not to difficult, see the sourcecode for how e.g. the ant task does it) you can combine this with an automated start of a process instance.
Many other ways are possible e.g. a startup servlet that checks the latest versions every x seconds and start a process instance if a new version is encountered.
There is just no functionality for this created by the jBPM team.
ok thanks. I will likely use the quartz scheduler to invoke the definition. The main reason for wanting this functionality is to allow a nightly process to inform users of certain tasks after completion of batch processes.
Hello,
I have got the following ECSS file, which I have successfully included into my page:
{code}@charset \"UTF-8\";
.rf-dt-shdr-c
{
color: red;
background-image: none;
background-color: #EAF0F8;
}{code}
The above shows my page with the red text color and blue table sub header cell background colors, however datatable these colors are skin-dependent. When trying any of
{code}
@charset \"UTF-8\";
.rf-dt-shdr-c
{
color: '#{richSkin.generalTextColor}';
background-image: none;
background-color: '#{richSkin.generalBackgroundColor}';
}
{code}
nothing happens, that is the text color is white and the background color seems to be transparent.
Is there anything you have to configure when trying to reference the inbuilt richSkin bean via EL? Am I using it incorrectly? What's wrong here?
Karsten
The answer is here: http://community.jboss.org/thread/163588
You have to use JSF 2 h:outputStylesheet for the EL replacement to work.
Karsten
Hello,
I am trying to migrate from Jboss4.0.2 to 4.0.5GA but am having jndi problems.
My application is deployed with three files in the $jboss/server/default/deploy:
- application.ear
- application.war
- application-ds.xml
The war contains all the client jars it needs to access the ear file which has its datasource defined in the *-ds.xml file.
When I run a TestCase which tries to access the ear services, I have no problems. i.e:
Object objRef = new InitialContext().lookup(LocalizationSessionFacadeHome.JNDI_NAME);\n LocalizationSessionFacadeHome sessionHome = (LocalizationSessionFacadeHome) PortableRemoteObject.narrow(objRef, LocalizationSessionFacadeHome.class);\n\n return sessionHome.create();\n
16:05:51,406 ERROR [STDERR] Caused by: java.lang.ClassCastException
16:05:51,406 ERROR [STDERR] at com.sun.corba.se.impl.javax.rmi.PortableRemoteObject.narrow(PortableRemoteObject.java
:229)
16:05:51,406 ERROR [STDERR] at javax.rmi.PortableRemoteObject.narrow(PortableRemoteObject.java:137)
16:05:51,406 ERROR [STDERR] at org.mdarad.global.ServiceLocatorBase.getLocalizationFacade(ServiceLocatorBase.java:33
77)
16:05:51,406 ERROR [STDERR] ... 27 more
16:05:51,406 ERROR [STDERR] Caused by: java.lang.ClassCastException: $Proxy566
16:05:51,406 ERROR [STDERR] at com.sun.corba.se.impl.javax.rmi.PortableRemoteObject.narrow(PortableRemoteObject.java
:212)
16:05:51,406 ERROR [STDERR] ... 29 more
Hi,
I have the same problem with JBoss 4.0.5 (ClassCastException when trying to narrow an EJB interface retrieved via a JNDI lookup).
As with you, my WAR and EJB are colocated in a single Jboss instance.
I figured setting \"Call by Value\" semantics everywhere would fix this (assuming the usual JBoss flat-classloader issues) - however that did not resolve the problem in 4.0.5 !?!?
Have you made any further progress ??
Hi,
Yeah I actually got it to work. I modified some configurations. You can see the details here:
http://mail-archive.com/mdarad-toolbox-devs%40lists.sourceforge.net/msg00127.html
Hope this helps,
Francois
Thanks Francois,
It seems you just mod'd the CallByValue semantics per your instructions:
- default/deploy/ear-deployer.xml\n - Isolated => true\n - CallByValue => true\n- default/conf/jboss-service.xml\n - NamingService: CallByValue => true\n- default/conf/jboss-minimal.xml\n - NamingService: CallByValue => true
Hmmm...sorry David. That's where my knowledge stops. I haven't yet used EJB3 and don't know anything about jems.
hi i used the installer as well to take advantage of ejb3 and im having the same ClassCastException when narrowing. Have you been able to solve this?
Thanx
i found a way around it, and it was a change to the way I initizlied my jndi context.
before i was doing:
Properties p = new Properties();
p.put(Context.INITIAL_CONTEXT_FACTORY, \"org.jnp.interfaces.NamingContextFactory\");
p.put(Context.URL_PKG_PREFIXES, \"org.jboss.naming:org.jnp.interfaces\");
p.put(Context.PROVIDER_URL, \"jnp://localhost:1099\");
return new InitialContext(p);
now i just simply returned new InitialContext() without the properties.
Hi,
I am trying to test some JMX client code from Arquillian.
I want to deploy a web application as a WAR and have the app call into remote JMX servers.
I have verified that the code implementing the JMX client is working fine from a standalone unit test.
When I deploy as a WAR archive in an Arquillian test, I cannot figure out what JARs and resource files need to be packaged to get the test to execute.
I am currently stuck on this exception;
Caused by: java.lang.ClassCastException: com.sun.corba.se.impl.orb.ORBImpl cannot be cast to org.omg.CORBA.ORB
at org.omg.CORBA.ORB.create_impl(ORB.java:295) [jacorb-2.3.1.jbossorg-1.jar:]
... 102 more
The CORBA dependency arises because I am using IIOP as the remoting protocol for JMX.
Any suggestions gratefully received...
Thanks,
I have an application based on springs and hibernate, which is working fine with Mysql database on
jboss 4.0.3. I am trying to change the database from mysql to postgres. The database porting has been successful and i have generted the hibernate files. Jboss configuration has also been modified w.r.t the newdatabase. When i start the jboss, i get the following error:
11:25:31,115 WARN [JDBCExceptionReporter] SQL Error: 0, SQLState: 42703
11:25:31,115 ERROR [JDBCExceptionReporter] ERROR: column usersadapt0_.user_id do
es not exist
11:25:31,130 INFO [STDOUT] org.springframework.orm.hibernate3.HibernateJdbcExce
ption: JDBC exception on Hibernate data access; nested exception is org.hibernat
e.exception.SQLGrammarException: could not execute query
11:25:31,130 INFO [STDOUT] Caused by:
11:25:31,130 INFO [STDOUT] org.hibernate.exception.SQLGrammarException: could n
ot execute query
at org.hibernate.exception.SQLStateConverter.convert(SQLStateConverter.j
ava:65)
the above column exists in the database..but it is not able to recognize.
I have - postgres 8.3
Postgres driver PostgreSQL 7.4 JDBC3 with SSL (build 210)
The 8.3 driver is giving some error.
Ok, so now that I have successfully implemented a plain JBossCache within JBoss 3.2.6, I have some basic, performance related questions.
1. In the plain JBossCache, it looks like I need to do a put everytime after an object is changed. Is that right? For example, if I have a vector assigned as value to a key, then in order to add an element in the vector, I need to do a put(fqn, key, myVector) after every change. Am I correct?
2. If #1 is correct, then does the replication take place on the entire object every time, or does is somehow only replicate the incrementals?
3. I am mostly using JBossCache to store the state information of my class. Right now, the class is a POJO. What is better performance wise - convert the POJO into a stateful session bean and use SFSN in state memory replication, or use TreeCache?
4. Are there any performance timings/statistics available for put and get using plain TreeCache?
5. What is the performance impact of doing a get on the TreeCache vs doing a get on a local hashmap? Is it worthwhile to invalidate the cache and only get values from cache is something has changed?
I guess thats a lot of questions. All answers are deeply appreciated.
To answer some of your questions:
1. Yes. You will need to specifically do a put whenever your object has been modified.
2. Replication takes place for the whole myVector in your case.
(This is where TreeCacheAop comes where you can do fine-grain replication.)
3. SFSN will manage the state change for you vs. you will need to manage yourself (unless again you are using aop for fine grain replication).
4. We have some old numbers on 2 nodes replication. But it's outdated so we will be working with new ones now.
-Ben
Thank you for the quick response Ben. That helps. My main concerns are in the area of performance.
From what I read, if I use the cache loader, then every single node is persisted on a backend store. Correct? Will that not be a big dent on performance?
What if I dont have/want a cache loader? Will the nodes still get evicted? If so, am I losing those values? What happens if I dont have a cache loader, and node gets evicted, and a client asks for those values?
When will the nodes get evicted if I set the max number of nodes to 0 (unlimited)? Is there a memory limit that I can set?
Is it possible to only persist the evicted nodes into the persistent store instead of all the nodes? This would be like the passivation of a session bean.
I apologize if any or all of my questions are naive.
Thanks.
1. Yes, using cacheloader implies every put is persisted so it has performance implication.
2. If the node is evicted and not persisted, it will return null. So it is application responsibility to check this.
3. We have overflowing persistence (or passivation) feature in our roadmap.
-Ben
Thank you. I would like to know about some production level deployments.
How many nodes, (with small objects associated) is typically recommended in a tree cache? Maybe a range would give me a better idea.
Are there any best practices for using TreeCache?
I've written a web service using JAX-WS. The service needs to load data from a file when it starts. I want to be able to specify when I deploy the web service what file to load the data from.
Seems like a simple enough task but I've not been able to figure it out. What's the best way to do this?
Here's one possibility. Have the web service open a properties file and get the info about the file to load from there. You can either place the properties file in the web service archive, in which case you would load it as any other resource (and you would probably want to deploy using an exploded archive), or you could place the properties file in the server/xxx/conf directory and use the jboss.server.config.url property to access it.
Thanks Peter,
I don't want to put a property file in the archive (not accessible enough), but a file in the conf directory would work. I thought of something like that but was looking at the data directory. I suppose the the conf directory is the 'official' place to put such things?
Either conf or data would work.
Hi,
I have a CDI bean which uses a conversation scope. In some situations I see the following exception in the server log:
2016-09-21 11:58:45,506 ERROR [io.undertow.request] (default task-4) UT005023: Exception handling request to /imixs-office/pages/workitems/workitem.jsf: javax.servlet.ServletException: /pa\nges/workitems/workitem.xhtml @30,74 src=\"/pages/workitems/forms/#{workflowController.editor}.xhtml\" WELD-000321: No conversation found to restore for id 6\n at javax.faces.webapp.FacesServlet.service(FacesServlet.java:671)\n at io.undertow.servlet.handlers.ServletHandler.handleRequest(ServletHandler.java:86)\n
My question is: how long is the default timeout. Should I set the timeout manually or can I use a configuration param?
===
Ralph
Hi Ralph,
Please check Weld 2.4.0.Final - CDI Reference Implementation and especially Weld 2.4.0.Final - CDI Reference Implementation
Thanks for your response. I thought that the timeout is coupled to the web session timeout. So thanks you for your hint.
I now placed the file 'weld.properties' into /WEB-INF/classes/
Is this the right place? Because all other files (e.g. jboss-web.xml) are located directly under /WEB-INF/
Yes WEB-INF/classes/ is the right place.
I now use the following code to synchronize my conversation timeout with the current httpSession timeout
conversation.setTimeout(((HttpServletRequest)FacesContext.getCurrentInstance().getExternalContext().\ngetRequest()).getSession().getMaxInactiveInterval()*1000);\n
I am trying to configure the timeout globally as well. I created WEB-INF/classes/weld.properties file and added
org.jboss.weld.conversation.timeout=1200000
I get:
14:54:30,695 WARN [org.jboss.weld.Configuration] (MSC service thread 1-4) WELD-001904: Unsupported configuration key found and ignored: org.jboss.weld.conversation.timeout
I use EAP 7. Any idea what I might be doing wrong? Thank you.
This configuration key has been added to Weld 2.3.4.Final - see [WELD-2113] Conversations - allow to configure defaultTimeout and concurrentAccessTimeout - JBoss Issue Tracker So if you are using the older Weld version then you will likely see this warning.
I've been experimenting some with jBPM and that's been going fine. Now I'm on the verge of recommending a candidate architecture that include jBPM for a new product my company is developing.
Before I go ahead with that recommendation I would like to get feedback on the current state of the product and the issue log. I see that Tom owns a whole lot of issues and unless he is superman (which he may be considering what he has accomplished so far) there seems to be insufficient resources to address all the issues in a timely manner.
Or maybe the issues are not things that would be encountered under normal circumstances and the product is quite stable enough to handle our product's workflow processing requirements.
Also, I would like to help out on this project if you need help.
Thanks,
Scott
Scott,
Not all issues are bugs, some/many are feature requests or some clarifications. Some are not even really assessed (they might be invalid for newer versions or not even valid at all). Getting them assessed and the list cleaned up is something that I'd really like to happen.
In most cases the issues are not encountered and the ones that are encountered are assigned and being worked on. E.g. issues about stale object exceptions. So in general it is quite stable
Help is always welcome. I personally would like to spend more time in the code, but keeping all the users in this forum happy (well almost all the users ;-)) is also important. Maybe we should, as the often do at mozilla, organize a assessment day. Do you have any idea in what area you would like to help out.
Thanks. That helps. I figured that was the case, but needed to ask anyway.
I saw somewhere that the next major version would be out by August and a minor release will be sooner. Even so, it sounds like the current release will be sufficient for us to get started developing.
As for helping out, I'd like to help with the Core Engine. I know that doesn't narrow it down much. I'm fine tackling whatever.
Scott
Scott,
Cool, I'll try informing Tom and see what is on the radar...
It would be *very* nice if someone could help out with the 3.2.x issues. As Ronald told help assess, prioritize and eventually implement them. Could you start a topic on the jbpm developer's forum for this?
Regards,
Koen
Ok. I'll start a topic on \"Design of JBoss jBPM\" to discuss 3.2.x issue tracking.
I am building a small application running on a Jboss server (jboss-4.0.0), that is utilizing form-based container managed security. Users access this application via a link on their portal page (that is not part of the jboss environment). Users have already been authenticated when they reach this portal page and the end user does not want to have to re-enter user-id or password on the login page.
The portal page is going to post the j_username and j_password information to a url, but I can't seem to find a way to get it accepted. I know you are not supposed to call the loginServlet directly, but has anyone found a way to get past this so that you can call the login servlet from outside the app server? I get the message: The request sent by the client was syntactically incorrect (Invalid direct reference to form login page).
Any thoughts!
Thanks
You cannot post directory to the form login page. This is used by the web container only when the it determines there is an unauthenticated session. To have seemless integration with other security contexts you need to look into sso configurations. Check the wiki:
http://www.jboss.org/wiki/Wiki.jsp?page=SingleSignOn
Hello there,
When I am doing file upload, the files are also getting created in \"bin\" directory of JBoss.
How to avoid this.
Thanks
Tara
Since you run the jboss server from the bin directory, thats the default location for all the uploaded files.
The MultiPartMimeServlet takes a constructor argument that can be set to be the file upload directory. Since its empty by default, it puts in the directory from where the server is started
As my computer science bachelor thesis I intend to build a web application in which users can design and share BPMN processes. A later objective would be to allow users to instantiate processes and track their progress in them, but this is strictly an optional target.
I intend to write this as a Rails application and I would like to use the jBPM Designer component as web based BPMN editor. So I though I just use JRuby with TorgueBox to let the Rails application run inside the Java AS.
My first problem now is, that TorgueBox only works on JBoss AS 7 at the moment, while the jBPM installer seems always to install a WildFly AS 8, which is still unsupported by TorgueBox. There doesn't seem to be much info on how to manually deploy jBPM into a pre existing JBoss AS 7 installation.
On the other hand, I have almost no experience with jBPM and I would like to know if my idea is at all a feasable approach. Any information helps, just tell me what you think about it.
Assuming you are using 6.1.0.Final, you can get a AS7 war here.
I tried to deploy that war file into my TorqueBox server. It didn't work, probably because TorqueBox is providing only enough dependencies to run JRuby and Rails applications. I tried to deploy the war file on a standard JBoss AS 7.1.1 and it didn't work either until I started it with the bin/standalone.sh -c standalone-full.xml configuration. At least that way, the deployment works, but a lot of warnings about missing database tables and stuff like that are thrown around. But as I cannot use a plain JBoss AS 7 it is not worth to investigate much. It seems not to be intended to get a full release of JBoss AS 7 and install TorqueBox into it, the only way to achieve something like this is to use EAP 6 and the so called TorqueBox EAP Overlay installed on top. Sadly, I guess that the license of EAP is not working for my project.
So what I tried now is to install the jBPM Designer as a standalone war into my TorqueBox specific JBoss AS 7 and that works well.
The next problem now is how exactly to embed the editor into my application. I found a video that says something about interacting through an AtomPub API with a component called Guvnor and that jBPM Designer would integrate with that Guvnor thing.
So how do I install Guvnor, integrate it into jBPM Designer and allow my application to talk to Guvnor through the AtomPub API? For all I know this could also be a wrong approach. If you know better, please let me know.
Unfortunately, I am not familar with TorqueBox. You will get some hibernate errors as the JPA entities are created for the first time.
Take a look at the user guide section on embedding. It might be what you need.
Just curious, why is the EAP LGPL license not kosher with your project?
Please tell me more about this EAP LGPL. All I found was a proprietary license which is free for \"development purpose only\".
I'll have a look at that guide.
It seems that this guide is only for embedding the \"workbench\". As far as I know, \"workbench\" is something different than the jBPM Designer BPMN 2 editor. Or am I mistaken here?
After someone named porcelli on Freenode IRC #guvnor channel pointed out to me that the former component Drools Guvnor is now replaced or fully integrated into KIE Workbench, the guide page about integrating the workbench into external application through an HTML iframe element made far more sense to me. With the jBPM demo deployment, I sucessfully managed to embed a BPMN 2 editor into a static web page.
Here is how I've done it:
\n<html>
\n<head>
\n<title>Test</title>
\n</head>
\n<body>
\n<h1>Test</h1>
\n<p>Editor frame below:</p>
\n<iframe height='800' src='http://localhost:8080/jbpm-console?standalone=&path=git://master@jbpm-playground/HR/src/main/resources/hiring.bpmn2' width='1000'></iframe>
\n</body>
\n</html>
\n
I extracted the path parameter from the metadata section inside the KIE workbench interface when viewing a BPMN process.
Although this is a nice start, now I need to be able to create new processes through some kind of API, so that they can then be edited in my frontend application. Also I wonder how access restriction can be implemented when using this iframe embedding. Thanks for any help so far and as before, any info, even if it is only bringing me one step forward, is highly valuable to me.
How can we call designer in iframe without passing already created bpmn process path. I want to create a new process witin my web app.
only localhost:8080/jbpm-console?standalone does not render designer.
Hi All,
I have downloaded nukes-snapshot. I have build the source code & deployed the ear files in JBoss. I am able to get the nukes home page. But bb module is not enabled. So i went to jmx-console of JBoss and started \"J2EEApplication=null,J2EEServer=Local,ServiceModule=nukes-bb.sar,j2eeType=MBean,name=nukes.modules:name=bb \" using mejbstart. Then forum is enabled. Is there any way to make the forum defaultly available.
Thanks in Advance
With Regards
Santhanam
Maybe a dependance is not resolved and prevent that module to start. I have to check that
Authentication - multiple login modules
I am attempting to use JBOSS 3.2/jetty & JAAS to authenticate against two seperate LDAP servers. It the first attempt fails then a second attempt is made against the second LDAP server. Seems that this fits well with the JAAS login module configuration. The first login module is define as \"SUFFICIENT\" the second is \"REQUIRED\".
The reality is that this configuration appears to work as if both login modules were \"REQUIRED\". I get authentication failure everytime.
If I remove either one of the login modules and define the remaining modules as \"REQUIRED\", I authenticate with no problem. Has anyone seen this behavior.
I have attched the login-config.xml file
I appreciate any help.
Romain,
You can probably produce a thread dump see what is going on, or run the process in the debug mode. With information provided it is very hard to guess. Using AdminShell is similar to using a Admin API, may be you can write a unit test to re-produce this behavior?
One another thing that really concerns me is that the pattern in which you are integrating Teiid into OSGI bundle. The \"EmbeddedConnectionFactory\" level integration is not publicly supported API, thus subject to change at any time in next versions (7.0). The safe bet is to use public APIs, like JDBC interface and Admin API, where Teiid team will suggest alternatives for migration paths if there are any changes to public API.
What I am trying to say is, wrap your OSGI bundle around Teiid JDBC Driver, once you make the first connection Teiid (either server mode, or embedded mode) will be up and running, then control the lifecyle with Admin API. See Designer's code to see how we used this same technique. What we found difficult in our development is that, Teiid uses lot of reflection based code, OSGI bundles with their classloading mechnisms it is hard to get it right. So, make your application only depend upon Teiid client driver, and bundle all the rest of the Teiid either in a separate bundle (Teiid embedded) or in external process (Teiid server) it is much easier to manage and upgrade in future.
Ramesh..
Does JBoss4 support distributed transactions?
Can I integrate JOTM(Java Open transaction Manager) into jboss?
Thank you for every advice.
I don't know about the version 4, but 3.x default tx manager does not support distributed transactions. However Tyrex, a distributed tx manager, is provided with JBoss. You only have to edit the config to enable it.
If you still wish to use JoTM, you have to write a service MBean for it that will bind JoTM to the right JNDI name. For more info look at the JBoss Docs and org.jboss.tm.plugins.tyrex.TransactionManagerService or org.jboss.tm.TransactionManagerService implementations.
If you succeed integrating JoTM in JBoss or find someone that already did that, please let me know. I'll probably have to do the same thing.
BTW is it true that Tyrex is not maintained anymore?
Jaksa
i have used Tyrex as DTM , but i found that it decreases the performance of Jboss. So i want to
use JOTM as DTM, the problem is that i am not familiar with JOTM and not very certain of how to do this work. If anyone knows it ,please give me some examples or some suggestions ,thank you in advance.
to Jaksa:
Now Tyrex has released its new version , but it doesn't
change any code related to DTM.
Hi,
New to SEAM, but well versed in Java...We are trying to use SEAM 2.1.2 along with SQL Server 2008. I am using the seam command line tool (at the moment) to try and generate (reverse engineer) our database (4 tables) into CRUD components. Part of a quick proto-type.
Our Application Stack is:
We consistently keep getting the following error in our SEAM web application after generating entities on the tables, all four generated entities/tables get the same error:
08:52:41,850 INFO [STDOUT] Hibernate: \n select\n top 26 swcmanufac0_.mfg_guid as mfg1_1_,\n swcmanufac0_.mfg_name as mfg2_1_,\n swcmanufac0_.num_discovered_products as num3_1_,\n swcmanufac0_.num_discovered_installs as num4_1_,\n swcmanufac0_.last_discovery_date as last5_1_,\n swcmanufac0_.restricted as restricted1_,\n swcmanufac0_.restriction_group as restrict7_1_,\n swcmanufac0_.restriction_on as restrict8_1_,\n swcmanufac0_.restriction_by as restrict9_1_,\n swcmanufac0_.reviewed_on as reviewed10_1_,\n swcmanufac0_.reviewed_by as reviewed11_1_,\n swcmanufac0_.source_system as source12_1_,\n swcmanufac0_.source_guid as source13_1_,\n swcmanufac0_.source_key as source14_1_,\n swcmanufac0_.excluded as excluded1_,\n swcmanufac0_.entered_on as entered16_1_,\n swcmanufac0_.entered_by as entered17_1_,\n swcmanufac0_.modified_on as modified18_1_,\n swcmanufac0_.modified_by as modified19_1_ \n from\n swc.dbo.swc_manufacturers swcmanufac0_\n\n08:52:41,884 INFO [SerializableType] could not read column value from result set: mfg2_1_; could not deserialize \n
The remaining StackTrace is large, but the key lines in it are:
\njavax.el.ELException: /SwcManufacturersList.xhtml: Error reading 'resultList' on type com.lmig.swc.action.SwcManufacturersList_$$_javassist_seam_3\n...\nCaused by: javax.persistence.PersistenceException: org.hibernate.type.SerializationException: could not deserialize\n...\nCaused by: org.hibernate.type.SerializationException: could not deserialize\n...\nCaused by: java.io.StreamCorruptedException: invalid stream header: 2D005800\n...\n
The initial SQL statement listed in the StackTrace can actually be run against the database without issues. So it appears
like the reverse engineering is working. We are also able to build, deploy and run the SEAM examples (booking, contactlist) that came with 2.1.2, so it looks like the SEAM/JBoss stack is working OK. Also, I went and defined a simple table on our SQL Server database of two columns (varchar(64)). I successfully reverse engineered this ONE table and that ONE table appears to be working in our SEAM web application.
Doesn't matter whether we deploy as EAR or WAR, same result.
The only difference I noticed between the table I defined and the other tables is the use of varchar() on my table and nvarchar() on the other 4 tables.
Can anyone give guideance on how we can pinpoint why this error is happening? Are there known issues with SEAM and SQL Server?
tia,
adym
This is fixed...I think.
I re-defined the SQL Server table using varchar() instead of nvarchar() column types...
We now get an error on 'resultCount':
\njavax.el.ELException: /SwcMfgList.xhtml: Error reading 'resultCount' on type...\n
see my other post...
thx,
adym
Hi all,
two questions about the Messaging configuration document at https://docs.jboss.org/author/display/WFLY8/Messaging+configuration
First: is my observation correct that WildFly 8 autocreates a queue if there is no config in standalone-full.xml and no \"...-jms.xml\" file in the ejb jar?
I observed this behavior with this ejb config:
@MessageDriven (activationConfig=
{
@ActivationConfigProperty(propertyName=\"destinationType\", propertyValue=\"javax.jms.Queue\"),
@ActivationConfigProperty(propertyName=\"destination\", propertyValue=\"jms/queue/MessageBeanQueue\")
})
public class MessageBean implements MessageListener
{
Is this an expected behavior? Or is it controlled by some switch?
Second: the doc contains some CLI snippets for adding a queue. It seems those are no longer correct. Instead of \"jms-queue add --name=myQueue --entries=queues/myQueue\" I had to use \"jms-queue add --queue-address=myQueue --entries=queue/myQueue\"
Similar problem for the sample \"read-resource\" call, which follows the \"jms-queue add\" snippet.
For me, it worked only with this syntax: \"/subsystem=messaging/hornetq-server=default/jms-queue=myQueue:read-resource\" (added \"hornetq-server=default\")
Could the doc be updated please?
Thanks
Wolfgang
Hi,
just want to bring this back.
Could someone take a look at the doc maybe-errors?
And it would be great if someone could bring some clarification on my question about autocreation.
Thanks
Wolfgang
\nWolfgang Knauf wrote:
\n\n
Hi all,
\n\n
two questions about the Messaging configuration document at https://docs.jboss.org/author/display/WFLY8/Messaging+configuration
\n\n
First: is my observation correct that WildFly 8 autocreates a queue if there is no config in standalone-full.xml and no \"...-jms.xml\" file in the ejb jar?
\n\n
No, queues are not auto-created in WildFly. You *must* define them prior to use them.
Note that if you are using JMS 2.0 you can use annotations (http://docs.oracle.com/javaee/7/tutorial/doc/jms-concepts005.htm#BABHFBDH) in your applications instead of defining them in standalone-full.xml or -jms.xml file.
\nCould the doc be updated please?
\n
I've fixed it. Thanks for the feedback
Thanks!
It turned out that the phantom auto-creation of Queues was some confusion on my side (probably to much playing around with/without standalone-full.xml and \"...jms.xml\" config)...
Many thanks for pointing me to the new JMS 2.0 config capabilities. Next thing to do for me: make it run in my sample ;-)
Best regards
Wolfgang
I have a JAAS Login Module which uses a custom principal.
Group callerPrincipalGroup = new MWareACLGroup(\"CallerPrincipal\");
System.out.println(\"PrincipalGroup: \"+callerPrincipalGroup.getName());
System.out.println(\"PrincipalGroupToString: \"+callerPrincipalGroup.toString());
callerPrincipalGroup.addMember(principal);
principals.add(callerPrincipalGroup);
The seems to do the proper thing in JaasSecurityManager. I put debugging code in there and info.callerPrinciple does get set correctly. However, it seems to get lost with the request.getUserPrincipal call. I still get
'org.jboss.jetty.security.JBossUserRealm$JBossUserPrincipal'
Where is the hook between JaasSecurityManager and request.getUserPrincipal?
I am using JBoss 3.0.6 with Jetty.
--John Fortin
> Where is the hook between JaasSecurityManager and
> request.getUserPrincipal?
Well, i'm afraid there isn't. The integration between Jetty and JBoss (JBossUserRealm) creates its own principal object, because it must implement jetty's UserPrincipal interface.
I'm afraid this is hard to change because of the way security is implemented in Jetty.
Maybe, Greg or Jules can comment on this....
Regards,
Peter.
So it is not possible to create an application which uses a customized principle? To me that seems to be a big disadvantage to using JBoss. We have data which must be returned to out applications in the principle object. Without being able to do that, we cannot develop with JBoss, and we definately cannot use it in a production environment.
Currently we are able to to this using Weblogic.
\"fortinj\" wrote:
BTW, I also tried this with JBoss/Tomcat with the same type os issue except that I get a SimplePrincipal instead of my custom principal.
--John
Hi,
I am using JBoss 3.0 final on Win2000 with Oracle 8.1.7 and experience the following problem/bug/whatever:
I have the following JBoss-QL statement in the jbosscmp-jdbc.xml:
<jboss-ql><![CDATA[
SELECT OBJECT(co)
FROM Companies co, IN(co.profiles) fp
WHERE fp.benuId = ?1
ORDER BY co.befiName
]]></jboss-ql>
Jboss generates the following two SQL statements out of this:
14:57:28,627 DEBUG [findByFirmenprofile] Executing SQL: SELECT t0_bf.BEFI_ID FROM BESTELLFIRMEN t0_bf, BEST_NUTZER t1_fp, FIRMENPROFILE t2_bf_firmenprofile_bf_R
ELATIO WHERE (t1_fp.BENU_ID = ?) AND (t0_bf.BEFI_ID=t2_bf_firmenprofile_bf_RELATIO.FIPR_BEFI_ID AND t1_fp.BENU_ID=t2_bf_firmenprofile_bf_RELATIO.FIPR_BENU_ID) ORDER BY t0_bf.BEFI_ANZEIGENAME ASC
14:57:28,657 DEBUG [Bestellfirmen] Executing SQL: SELECT BEFI_ID,BEFI_ADRE_ORT,
BEFI_ADRE_LAENDERKENNZEICHEN, BEFI_ANZEIGENAME, BEFI_FIRMENNAME, BEFI_LOGINNAME,
BEFI_ADRE_PLZ FROM BESTELLFIRMEN WHERE (BEFI_ID=?) OR (BEFI_ID=?) OR (BEFI_ID=?
) OR (BEFI_ID=?) OR (BEFI_ID=?)
The problem is, the first statement returns the primary keys in the right order, but when JBoss fetches the actual data, the order get lost, because Oracle returns the records in its own order, when you don't attach an ORDER BY clause.
Is there any way to get around this?
Thanks a lot,
Thorsten.
This JBossQL query is not related to the sql you posted. JBossCMP will generate an alias like t0_co for the Companies table, and specifically will not generate the alias t0_bf.
Also the second query is simply a load query and does not affect the ordering of a finder.
There is an existing bug that occurs when query results returned as a java.util.Set. When the results are returned they are loaded into a Set and the set scrambles the order.
Can someone give more detail on the \"order by\" bug? Where is the order scrambled by the usage of a Set? Thanks.
if your finder is declared to return a java.util.Set rather than a java.util.Collection, then the effect of the ORDER BY in the query is lost because the Set implementation used is unordered.
Steve
I use Jboss3.0.1RC1/Tomcat4.0.4£¬I have got a strange problem:
System display \"null\" on page When a String gotten from DB is empty.
I hate \"null\", I can accept a empty String, \"\",etc.,rathen than a string whose content is \"null\".
Weblogic6.1 have not the same problem.
How can i config System to return a empty String instead of a \"null\" String? because it destoy my page.
Must I check every String if it is empty and convert it to a empty String by myself,not to a \"null\" String automatically by system?
I need your help.
thanks
eh? If your String is null, then it will display as null, if it is empty string it will display as empty string. I don't think jboss has too much to do with this
Yeh,
if the String is empty with nothing it display \"null\", but i hope it display \"\"(a String with \"\",not a String \"null\").
Weblogic6.1 has not this problem,Maybe Weblogic have handled the empty String to a \"\" String before sending to page.
?
....
?
..........
!
I have to check the every String in every getXXX() method of Bean Class of every Entity Beans.
Greetings all!!!
I have upgraded to Richfaces 3.1.3.GA in hopes that it would correct a problem whereby the RichFaces related stylesheets are not being loaded in my app. The strange thing is that sometimes they are loaded but more times than not, they are not loaded. I have tried the refresh button within the browser, restarting the app server (JB 4.2.2.GA) and have flushed the browser (IE6 & Firefox 2.0.0.11) cache all to no avail.
Any one else run into this problem? Any ideas on what else I can try?
Thanks in advance for any and all help!!
Earnie!
You can try new feature: http://www.jboss.com/index.html?module=bb&op=viewtopic&t=126207
Thanks for the reply Nick. I tried:
<context-param>\n <param-name>org.richfaces.LoadStyleStrategy</param-name>\n <param-value>ALL</param-value>\n</context-param>\n
Have done some more research and discovered when my login page is first displayed, the link to the richfaces stylesheet looks like this:
<link rel='stylesheet' type='text/css' href='/WAITC/a4j_3_1_3.GAorg/richfaces/skin.xcss/DATB/eAFjlrb0AAAC6gHS'/>
I added this to my components.xml and flushed the browser cache and pages subsequent to my login page now use the style. The login page still does not use it however. Not to mention that this makes the app VERY slow!
<web:ajax4jsf-filter enable-cache=\"false\"/>
in files, that you attached to JIRAs issue, in web.xml missing filter for rich* components.
Add following strings in your web.xml:
<filter>\n <display-name>Ajax4jsf Filter</display-name>\n <filter-name>ajax4jsf</filter-name>\n <filter-class>org.ajax4jsf.Filter</filter-class>\n </filter>\n\n <filter-mapping>\n <filter-name>ajax4jsf</filter-name>\n <url-pattern>*.seam</url-pattern>\n </filter-mapping>\n\n <context-param>\n <param-name>org.ajax4jsf.SKIN</param-name>\n <param-value>wine</param-value>\n </context-param>\n
and please add one more context-param:
<context-param>\n <param-name>org.ajax4jsf.VIEW_HANDLERS</param-name>\n <param-value>org.jboss.seam.ui.facelet.SeamFaceletViewHandler</param-value>\n </context-param>\n
I ran into the same problem using 3.1.3 or 3.1.4 snaps with JSF RI 1.2.07 and a JDK 6 on Glassfish V2b58.
Still unexplained to me.
For your infos, I just installed the latest glassfisg version glassfish-installer-v2ur1-b09d and all works fine now !
Another computer science mystery...
More clues: I had upgraded JSF RI to version 1.2.07 on my previous glassfish. The Glassfish I just installed contains 1.2.04.
In order to test the assumption that RI 1.2.07 has some regression, I upgraded the JSF lib to 1.2.07 again, and...css don't load anymore !!
Ok, rollback to 1.2.04...css load again !!
This is clear, something is wrong with the JSF version. I checked on the JSF forums, nothing like that matches. To be continued...
Good afternoon:
I'm developing a Richfaces application and I'm using 4.1 version. I'm having a problem when I try to add a new created skin. I followed the Richfaces 4 guide in witch it's specified that the new skin must be named mySkin.skin.properties and adding to META-INF/skins directory. After that, I modified the web.xml, including the appropiated declaration (considering the guide):
<context-param>
<param-name>org.richfaces.skin</param-name>
<param-value>mySkin</param-value>
</context-param>
When I add the application to the Tomcat Server 7.0.22, and I start it, I get the next exception:
org.richfaces.skin.SkinNotFoundException: skin with name mySkin not found.
Looking for help in the disscusions I found this topic:
http://community.jboss.org/thread/162295
the proposed solution works for me, that is by adding my own skin with the rest of the properties files of my application. However, I'm not sure if it's the most appropiated way to do that.
Anybody else has other solution?
Thank you very much in advance
Regards
I wonder how to set current page in dataScroller in managed bean when I come back to page. What am I doing wrong?!
jsp:
<rich:datascroller align=\"center\" for=\"list\" maxPages=\"20\" binding=\"#{bean.scroll1}\" page=\"#{bean.pageNumber}\" id=\"sc1\" reRender=\"sc2\"/>
bean:
scroll1.setPageIndexVar(String.valueOf(page));
Thanks,for all responses.
We do it a different way where we inject the uicomponent into the bean using seam, like
@In(value=\"#{uicomponent[idOfDataScroller]}\"
private transient UIDatascroller scroller;
where the UIComponent is the org.richfaces component.
Then in the code we do a....
scroller.setPage(\"nameOfFacetInRichDataScroller\");
But in your case I would have thought the value of the page property on the rich:datascroller should do the job.
Are you rerendering the \"list\".
I have seen the error before where you set the page but it does not get propergated back to the page, havent figured it out as yet.
you should use page binding instead of pageIndexVar which is an attribute to define request scope variable for page access to current index(like var in dataTable).
I tried to deploy one JCA 1.5 Adapter RAR file to the JBoss JCA 1.0.0.Alpha11, but failed with the following error:
Caused by: javax.naming.NoInitialContextException: Need to specify class name in environment or system property, or as an applet parameter, or in an application resource file: java.naming.factory.initial
at javax.naming.spi.NamingManager.getInitialContext(Unknown Source)
at javax.naming.InitialContext.getDefaultInitCtx(Unknown Source)
Is it because the RAR file does not follow JCA 1.6 Spec ?
Looks like the naming configuration isn't found by the RA. The configuration is located in
config/jndi.properties\n
in the distribution.
Try and see if you can debug what goes wrong and post a bug report with a sample resource adapter which shows the problem in our JIRA installation, as JCA 1.5 resource adapter are supported by the project.
Thanks.
PS: These types of questions belongs in the user forum.
http://www.jboss.org/wiki/Diff.jsp?page=MailServicesBuilding&r1=-1
Have you tested 1.6? Last I tried it failed. Never bothered to see why.
Okay.... cool. I'll bug ya then :-)
This is going to be awesome. If this goes well we'll have a hell of a test.
1.6.1 works. 1.6.0 gives some weird error, which I didn't bother to investigate either.
Ant > 1.5.4 is needed to use the cmpversion attribute to the ejbjar task. Unless cmpversion=\"2.0\" is added to ejbjar task in build.xml, the jbosscmp-jdbc.xml file does not seem to get added to the ejb.jar files. I need this for my entity mail list stuff. I guess an alternative would be to change the packaging code, but give 1.6.1 a go.
Cheers,
Kab
Should read:
Ant > 1.5.4 is needed to use the cmpversion attribute in the ejbjar task.
If it works I dun mind using it.
Shit, you're working on entity mail list? I'm about 1/2 way through... When will you have it in? I promised tomorrow so..
Just testing and tidying up now, so tomorrow morning :-) Unless something majorly wrong happens...
Can't figure out the unit tests for EJB stuff, but have added a test method in the MBean interface to prove that it all works.
Hi,
I noticed that in AS 7.1.1.Final UserTransaction that is returned from JNDI look up is different that one in AS 7.0.2>
7.1.1 is org.jboss.tm.usertx.client.ServerVMClientUserTransaction
and on 7.0.2 is com.arjuna.ats.internal.jta.transaction.arjunacore.UserTransactionImple.
And this seems to cause me some troubles on the application side. On commit some fields on an abject are reset (at least I was able to observe such behavior). Why that has changed and can it be somehow configured to use arjuna usertransaction be default or any other thoughts are more than welcome.
Cheers
Maciej
The internal implementation class shouldn't really be a concern to the application code. If it does, that we should find out what the real issue is. Can you explain a bit more about what you are running into?
After bit more digging into the issue it turned out it is not a matter of implementation class of UserTransaction but missing TransactionManager bound to expected JNDI context. In 7.0.2 it was bound to java:jboss/TransactionManager and in 7.1.1 it is bound to java:/TransactionManager
So, in fact on 7.1.1 transaction manager was found and transaction synchronization was registered and on commit (after completion) there was additional logic executed on application side causing reset of mentioned fields.
Thanks, I consider this as answered as it has nothing to do with different implementation classes of UserTransaction.
UI Bindings
<rich:dropDownMenu direction=\"bottom-right\" jointPoint=\"tr\"
submitMode=\"server\" binding=\"#{backing_activityCreate.menuComponent}\">
<a4j:support event=\"onitemselect\" action=\"#{backing_activityCreate.changeActivityType}\" >
<f:setPropertyActionListener value=\"#{backing_activityCreate.activityTypesMenu}\" target=\"#{backing_activityCreate.selectedActivityTypeCode}\" />
</a4j:support>
</rich:dropDownMenu>
for (AppComActivityType menuItem : activityTypesMenu)
List<AppComActivityType> childActivity = getActivityBaseEJB().getAllActivityTypesByParentActivityTypeId(firmId, menuItem.getActivityTypeId());
<f:setPropertyActionListenerHi,
You can assign action to htmlMenuItemSub/htmlMenuItem components.
Actually i need to get the value for the selected menuItem. for eg: if i have (Bozho,Skaffman,anandhinaveen) in MenuItem. And if i have selected Bozho item from the dropDownMenu, i need to get the value of the menuItem in the backing bean as Bozho. so that i can proceed further with that value..
You do not need any additional supports, bindings and so on. Just pass parameter using a4j:actionparam or f:param and get it from request map in your item action or listener
Hullo,
anyone ever had issues with an application that works 100% OK, but when calling HistoryService queries only gets zero results?
This is probably not an issue in the \"JIRA\" sense, rather I am looking for something arcane I may be overlooking. I am using 4.3 on a mem: hsqldb db, wrapped the ProcessEngine into a signleton, and accessing it from an application being tested with plain Junit TestCases. Everything running smooth as expected, except for ProcessInstanceQueries and ActivityInstanceQueries consistently returning an empty list.
I tried turning autocommit to true (the only hint in that direction I coudl find) in the hibernate config file, to no avail.
Anyone that ever experienced a similar problem has any hints as to what I should be double-checking?
TIA
Did you try to enable logging for Hibernate to check if Hibernate is issuing queries?
I have similar setup and there are no problems with running history queries. Perhaps you could past a unit test...
Cheers,
Maciej
Maciej,
thanks very very much for the hint - activating hibernate logging results in two queries being issued against tables JBPM4_HIST_PROCINST and JBPM4_HIST_ACTINST when asking for the process and activity instances.
However, to the best of my understanding, no inserts or updates ever get fired in the course of the execution. It seems I somehow managed to disable history completely . Any hints about how I could possibly have done this?
<history-service /> is in jbpm.default.cfg and I am not sure what else I could be checking...
Bye and TIA,
Emp
For the record, I found the issue, largely thanks to Maciej's suggestion of looking at the hibernate log, which pointed me in the right direction (tables OK, nothing logged).
I apparently had carried over a jbpm.default.cfg.xml file from an older (4.0?) version, and that version did NOT have the following lines:
<history-sessions>
<object class=\"org.jbpm.pvm.internal.history.HistorySessionImpl\" />
</history-sessions>
This apparently (correctly) causes all calls to the history service to succeed, without anything getting recorded. So, for anyone out there, if you find yourself with getHistoryService().createHistoryProcessInstanceQuery().list() or getHistoryService().createHistoryActivityInstanceQuery().list() returning zero results, I suggest you check your jbpm.default.cfg.xml file.
Bye,
Emp
Hi all,
i have made some changes in the source of the rich:tab renderer,
file name (org.richfaces.renderkit.html.TabHeaderRenderer.java) in the function writeLabel() to achive a look and feel which otherwise would not have been possible through the attributes of richfaces.(i have added the html tags and in the that contains the label).
I have achieved the look & feel but now i would like to write a class of my own that would maybe extend this class or some other tab render class where i will write only my changes and bundle it in a separate jar. the rest of the things should be taken from the existing renderer of richfaces. I'm trying to do this cause in future if i have new version of richfaces(with extra features) then i don't have to again chance the new richfaces source to get what i want and make a new jar. i will directly deploy the new jar and have my changes.
can anybody please help me out with it?
thanks & regards
Jinson
Hello Folks,
There is a case:
I'd like to to expose some data from DB via REST service on GateIn, but I'd like the REST service to know who is requesting the data. How to check whether a request comes from user who is already logged in ( to make it simplest : everything happen in the same browser), or from an anonymous user.
Building an JAVA-REST service it's not a big problem from now on - we've got a sample application by courtesy of Trong Tran:
(...) I just created a simplest rest services for GateIn at https://github.com/trongtt/examples/tree/master/simplest-rest-services.
with the example, it could show the hello messages when you go to URL like following : But the qustion is how to deal with SSO/Oauth or another way to pass the user session into REST service ? ps. adding ?JSESSIONID=XXXXXXXXXXXXXXX to the requested resource path also seems to not work Any info, comments will be highly appreciated Cheers, Paweł
Hello Pawel,
You would need to provide more information on what you are trying to accomplish. I.e. what do you need user identity for? Is it just to authorize access to the resource or your response depends on user identity attributes - age, service contract level, etc. Without that I can only give you very general response. First, for anonymous users it is pretty strightforward - you just don't have user idientity in the http request. For authenticated users it is a bit more complicated. Generally speaking you have portal as identity provider i.e. party responsible for user authentication and your REST service as a service provide that needs to accept request made on identity provider behalf. There are number of ways to pass user idenity - either in the http header or even as request parameter. This part is easy you ca use a simple servlet filter to extract the identity. The challenge here is that your service needs to 'trust' the identity supplied in the http request. If it is to simply authorize access to the resource something like oAuth may work well for you. JBoss RestEasy can handle oAuth tokens for you. Or you can have some security gateway component e.g. Siteminder or Webseal in front of your service handle SSO with portal container and authorize access to the protected resource(s). In more complicated scenario service provider typically requires access to user identity attributes. The right solution depends on your requirements. It could be as simple as SAML assertion in the http header signed for trust purposes. A security gateway can do the job of generating SAML assertion and inserting idenity attributes. Or it could be a signed reference to the resource your service has to call to retrieve identity attributes. I have also seen implementations using private resources i.e. something like http://hostname/rest/HelloWorldRest/<user idenity>/hello. Identity provider can supply reference(s) to the private resource(s) via hypermedia links.
If you don't mind being dependent on the underlying portal container, I believe you can use
ConversationState state = ConversationState.getCurrent();\n
In order to be authenticated you need to access the private context so for the helloworld service you mentioned it would be
http://localhost:8080/rest/private/HelloWorldRest/hello
Hello ,
Thank you Guys for your time and for the answers! It everything is true - no doubts about it.
In regards to Andrei suggestion to give more details on this topic I've prepared a simple picture ( as a picture is worth of thousand words ) to give more details about the context (hope it does).
We are not planned (yet) to use any external IDM or oAuth - for the time being. I'd like to leverage a GateIn's standard authorization mechanisms. Means if a user is already logged into the application, can view a portlet (jsp+ajax+extJS). The portlet, to be more precise, the extJS stuff will make call to a RESTful service exposed via GateIn way.
My basic question is, how to make that the REST service will be informed that the request came from logged user (from logged user's portlet) ?
http://localhost:8080/rest/private/HelloWorldRest/hello - even though we access the resource through the private we are being asked for input user/password - it isn't desired effect in our case (storing plain password at client side isn't a solution)
There is also another way to expose REST services - via RestEasy. As a matter of fact our app's stack utilize Seam/EJB3 so RestEasy is also an option. Creation of a REST services via RestEasy is even easier then in GateIn way - but the more ingridients the more potential risk of misunderstanding, especially that Seam tends to make many things behind the scene (-: - that's why we've ve chosen the GateIn way as a first approach - to gain better understanding of SSO stuff.
If we succeed with this story I think it would be a good marerial to write a smal article, because REST and DataCentric approach seems is getting more popular day-by-day.
Cheers,
Hi Pawel,
I'm wondering if you were thinking about using serveResource ( in your portlet ) instead of a Rest service for your usecase.
Actually, i don't really get why you chose using Rest for the usecase which the Rest service only serves your portlet, right ?
Hi,
I think this howto can help you : http://tugdualgrall.blogspot.com/2011/04/how-to-protect-your-rest-service-and.html
I'm trying to wrap my head around Arquilian and am just not fundamentally understanding the difference between in-container mode vs. as-client mode. My understanding is that, with Arquilian/ShrinkWrap:
So, from my perspective, everything just \"feels\" like it would always be in-container, since the tests are always executing inside a container.
Please help me understand the differences between these two modes and what types of integration tests both modes are suited for. A concrete code example that explains why/how you run the tests in both modes would be enormously appreciated! Thanks in advance!
Hi Zac,
your description fits exactly in-container mode (plus your test class and additional arquillian modules such as enrichers for EJB etc will be deployed to the server as well).
When it comes to client mode, you deploy your shrinkwrap test archive (without repackaging as it's the case for in-container testing), but tests are performed from your local jvm (outside of the container in remote way).
More information can be found in our confluence space - https://docs.jboss.org/author/display/ARQ/Test+run+modes
I hope that helps.
Cheers,
Bartosz
Thanks Bartosz!
So, to summarize:
Is this the gist of it or am I missing something here? Please understand I have already read the \"Test run modes\" page that you cited and am still having a tough time \"seeing the forest through the trees\" here! Thanks again!
You are welcome.
Just to clarify.
in-container mode: Arquillian deploys your Shrinkwrap archive together with some additions (namely extensions you are using or which your container adapter relies on) to your target environment (through container adapter such as JBoss, Tomcat, Websphere to name a few). Then it runs your tests within this environment, collects the result and send them back to your test runner - could be IDE or a build tool (such as mvn's surefire).
as-client mode: Arquillian deploys your test archive and runs tests locally talking with your target environment from outside (local jvm)
I hope this gives you slightly better picture
Cheers!
So \"As-client\" mode fits perfectly functional test of the web application or testing rest services using client.
Ahhhh, so as a rule of thumb, one should probably use \"as-client\" for functionality testing and \"in-container\" for integration testing, correct? Thanks again!
Or in other words you might think of it in the following way:
in-container mode for white box testing
client mode for black box testing
but obviously that's only one angle
I have installed JBoss-3.0.0_Tomcat-4.0.2.tgz under
Windows 2000. The server seems to work except for the
JSP-engine. Some classes cannot be found:
org.apache.jasper.JasperException: Unable to compile class for JSP
C:\\jboss-3.0.0beta\\jakarta-tomcat-4.0.2\\work\\localhost\\html\\main$jsp.java:3: Package javax.servlet not found in import.
import javax.servlet.*;
... and so on ...
Seems to be some kind of classpath problem. But even if
I ensure that the jars are in the classpath, it didn´t
work. But if I start tomcat standalone, the JSP engine
works fine.
I couldn´t figure out how the tomcat integration
works and which configuration files are affected.
Can anyone give me a hint ?
Thanks,
Felix
One of the troubles I come across when working with JSF applications is the inability to verify that a user should even be on the page that is requested. I am wondering if Seam helps out in this area in any way.
Let me give an example scenario.
In the hotel booking application, a request for /hotel.seam or /book.seam without a conversation id renders a partially broken page (empty hotel data). Some of the logic in the backing bean could assume that a valid hotel is available, but in this case it is not. While the booking application doesn't trip on this, it is very likely that another application could (depending on what is going on in those backing bean getter methods). From the standpoint of a user, I would expect the application to redirect me to a start page if I attempt to access the application somewhere in the middle of a flow.
Is there any place I can perform validation logic before PhaseId.RENDER_VIEW? If the conversation is not valid or a certain variable is not present, I would like to redirect the user somewhere else.
ah, answered my own question (or more accurately, the Seam developers already answered it). Page actions are the feature for which I am searching and they are covered in section 4.1.1. of the Seam manual.
My followup question is, why not implement this in the Hotel Booking app to prevent access to the two pages mentioned if the hotel is not set?
Hi,
I want to use JBossSX with the DatabaseLoginModule. I'm really confused with this groups. What they are for? I was thinking that in a group you can put a sertain number of roles and after that assign a principal to a group and the principal will have all the roles which are in the group.
I think that's not the case. In the Roles table I have to have a principal, a role and a group. Why do I need both, a role and a group?
In the Scott Stark's article in JavaWorld there is
java.CallerPrincipal=caller_java
duke.CallerPrincipal=caller_duke
Here, there are two principals in one group, but with two different roles. This I can't understand at all. I was thinking that all the principals in one group have the same roles.
I looked in JBoss docs, but I didn't see any explanations how the roles, principals and groups are related. Please help me!
Thanks in advance, I appreciate!
IMHO, this is the deal:
the roles table(file) does not describe only roles, it describes the association between a user and it's roles and the association between a user and a caller principal. In this table(file) there are two categories of associations:
1--A one to one association between an user and a principal: the principal associated with the user here, will be returned by the context.getCallerPrincipal() method call inside a bean.
2--A many to many association between a user and a role. The roles defined by this associations are used in <method-permission> tags from ejb-jar.xml
Greetings,
seven
I have the following setup:
JBoss 2.4.4/Tomcat using JbossMQ
A topic that various publishers post messages into.
A MDB that has a durable subscription to the topic.
Problem:
If messages are not handled before shutting down JBOSS (ctrl-C, for example - but it could be a machine or hard software failure instead), the messages are not sent to the MDB upon restarting JBOSS.
Here's what I see when I look in the db/jbossmq:
- there is a file for the topic/subscriber combination with a .dat1 extension which is non-zero in length. This file exists during the first JBOSS run.
- a second file is created for the topic/subscriber combination when I do the restart - it has a .dat2 extension.
So, how can I get those messages to survive across a JBOSS stoppage or failure?
Also, I notice that the message files mentioned above continue to grow (at least when using a small number of messages) even if the messages are properly delivered and committed (at least I think they've been properly committed). I'm using the rollinglog persistence manager.
Hello,
I posted a similar question already to the Springframework support forums without success so I wonder if somebody here had/has the same problem:
In order to reconnect a Springframework based JMS client application after connection failure or cluster failover I want to implement and set an ExceptionListener. I have done that successfully with a plain JMS test client but I have no idea how to do this with Springframework, as it hides the API away from the programmer. I read somewhere that a way to solve that would be to write a ConnectionFactory proxy that takes care of returning only valid connections similar to a connection pool.
Does anyone have done something like that or can explain me a little bit how this would be done?
Thanks in advance,
Torsten
Hi,
We are using JBoss Portal version 2.6.5SP1 in our comopany. We didn't change the configuration of JBoss Portal that come with a default option of EHCache for the portal DB. I also checked the configuration of the hibernate and it seems correct which mean that the first\\second levels are true and everything there seems fine.
The issue is that we have some gaps in our log files and we found out that the times spent on the Jboss Portal DB. The issue is that in every login we spend at least 300ms for it and also there are some places during the login that we spend the same amount of time on them (another 2-3 times).
When I checked the Hibernate logs I found out those line (this is from the first login, and those are just samples from 2000 lines):
[org.hibernate.jdbc.AbstractBatcher] about to open PreparedStatement (open PreparedStatements: 0, globally: 0)
[org.hibernate.SQL] select displaynam0_.INSTANCE_PK as INSTANCE1_0_, displaynam0_.TEXT as TEXT0_, displaynam0_.LOCALE as LOCALE0_ from JBP_PORTAL_OBJECT_DNAMES displaynam0_ where displaynam0_.INSTANCE_PK=?
[org.hibernate.cache.EhCache]: key: org.jboss.portal.core.impl.model.portal.PortalImpl.windowStates#119 \\\\this line showen many times with diff keys
[org.hibernate.loader.Loader]: [user:] [context:] result row: EntityKey[org.jboss.portal.core.impl.model.portal.PortalObjectImpl#95] \\\\this line showen many times with diff rows
and finally I see those 2 lines at the end:
[org.hibernate.engine.StatefulPersistenceContext]: initializing non-lazy collections
[org.hibernate.loader.Loader]: done loading collection
I understand that at those times we are loading things from the DB but: a. Why this is happenned every time instead of using EHCache. b. how I can avoid it or minimize the times that spend.
* I uploaded the full log file in case that you want to see everything
Thanks for any response!
JBoss Portal community project is not currently in active development. The full force is on GateIn Portal (http://www.jboss.org/gatein/) so that's something to keep in mind when you start your work.
If you need to use JBP for whatever reason, you should at least use 2.7.2 which has several fixes found in 2.6.* versions. The supported version of JBP is Enterprise Portal Platform (EPP 4.3) and I believe lots of these perf problems are either solved or support folks will give you tips on how to do that.
First of all thanks for you r quick response.
I saw the GateIn site and I went there. When I click on Developer Forums at this page - http://www.jboss.org/jbossportal I achieved the same forum where I wrote this discussion (question).
I understand the first part of upgrade the version to 2.7.2 and I will try it right now but I didn't understand the EPP part, this is something else that I should do or this is part of the upgrade? which support folks can help me?
Thanks again!
You can see details of EPP at http://www.jboss.com/products/platforms/portals/. In short:
JBP is a community project and, as such is not supported with an SLA. Any questions or problems can be raised through user form (this forum you are using) and where the community will provide their best efforts to assist. This project is included in the JBoss Enterprise Portal Platform for which a support subscription is available.
Hi,
First of all I found out that it isn't so wasy to upload to version to the new one (2.6.5SP1 --> 2.7.2) and there are a lot of changes. I had a lot of problems and some of them I still didn't fix and don't have any idea how to.
At the current situation I found out that even with the new version we still have the same amount of delay at the beginning and between each protlet download so at the current situation I don't know how much is going to be helpful to upgrade it to 2.7.2. There is something that I need to change when I move to 2.7.2 in order to make it faster? Should I do the recommendation at JBP site to upgrade the JBP even at the new version of JBoss Portal?
Thanks again!
Hi,
The doc says (http://docs.jboss.com/seam/2.0.3.CR1/reference/en-US/html/events.html):
Seam rolls back any active transaction when an exception occurs and is uncaught in the Seam component layer
so I expect that if code throws an ApplicationException, the transaction will be rolled back.
However what happens is this:
The rollback interceptors check it. Because it is an application exception it doesn't rollback the transaction The SeamPhaseListener see a transaction and commit (status from 0 to 6)
The exception filter sees the exception, but no transaction(status is 6) so do nothing.
We want the behavior describes in the doc, that is Application Exception rolls back the exception without having to change the exception code (not adding annotation to it).
How can we do this ?
alex
Hi all,
I am new in Jboss as 7.
I have Domain of 4 Node . N1 , N2 ,N3 , N4
Here N1 as master node (Domain controller).
I have simple web application in which i want to create page which have functionality , something like to view all node in cluster
and to make Node unregistor from the cluster using checkbox or radio button .
I know about domain managment using \"Commnad Line Interface\" , \"Web Managment Interface\" and \"Configuration Files\".
But i have no idea how to manage Domain from simple web application .
I have no idea how to communicate with \"CLI\" or \"Configuation File\" from web application to domain controll.
My requirement is very basic means, i want to show only all node's name and node's Ip address
and want capability to unregister particular node from Domain using simple web page.
You might use the management API directly from your application.
A good point to start is here
I have a few dataTables (which are refreshed usign a4j poll) in my application and when you right click on row the menu is shown (something like enable/disable).
What I would like is when you click on enable/disable the dataTable row style/color would change but the data would not be retrieved since it's lightweight client and I refresh the page only every 1 min.
Any ideas?
Tnx
Not RF specific but just JS and css usage. Params passing to the menu shown at our demo. All you need - to use your own JS to change the style for the current row and to store the state in some hidden. hiddens will be passed to server by poll periodically.
Is there a way to get the rich component instance in bean? That way I could check if there is about 2sec before refresh and then I could delay the request for disable/enable for about 2-3sec. That would also be an option. Since the application refresn rate is low.
Hi all,
I use Selenium to test my UI. It work fine with IE, even ajax request with waitForCondition can be tested (only with IE, it does not work with Firefox so I cannot use it for my tests). But there are problems even with IE ...
In the UI I use a lot of rich:trees but I can't select any nodes of them by selenium. I tried mouseDown (which works in Firefox), mouseUp, click, keyDown(\"\\13\").
Can anybody help me with this problem ?
Greeting
Matze
Hi,
I am using JBoss LDAP authentication module for LDAP authentication for my application. If i enter the username with incorrect case , i am not able to authenticate...The issue here is that JBoss authentication passes but it returns the user name entered and not the retrieved user name from the repository.
In Weblogic we have some thing called as
Go to Security > Realms > Providers > Authentication
Click the LDAP authenticator. Click Configuration > Details.
Select \"Use Retrieved User Name As Principal\" checkbox.
If we check this box it returns the Reteived user name as principal and authentication passes successfully...
Do we have the similar settings for Jboss 3.2.1
Thanks in Advance
Debra
Please respond to my Query
Thanks
Is there a way to set the principal of the callBackHandler explicitly
Your request is hard to understand, but I think what you are asking for is to have the JAAS layer set a new user principal (different from the username principal) obtained from realm to replace the username principal. If this is correct, then I think that is why JBoss has the 'CallerPrinicpal' Subject usage pattern. So I think you may need to extend the LDAP login module to handle this in a custom fashion.
Does this solve your problem? cgriffith
Hi,
I have a question regarding the transaction attributes and Jboss behaviour. Here's my configuration:
Jboss 2.4.10
Oracle 8i
P6Spy driver to log the DB calls
EJB application combining around 50 beans (EJB 1.1)
Now, if I set the transaction attribute to Supports for a finder method belonging to a CMP entity bean, the p6spy reports a rollback after the specific SELECT statement (I suspect other methods with the 'Supports' tx attribute or more generraly, the unspecified tx context as defined in EJB 1.1 spec section 11.6.3, behave the same way). In fact, I have lots of places in which the Supported transaction attribute is used and I have observed that a rollback occures in such situations. However, no error condition is reported either by the container or the database and the application behaves as expected (that's why I have discovered this so late).
My question is: is this a correct behaviour? Why the plain select statements could not be executed without a transactional context, thus avoiding an unnecessary rollback?
I can only guess that the rollback itself takes some time to to execute and this is in my oppinion a complete unnecessary action.
This post also relates to one earlier on a related subject:
http://www.jboss.org/modules/bb/index.html?module=bb&op=viewtopic&t=forums/ idea/thought would be greatly appreciated.
florin
The delayed operation optimization in PersistentBag and PersistentList can corrupt the second level cache under specific conditions.
Assume we two entities Parent and Child with a bi-directional one-to-many from Parent to Child with 2nd level caching enabled for both entities and both sides of the relationship and at the beginning of the transaction the 2nd level cache is empty. The transaction flow as follows will result in a corrupt 2nd level cache:
// add a new Child to Parent
// the add operation is added to the delayed queue and the collection is not initialized or marked as dirty
Child newChild = new Child();
newChild.setParent(parent);
parent.getChildren().add(newChild);
// do something that causes an entity manager flush, either explicit call or a query
// this causes newChild to be written to the database
entityManager.flush();
// remove an existing child from parent
// the remove() call will cause the collection to get initialized from the database, which includes the newly added child
// because the collection does not think it is dirty, it sticks the relationship data in the 2nd level cache
Child oldChild = parent.findChild(\"name\");
oldChild.setParent(null);
parent.getChildren().remove(oldChild);
entityManager.remove(oldChild);
// at this point if the transaction fails, the 2nd level cache is not cleaned up and the Parent->Child relationship contains an ID that does not exist in the database
// even if the transaction eventually succeeds, there is a period of time when the 2nd level cache contains uncommitted data not visible to any other transaction
The code in AbstractPersistentCollection.afterInitialize() that controls whether or not a collection is added to the 2nd level cache does check the delayed operation queue and block the 2nd level cache put if it is non-empty. But it does account for the fact that a previous flush() within the same transaction may have cleared the delayed operation queue. It seems to me that once a collection has been modified in the course of a transaction, it should veto the 2nd level cache put for the duration of that transaction.
Our workaround is to provide our own UserCollectionType that forgos use of the delayed operation queue or overrides the endRead() method to veto caching.
We decided instead to modify the conditions under which a collection is placed in the 2nd level cache during loading. Our patch to Hibernate 4.1.7 is below.
diff --git a/hibernate-core/src/main/java/org/hibernate/collection/internal/AbstractPersistentCollection.java b/hibernate-core/src/main/java/org/hibernate/collection/internal/AbstractPersistentCollection.java\nindex 80db62e..d90ab51 100644\n--- a/hibernate-core/src/main/java/org/hibernate/collection/internal/AbstractPersistentCollection.java\n+++ b/hibernate-core/src/main/java/org/hibernate/collection/internal/AbstractPersistentCollection.java\n@@ -65,6 +65,8 @@ import org.hibernate.type.Type;\n public abstract class AbstractPersistentCollection implements Serializable, PersistentCollection {\n private static final Logger log = Logger.getLogger( AbstractPersistentCollection.class );\n \n+ private static final boolean useModifiedFlag = !Boolean.getBoolean(\"geneinsight.hibernate.disable-modified-flag\");\n+ \n private transient SessionImplementor session;\n private boolean initialized;\n private transient List operationQueue;\n@@ -78,6 +80,10 @@ public abstract class AbstractPersistentCollection implements Serializable, Pers\n // collections detect changes made via their public interface and mark\n // themselves as dirty as a performance optimization\n private boolean dirty;\n+ // track if any change happens to this collection\n+ // set when dirty is set, but wont be cleared \n+ private boolean modified = false;\n+ \n private Serializable storedSnapshot;\n \n private String sessionFactoryUuid;\n@@ -105,8 +111,17 @@ public abstract class AbstractPersistentCollection implements Serializable, Pers\n \n public final void dirty() {\n dirty = true;\n+ modified = true;\n }\n \n+ public final boolean isModified() {\n+ if (useModifiedFlag) {\n+ return modified;\n+ } else {\n+ return false; \n+ } \n+ }\n+ \n public final Serializable getStoredSnapshot() {\n return storedSnapshot;\n }\n@@ -424,7 +439,7 @@ public abstract class AbstractPersistentCollection implements Serializable, Pers\n operationQueue = new ArrayList( 10 );\n }\n operationQueue.add( operation );\n- dirty = true; //needed so that we remove this collection from the second-level cache\n+ dirty(); //needed so that we remove this collection from the second-level cache\n }\n \n /**\ndiff --git a/hibernate-core/src/main/java/org/hibernate/engine/loading/internal/CollectionLoadContext.java b/hibernate-core/src/main/java/org/hibernate/engine/loading/internal/CollectionLoadContext.java\nindex 72a5e68..83596f2 100644\n--- a/hibernate-core/src/main/java/org/hibernate/engine/loading/internal/CollectionLoadContext.java\n+++ b/hibernate-core/src/main/java/org/hibernate/engine/loading/internal/CollectionLoadContext.java\n@@ -38,6 +38,7 @@ import org.hibernate.EntityMode;\n import org.hibernate.HibernateException;\n import org.hibernate.cache.spi.CacheKey;\n import org.hibernate.cache.spi.entry.CollectionCacheEntry;\n+import org.hibernate.collection.internal.AbstractPersistentCollection;\n import org.hibernate.collection.spi.PersistentCollection;\n import org.hibernate.engine.spi.CollectionEntry;\n import org.hibernate.engine.spi.CollectionKey;\n@@ -258,7 +259,15 @@ public class CollectionLoadContext {\n persister.hasCache() && // and the role has a cache\n session.getCacheMode().isPutEnabled() &&\n !ce.isDoremove(); // and this is not a forced initialization during flush\n- if ( addToCache ) {\n+ \n+ boolean isModified = false;\n+ PersistentCollection pc = lce.getCollection();\n+ if (pc instanceof AbstractPersistentCollection)\n+ {\n+ isModified = ((AbstractPersistentCollection)pc).isModified();\n+ }\n+ \n+ if ( addToCache && !isModified) {\n addCollectionToCache( lce, persister );\n }\n \n\n\n I don't think that the Hibernate team is monitoring this forum, please try using the Hibernate forum instead (see http://hibernate.org/ for forum link and mailing list info). You can link to this message if you like. If you create a Hibernate jira for this issue, please attach your patch to that (should include a unit test similar to the other second level cache tests).
Scott
Hi,
I need help with calling a web service. I have a staeless bean: MyWebService and his JNDI_NAME is \"ejb/myproject_MyWebServiceLocal\"
I want to call this web service and I use:
call.setTargetEndpointAddress(paramSoapUrl), where
paramSoapUrl = \"https://localhost:443/axis/services/MyWebService\"
I get an exception:
javax.naming.NameNotFoundException: MyWebServiceLocal not bound
at org.apache.axis.message.SOAPFaultBuilder.createFault(SOAPFaultBuilder.java:260)
In Browser there is no problem, I can call it using the url:
https://localhost:443/axis/services/MyWebService
Thanks.
You probabaly need to go to the Apache Axis support forum. Here we discuss J2EE compliant web services with JBossWS
I created a singleton service based on the quickstart example (). For a single cluster group, things work perfectly. i.e., only one cluster node runs the service and when the same goes down, another cluster node takes over. We observe that in the case of split brain, while the groups splits into two, two singleton services run (this is expected). However, if the groups merge, we expected one of the singleton services to stop which wasn't working as expected. Rarely did one of the singleton service stop. In simple terms, we had two singleton services running in a merged setup. Is this a known issue? if yes, do we have a workaround.
PS: We used the default standalone-full-ha profile and we didn't modify the cache properties.
That shouldn't be. Upon merge, a singleton master re-election should take place, and any master should stop.
The workaround for this is to preempt the multiple master on split by requiring a quorum when building your singleton service.
e.g.
wildfly/MyServiceActivator.java at master · wildfly/wildfly · GitHub
According to this site http://download.jboss.org/jbosstools/updates/stable/helios/,
in eclipse, I have this site added for the nightly builds:
https://www.jboss.org/tools/download/nightly.html
or
http://www.jboss.org/tools/download/nightly.html
However, I cannot update my plugins anymore.
I get this error:
Error reading update site http://www.jboss.org/tools/download/nightly.html.
The XML stream is not a valid default \"site.xml\" file. The root tag is not site.
In the past, this link worked. Is there a newer one ? I cannot find it.
I am trying to use the following login form to send the authentication request to the JBoss container.Can anybody let me know how the JAASSecurityManager intercept the J_SECURITY_CHECK action, and send the correcpoding j_username, j_password to the Subject, Principle. or others.
<!DOCTYPE HTML PUBliC \"-//W3C/DTD HTML 4.0 Transitional//EN\">
<META HTTP-EQUIV = \"Pragma\" CONTENT=\"no-cache\">
Form Login Page
<h2>Sample Form Login</h2>
<FORM METHOD=POST ACTION=\"j_security_check\">
Please Enter user ID and password:
User ID
Password
And then click this button:
Hi
I have a configuration in which my application servers creates MDB(S) as consumers to topics in remote JbossMQ server. In addition there is security configuration, while a special MDB is responsible for handling security messages and logic. The problem is when losing connection to remote JMS server: once the server is back the local MDB trying to reconnect but fails due to the fact that the MDB responsible for the authentication is not ready (was not reconnected yet). I tried to configure dependency between all of the MDBs and the special MDB responsible for the authentication, but it doesn?t works as well, since that in the deployment part I?m getting an error says that MDB depends on other MDB was not installed yet.
Please advice how to solve the problem.
10x
Arie
Although getting help here is fairly rare these days ;-) you should post your server version, configuration files, and relevant error messages.
I assume you're using in your jboss.xml ?
using ...
< depends / >in your jboss.xml
Hi, 10x for the response
In general I've managed without dependencies, though for general knowledge I used < depends / >, the problem is that both the depend and the depended MDBs are in the same configuration files (jboss.xml and ejb-jar.xml) so while deploying I get messages saying that the depended MDB is not install yet.
It would be great to get response how to solve this issue for further possible needs.
Arie
I'd bet that there was a bug in the MBeans created by the EJB deployer, which I've seen dealing with MDBs and \"depends\". (but only on redeploy.) I honestly don't know what it is, but it has never worked right since 3.2.
If you can make things work without \"depends\", then that might be good enough. If you just need it to work, there is a way to have JBoss deploy modules in a certain order:
http://www.jboss.org/wiki/Wiki.jsp?page=URLComparator
Again, explain your module layout and messages you see.
Hi,
I'm evaluating this promising product, and try to get the sources.
I've been following these instructions:
http://www.jboss.org/jbossjbpm/gpd_sources/
I can see the folder structure, but no files inside it.
I've tried this folder:
HEAD/jbpm.3/designer/jpdl/org.jbpm.gd.jpdl
Of course, I've used anonymous access.
Is there any known problem?
I haven't seen anywhere else to get the source code.
Any news for this?
Have the repository been changed like BPEL from CVS to SVN?
You should check out the branch branch_jbpm_jpdl_gpd_3_1_x to have the latest sources of the 3.1.x branch.
As for the new GPD 4, the sources are in the jboss tools svn. I'll write up a wiki article that describes how to get/build it.
Cheers,
Koen
Thanks. :-)
Hi, i m behind firewall and trying to connect EJB thru. applet client, but i am getting following error
---------------------------------------------------------------------------
java.rmi.ConnectException: Connection refused to host: 200.133.130.100; nested exception is:
java.net.ConnectException: Connection refused: connect
at sun.rmi.transport.tcp.TCPEndpoint.newSocket(Unknown Source)
at sun.rmi.transport.tcp.TCPChannel.createConnection(Unknown Source)
at sun.rmi.transport.tcp.TCPChannel.newConnection(Unknown Source)
at sun.rmi.server.UnicastRef.invoke(Unknown Source)
at org.jboss.invocation.jrmp.server.JRMPInvoker_Stub.invoke(Unknown Source)
at org.jboss.invocation.jrmp.interfaces.JRMPInvokerProxy.invoke(JRMPInvokerProxy.java:135)
at org.jboss.invocation.InvokerInterceptor.invoke(InvokerInterceptor.java:96)
at org.jboss.proxy.TransactionInterceptor.invoke(TransactionInterceptor.java:46)
at org.jboss.proxy.SecurityInterceptor.invoke(SecurityInterceptor.java:45)
at org.jboss.proxy.ejb.HomeInterceptor.invoke(HomeInterceptor.java:173)
at org.jboss.proxy.ClientContainer.invoke(ClientContainer.java:85)
at $Proxy1.create(Unknown Source)
at testclient.ServerManager.lookup(ServerManager.java:82)
at testclient.ServerManager.(ServerManager.java:55)
at testclient.ServerManagerFactory.getServerManager(ServerManagerFactory.java:28)
at testclient.gui.TestAppletClient.initialise(TestAppletClient.java:230)
at testclient.gui.TestAppletClient.init(TestAppletClient.java:165)
at sun.applet.AppletPanel.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)
Caused by: java.net.ConnectException: Connection refused: connect
at sun.rmi.transport.proxy.RMIMasterSocketFactory.checkConnector(Unknown Source)
at sun.rmi.transport.proxy.RMIMasterSocketFactory.createSocket(Unknown Source)
... 19 more
-------------------------------------------------------------------------
Can any one help me in this regard please?
Thanks
Parag
Parag,
EJB containers don't speak HTTP. When you say HTTP, do you mean internet? Connecting to an EJB requires ports that most firewalls do not allow. (I don't remember which port numbers; its documented somewhere.) You should probably put a servlet between your applet and any EJBs in your app. The servlet can be programmed to pass the resulting objects in a serialized stream, as XML, or as clear text. This allows you to communicate entirely on port 80 and keeps your applet much thinner, because it will not have to load all that RMI stuff.
HTH
Larry
EJB containers don't speak HTTP
Hi all,
I'm running the JBoss2.4.4_Tomcat4.0.1 bundle on Solaris 8. My problem is that JBoss doesn't seem to want to recognize the fact that the jboss-j2ee.jar is in the classpath even though from the print out, I can see that it is loaded from lib/ext:
...snip...
[18:21:24,958,ClassPathExtension] Added library:file:/export/home/openspace/work/development/OpenSpace/Dep
loy/AppServer/jboss/lib/ext/jnet.jar
[18:21:24,959,ClassPathExtension] Added library:file:/export/home/openspace/work/development/OpenSpace/Dep
loy/AppServer/jboss/lib/ext/jboss-j2ee.jar
[18:21:24,960,ClassPathExtension] Added library:file:/export/home/openspace/work/development/OpenSpace/Dep
loy/AppServer/jboss/lib/ext/catalina-service.jar
...snip...
I also looked in the JBoss's server.log and the jar file is in java.class.path too. It deploys the tomcat-test.ear fine so I thought it was my code, but when I explicitly add jboss-j2ee.jar to JBOSS_CLASSPATH in run.sh, it works ( actually, it then complains that it can't find log4j, which has also been loaded from lib/ext, but at least it finds jboss-j2ee.jar). I found another post with the same problem, but no reply:
http://www.mail-archive.com/jboss-user@lists.sourceforge.net/msg12536.html//www.mail-archive.com/jboss-user@lists.sourceforge.net/msg12536.html
My output is below. Does anyone have an idea of what's going on? Thanks very much in advance and for this forum.
abe
[18:21:47,824,J2eeDeployer] Create application smi-adm-ejb.jar
[18:21:47,825,J2eeDeployer] install EJB module smi-adm-ejb.jar
[18:21:47,897,ContainerFactory] Deploying:file:/export/home/openspace/work/development/OpenSpace/Deploy/Ap
pServer/jboss/tmp/deploy/Default/smi-adm-ejb.jar
[18:21:48,180,ContainerFactory] Verifying file:/export/home/openspace/work/development/OpenSpace/Deploy/Ap
pServer/jboss/tmp/deploy/Default/smi-adm-ejb.jar/ejb1004.jar
[18:21:48,349,ContainerFactory] Verify failure
java.lang.NoClassDefFoundError: javax/ejb/SessionBean
at java.lang.ClassLoader.defineClass0(Native Method)
at java.lang.ClassLoader.defineClass(ClassLoader.java:493)
at java.security.SecureClassLoader.defineClass(SecureClassLoader.java:111)
at java.net.URLClassLoader.defineClass(URLClassLoader.java:248)
at java.net.URLClassLoader.access$100(URLClassLoader.java:56)
at java.net.URLClassLoader$1.run(URLClassLoader.java:195)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(URLClassLoader.java:188)
at java.lang.ClassLoader.loadClass(ClassLoader.java:299)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:286)
at java.lang.ClassLoader.loadClass(ClassLoader.java:292)
at java.lang.ClassLoader.loadClass(ClassLoader.java:292)
...and so on
Can you give some details about how you have packaged
this jar.
You shouldn't have to add anything to your classpath.
There is a service at the top of jboss.jcml that adds
everyting in lib/ext to it. This is obviously running
because that is the first part of your post.
There are lots of problems posted about classpaths
that involve confusion between Sun's j2ee.jar and jboss-j2ee.jar.
Make sure you don't have Sun's version in
your classpath or lib/ext of JAVA_HOME/jre/lib/ext
or anywhere else this might cause confusion.
Regards,
Adrian
> Can you give some details about how you have
> packaged
> this jar.
That's right, I did forget to mention that I separated out the EJB code and the business logic using inheritance so that the EJB Remote interface, Home interface, and Bean class do not actually contain any functionality. Ie, the Bean class only has ejbCreate, ejbRemove, ejbActivate, ejbPassivate, and setSessionContext; it inherits all of the business logic from another class packaged in a jar file in the client dir (same goes for the Remote interface: it inherits all the business logic by extending another interface). I explicitly include this client jar file in the JBOSS_CLASSPATH in startup.sh.
I don't think it's the inheritance that's causing this though cause I tried reorganizing so that all the functionality is in EJB and it still gave me the same error.
Besides the Home, Remote, and Bean classes i just have my META-INF/jboss.xml and META-INF/ejb-jar.xml in the jar file.
> Make sure you don't have Sun's version in
> your classpath or lib/ext of JAVA_HOME/jre/lib/ext
> or anywhere else this might cause confusion.
Nope, I just the jboss-j2ee.jar
abe
Have you tried putting client.jar
in lib/ext
another alternative is to create a .ear
Don't use JBOSS_CLASSPATH.
I saw a post some time ago that said it was called
this to show JBoss didn't use the CLASSPATH
environment variable, although I've also seen posts
where people say dump whatever you like in there :-)
Everyting in lib/ext is easy and doesn't require
any editing of scripts.
It also stops confusion between the class loaders
like your log4j problem.
Regards,
Adrian
> Have you tried putting client.jar
> in lib/ext
> another alternative is to create a .ear
>
> Don't use JBOSS_CLASSPATH.
Yep...it looks like JBOSS_CLASSPATH is pretty sensitive to changes I've made and that's confusing the class loaders. I put the jar files in lib/ext, used the original JBOSS_CLASSPATH and it deployed fine. I was hoping to put folders in lib/ext though but that could be getting nitpicky :)
Thanks for the help.
abe
>
> Everyting in lib/ext is easy and doesn't require
> any editing of scripts.
> It also stops confusion between the class loaders
> like your log4j problem.
>
> Regards,
> Adrian
You can have directories.
Have a look in conf/????/jboss.conf at classPathExtension.
NOTE: Everything is relative to the location of jboss.conf.
If you add
Providing you don't put a .jar or .zip in there
it will add the lib/ext/mydirectory to the classpath.
Putting a .jar or .zip in there, will only add the .jar
or .zip to the classpath.
Make a sentence out of the following words.
it eat and cake your have :-)
Regards,
Adrian
Sweet! Thanks much Adrian...
> You can have directories.
>
> Have a look in conf/????/jboss.conf at
> classPathExtension.
>
> NOTE: Everything is relative to the location of
> jboss.conf.
>
> If you add
>
> ARCHIVE=\"jboss.jar\" CODEBASE=\"../../lib/ext\">
> VALUE=\"../../lib/ext/mydirectory/\">
>
>
> Providing you don't put a .jar or .zip in there
> it will add the lib/ext/mydirectory to the
> classpath.
>
> Putting a .jar or .zip in there, will only add the
> .jar
> or .zip to the classpath.
>
> Make a sentence out of the following words.
> it eat and cake your have :-)
>
> Regards,
> Adrian
Hi Guys
I'm a little confused by the user guide, it said that
autoDataGravitation - Whether data gravitation occurs for every cache miss. By default this is set to false to prevent unnecessary network calls. Most use cases will know when it may need to gravitate data and will pass in an Option to enable data gravitation on a per-invocation basis. If autoDataGravitation is true this Option is unnecessary.
I've been reading through the SEAM documentation to try and gain a better understanding of what level of reliability SEAM provides. What I've gathered is that SEAM doesn't provide any reliability features itself, but rather depends on the container to provide replication of the different components. (This may be an incorrect statement, but I haven't found anything in the documentation that states otherwise).
So my question is: yes, it is important to have replication of components for reliability, but what about replication of the context cache? SEAM keeps a cache of all the different users' contexts. If a JBoss instance goes down, do all of those user contexts that were in memory on that instance get lost? Do all of my users have to login, create new shopping carts, start new business process flows, etc.?
Maybe I'm missing something?
Seam is built to take advantage of the features of the underlying Java EE implementation. For example, the conversation context data is actually kept in the HttpSession when server-side conversations are used.
So no, there should be no loss of data when a server fails ;-)
Wonderful!
I was just putting together a small clustered environment to test this first hand. I think I'll go ahead and prove it to myself anyways. :)
Nice, let us know if you have problems.
Hello,
I have an AWS (T2-small) Instance that is running Windows Server2008R2. I have installed JBoss 7.1.1 (from jboss-as-7.1.1.Final.zip) and am able to launch it with \"standalone.bat\". In this case I am able to access the server with a local browser @ http://localhost:8080 and http://localhost:9990.
However, if I try to launch the server for remote access with \"standalone.bat -b=PublicIPAddressOfTheAWSInstance\" an attempt to reach to reach the server from a remote system using http://PublicIPAddressOfTheAWSInstance:8080 results in a \"Timeout\" and we don't connect. Further, using a local browser:
http://localhost:8080 still connects
http://PublicIPAddressOfTheAWSInstance:8080 does not connect
http://localhost:9990 no longer connects
http://PublicIPAddressOfTheAWSInstance:9990 does not connect
What may be an important hint is that I see the following error in the console log:
...
11:48:54,459 INFO [org.jboss.as.mail.extension] (MSC service thread 1-1) JBAS015400: Bound mail session [java:jboss/mail/Default]
11:48:54,475 INFO [org.jboss.as.server.deployment.scanner] (MSC service thread1-1) JBAS015012: Started FileSystemDeploymentService for directory C:\\jboss-as-7
.1.1.Final\\standalone\\deployments11:49:00,403 INFO [org.jboss.as.controller] (Controller Boot Thread) JBAS014774
: Service status reportJBAS014777: Services which failed to start: service jboss.network.public: org.jboss.msc.service.StartException in service jboss.network.public: JBAS015810: failed to resolve interface public
11:49:00,419 INFO [org.jboss.as] (Controller Boot Thread) JBAS015954: Admin console is not enabled
11:49:00,419 ERROR [org.jboss.as] (Controller Boot Thread) JBAS015875: JBoss AS7.1.1.Final \"Brontes\" started (with errors) in 9142ms - Started 119 of 214 services (44 services failed or missing dependencies, 49 services are passive or on-demand)
>>> Note the error: JBAS015810: failed to resolve interface public.
Additional info: The AWS Instance is deployed in the default VPC and an AWS Security Group is defined to permit access to PublicIPAddressOfTheAWSInstance from all IP addresses using TCP on 8080. Also, the Windows Firewall is disabled.
I am guessing that \"interface public\" failing to resolve is related to why I can't get remote access.
Note that if I launch with \"standalone.bat -b=0.0.0.0\" EVERYTHING WORKS! There is no error in the log and remote access is available. However, using 0.0.0.0 is not advisable for production.
Can anyone help me understand why this is failing if I specify the AWS Public address?
Thank you.
Hi,
I have deployed an ear application in JBoss3.0.4 (ejb + war modules).
When i tried to access my web application, i encounter NoClassDefFoundError: org/apache/struts/util/MessageResources
The struts.jar is located at the web module (WEB-INF/lib/struts.jar).
I noticed that the jboss console indicates loggings of the struts action forwardings when i access my web application. It shows that it has no problems traversing from 1 action to another action (as per what i have defined in my struts-config.xml)
Since the struts.jar has successfully loaded and it has no problems in traversing from 1 action class to another action class (before the no class found exception occurs), it really puzzles me that NoClassDefFoundError: org/apache/struts/util/MessageResources is encountered at this stage ...
Can someone help me to resolve this cos i'm really at my wit's end ....
Thanks,
- SY
look at the posting I made on classloader problems with JetSpeed just now - your problem is probably the same thing.
Jules
I am trying to create a custom workItemHandler and have been reading posts for the better part of a week. I have
the workItemHandler showing up in my KIE interface but every time I deploy and run my process, JBPM reports that
it can't find my custom workItemHandler.
I added the session.template file in my /standalone/configuration/ directory but the file does not seem to be being read.
I proved this to myself by adding the word \"blart\" to the start of the definition and there were no errors in the server.log
I found another post that talked about exploding the jbpm.console war and I have tried that as well... I moved my jar containing my customer workItemHandler to
the \"../standalone/deployments/jbpm-gwt-console-sver.war/WEB-INF/classes/META-INF/\" directory and still no joy.
Question1: Is there something else that has to be done to have jbpm read the session.template file?
I have read Chapter12 in the documentation several times and I don't see the answer.
Question2: Is there some other source of information that provides a step by step example for custom workItemHandlers?
I checked my boot.log and all of the directories seem to be set properly.
Hi there, which version of jBPM are you using?
Maybe this post will help you with the configurations: http://salaboy.com/2013/10/22/kie-wb-jbpm-console-ng-configurations/
Thanks for the reply Mauricio,
I am using JBPM6. I read the article/link but I think my problem is more basic. I found a post last night where the author created a process with a custom service and an associated WorkItemHandler and
everything was contained in the Eclipse project. I downloaded his code and the project worked for me. That would imply that I don't need to explode the console WAR and place my jar files
in the lib directory.
I suspect that I need a basic tutorial on how to create a custom service with an associated WorkItemHandler that can be deployed to JBOSS. The example that I found last night
worked in Eclipse, but I could not figure out what needed to be done to deploy it so that it works from inside KIE.
I have managed to checkout projects that I create in KIE using GIT, but when I make changes and try to submit them back the GIT repo is not updated. That's probably a topic for
another thread.
THE BOTTOM LINE: I am looking for a tutorial on how to create and deploy a custom service with an associated WorkItemHandler for use inside KIE.
for 6.0 and 6.0.1 you still need to place your handler class/jar inside jbpm-console.war/WEB-INF/lib so it can be registered properly. Then the definition of the handler name and class name should be either specified in kmodule.xml or CustomWorkItemHandlers.conf. See docs of RuntimeManager configuration on various options to define handlers. This applies only to version 6 as in 5 it was different mechanism for that.
HTH
ANSWER: The problem turned out to be related to the copious amounts of misleading documentation on the Internet. In the end the solution was to edit the WEB-INF/classes/META-INF/CustomWorkHandlers.conf file and add my tags and the associated classes. I tried adding definitions to the kmodule.xml file, session.template files and nothing worked. I'm not sure why this was so hard to find (8 days total), but all is working now.
Hi,
Does richfaces supports validation for schema objects created by CXF?
Like it supports for Hibernate annotations with the following:
CXF generate java class would have attributes
@XmlElement(name = \"Attribute\", required = true)
protected List<AttributeType> attribute;
Can someone suggest me a best approach for validating the input fields which is directly mapped to Schema/Client Objects generated by CXF for given WSDL?
Thanks,
Shabeer.
Hi ,
I have been reading again many of the topics about BPEL vs JBPM, differences, etc since I was too new to understand.
Let me know if i am wrong.
I understand the difference btw web service orchestration (BPEL) and java components (JBPM).
The problem is that I want to get web service orchestration. Imagine a process where several web services are called (think of the loan approval example) and at some point I want a human to accept or refuse my loan request.
My first question is (): is there any library to manage this case using jbpel ?Is this case out of the scope of BPEL ? do I have to use jbpm then ?
In my first question I am thinking of an easy way or automatic way to do such a thing.
My second question, idea is : Since BPEL process is only aware of web services is it possible to create (code) a web service that can be called by the process and by another user so that the process response will be release after the user called to the same web services ??
Any other suggestion to deploy such an example ??
thanks
I ma having a heck of a time doing what should be the most simple thing - getting SwitchYard to install in Eclipse.
I am following instructions from the SwitchYard installation page:
Home - SwitchYard - Project Documentation Editor
When i attempt the install from this site:
http://download.jboss.org/jbosstools/updates/development/kepler/integration-stack/aggregate/4.1.1/
I get the option to install these tools:
SOA Development
JBoss ESB Tools 1.5.200.Final-v20130513-1817-B83
JBoss Runtime ESB Detector 1.5.200.Final-v20130513-1817-B83
SAVARA Tools Feature 2.1.2.Final
SwitchYard 1.0.0.v20130610-2034-H331-CI
SwitchYard BPEL Support 1.0.0.v20130610-2034-H331-CI
SwitchYard BPM Support 1.0.0.v20130610-2034-H331-CI
I select the root (SOA Development) and the process fails with the generic message \"Cannot perform operation. Calculating Alternatives\"
installing the alternatives also fails.
Any help is appreciated.
Hey Patrick,
Which version of Eclipse are you using?
The interesting thing is that the versions you list above do not align with the versions listed in the 4.1.1 link you posted (1.0.0.v20130610-2034-H331-CI vs. 1.0.0.v20130722-1948-H360-CI).
If you could provide more details regarding which features are being installed (including dependent features), which features cannot be installed (e.g.features/plugins with missing/incompatible dependencies), that would be helpful.
Rob
I am fairly new to JBoss so apologies if my responses are not as coherent as one might hope. I am following instructions verbatim from the switchyard installation guide per the link in the OP.
My first mistake was using Eclipse rather than Eclipse J2EE (Kepler). Fixed that but still having problems:
Software being installed: SwitchYard 1.0.0.v20130722-1948-H360-CI (org.switchyard.tools.feature.feature.group 1.0.0.v20130722-1948-H360-CI)
Missing requirement: SwitchYard 1.0.0.v20130722-1948-H360-CI (org.switchyard.tools.feature.feature.group 1.0.0.v20130722-1948-H360-CI) requires 'org.jboss.ide.eclipse.freemarker.feature.feature.group 1.2.0' but it could not be found
At least that is a clear error with a recommended course of action: I have to install org.jboss.ide.eclipse.freemarker.feature.feature.group 1.2.0
Looking for and not finding installation instructions for freemarker. Any help is appreciated.
Got freemarker from
http://download.jboss.org/jbosstools/updates/development/kepler
Switchyard install is progressing now as it seems the final dependency has been met.
Hey Patrick,
There are few things you can do: (I'd recommend the last option one of the Beta URLs)
It appears the site you are using does not include a reference to the core JBT site, which is why it can't find that dependency. The other sites have been restructured and include a reference to the core JBT site, so no additional configuration is needed. I'll update the documentation appropriately.
Sorry you had to experience this. The JBTIS project has been a bit of a learning experience and has evolved pretty significantly over the past year. It's goal is to provide a single site from which JBoss integration tooling can easily be installed. Obviously, that goal was not met with the 4.1.1 site. Hopefully your experience will be better with one of the beta sites. If not, please let me know. (As an aside, the earlier sites also polluted the list of configured available sites with a bunch of extra sites. The latest version should only add sites which are used to resolve required dependencies, specifically, JBT core tooling and JBTIS \"target platform\" which are provides dependencies not provided as part of core JBT or its target platform.)
Hope that helps.
Rob
I've edited my previous comment. I recommend using one of the \"beta\" URLs as they are more mature. Also, the category names have changed from version to version and different categories have different versions of the SwitchYard tools. The latest name appears to be \"JBoss Integration and SOA Development.\"
All of these kinks should be worked out by the time a \"stable\" release is available (i.e. http://download.jboss.org/jbosstools/updates/stable/kepler/integration-stack/aggregate/)
Once again, sorry this was much more difficult than it should have been.
Best,
Rob
We NEED a way to moderate these forums.
The amount of drivel http://www.thefreedictionary.com/drivel
I read everyday is ridiculous.
As soon as we convert these forums over to JBoss Portal, I'm going to re-write
the moderation function myself, even though I hate GUI programming. :-)
Buttons to move posts to:
* The beginner's forum
* An RTFM didn't read the FAQ forum.
* A \"your question is unanswerable\" forum
* An un(der)educated comment forum
* Don't post user questions in the development section forum
And
* Delete the user's post and send them an Email to stop ME TOO/CROSS POST/HIJACK
By way of an example, I'm going to \"answer\" this morning's posts in the JCA
forum.
I'm not going to actually answer the question since most of them are unanswerable
http://wiki.jboss.org/wiki/Wiki.jsp?page=BadPosts
or have been answered so many times before it is ridiculous.
I'm going to answer them to show why they can't/won't be answered.
The sample posts are:
http://www.jboss.org/index.html?module=bb&op=viewtopic&t=67445
http://www.jboss.org/index.html?module=bb&op=viewtopic&t=68015
http://www.jboss.org/index.html?module=bb&op=viewtopic&t=68026
http://www.jboss.org/index.html?module=bb&op=viewtopic&t=68042
And here's another one for good measure;
http://www.jboss.org/index.html?module=bb&op=viewtopic&t=68007
So that took *one hour* to answer five questions and provide all the links
that the users couldn't be bothered searching for themselves.
With the exception of the first question, we aren't any closer to answering these
questions, but hopefully this is five users that will provide more useful information in future.
And even do some of the work for themselves?... :-)
http://wiki.jboss.org/wiki/Wiki.jsp?page=BadPostNoWork
Here we go again.
A group of users who think this is the IBM AIX JDK Support forum:
http://www.jboss.com/index.html?module=bb&op=viewtopic&t=72470
I've given up even trying to moderate to the forums, and just created a bad post forum
instead :-)
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3911887#3911887
Now if I could figure out how to exlude that from the search function? :-)
You really are one arrogant prick aren't you?
Arrogant, No
Frustrated, Yes
Incidently, I no longer read these forums (well maybe once a month looking for
potential bug reports). I would be bald from pulling my hair out otherwise. :-)
Maybe when we move to the JBoss Portal version of the forums with
a reasonable system of moderation (that removes these useless posts from the
search index so it becomes usable again), I'll go back to contributing regularly.
On a related story, I had a problem with installing MySQL back in January,
but I couldn't find the answer to the problem, just \"a million\" posts mostly unanswered,
with a few saying this is an FAQ or use search.
I'd like to thank all those time wasters that obscured the post where it was
really answered (I couldn't find it),
because it made me go through the MySQL source to figure
out what the problem was for myself, so I learnt something.
In fact, it was a none problem, at least for me, just an irrelevant warning. :-)
This just shows the problem is a generic one, not just these forums.
In general, people that have real problems suffer because of the laziness of a
minority.
Adrian, you are the lizzard master :-)
Adrian, I would like to give you a user's perspective on this issue.
I do not know if there is such a thing as a typical user, or even if that has any meaning. However, for the purposes of this argument, consider that I am it. I managed to get the JBoss advanced developer's certificate, so I guess that means that I am not completely stupid, and that I do have a certain understanding of JBoss. On the other hand I am certainly not brilliant enough ever to be a JBoss core developer. I do not think that I will ever understand all the intricacies of JBoss Class Loading. (Well, maybe if I devoted myself exclusively to the subject for a month. unfortunately a month I do not have). I love JBoss because it does give you some chance of understanding what is going on under the hood, even though, given the constraints of time and my intellectual capacity, I know that it will always be limited. Unfortunately I need eight hour's sleep a night, I work on projects with time constraints, and I do have a life outside JBoss.
People like me use JBoss, and act as champions for JBoss in our different companies. I guess you could even say, without exaggerating too much, that we provide your bread and butter.
I guess the seminal article on Forum questions is that of Eric Raymond
http://www.catb.org/~esr/faqs/smart-questions.html
. When I ask a question in the forums I try to make it intelligent. Try but obviously fail. I have the distinction of never having had a question answered. My latest effort is a good example: http://www.jboss.org/index.html?module=bb&op=viewtopic&t=107384 . With 22 views it remains gloriously unanswered. However I thought I had done all my homework. I went right through the PojoCache User Documentation. I went through the FAQ. I went through the examples and the tests. I tried to intelligently trace the classloaders concerned, and give sufficient information without giving too much. However there was one document I had overlooked: the TreeCache User Documentation. Silly me! This was definitely a stupid question. No, worse than that: a really idiotic question. I definitely deserve roasting over the burning embers of a TreeCache for at least a week, while assorted JBoss gurus gleefully stoke the flames with odd bits of PojoCache to keep the heat up. But, really, would it have been too difficult for someone to briefly point me in the right direction?
I think also, looking through the forums, that other people not quite so stupid as me feel the same sense of frustration. I take as an example this: http://www.jboss.org/index.html?module=bb&op=viewtopic&t=99432 . I know this entry because I hit the same problem. The question was answered by Ben Wang, so obviously it was an intelligent question. However note how it starts:
I have read all of the documentation I can find on the JBoss AOP AND the JBoss Cache, and even looked through the JBoss Cache public forum from beginning to end, drilling into ANY of the cases whose subject might remotely indicate a potential solution to this. If there is any place my question indicates I have not looked, please do point me in that direction.
Often the problem isn't that the user doesn't know he's posting incomplete questions or posting them in the wrong place but that the user doesn't care.
This may be less prevalent here than in other forums (like the Sun forums for example).
Combine that with an attitude of expecting instant gratification without having to put in any effort of their own, and you have a recipe for disaster.
This has in part to do with education (or the lack thereof that nowadays is called education), in part with cultures that raise children in an atmosphere where their every wish is your command, noone dares say no to them or punish them for transgressions for fear of being termed a bad parent or worse a child molester.
Together they create a user (or more generally a person) who doesn't know how to use the tools he has at his disposal, and doesn't care to figure out how to use them, instead expects others to instantly drop what they're doing and provide service free of charge without even being told what the problem is.
Comments
\n\n- \n \n 0 Comments\n \n
\n\n
\n \n\n \n\n \n\n \n