Skip to main content

EJB called from another, specifically Message driven bean with container transaction.

The other day I had a weird defect in the application I worked on, we had a Message driven bean [MDB] calling another EJB inside the message processing code and everything was working fine and passed two different testing periods.

after going live on production something happened, the MDB received a message and invoked the EJB to do some work, that work included calling a web service, this service client thrown time out exception and thrown an exception and the EJB couldn't finish its work because that's what is expected, the weird thing is that the exact same message was reprocessed by the MDB automatically.. we don't have anything in our code that did that.

this didn't happen before with other EJBs called from an MDB..

after thorough investigation I found the following :

1- The MDB uses container transaction [ found configuration in the ejb-jar.xml] (this is used in EJB2)
2-  The EJB method that was invoked also had container transaction configured with transaction attribute as Required
3- When an exception is thrown from the EJB method, it calls : context.setRollBackOnly(); which means the container transaction will roll back

but why was the message reprocessed again? Luckily in production the service didn't time out the second time and the message was processed successfully, but I did a test where the EJB always fails and that led in having the MDB reprocessing the message indefinitely until I shutdown the server.

I tried and changed the EJB transaction attribute in the ejb-jar.xml to 'RequiresNew' instead of 'Required' because at that point I felt that there is a relation between the transaction failing in the EJB and the MDB transaction [Yes MDBs have transactions] and that WORKED.

it's critical to configure the EJBs [ MDB is an EJB ] transactions to match the need of the application, in our case we didn't want the message to be reprocessed everytime it failed, so a new transaction in the EJB did the trick


Popular posts from this blog

Spring 4 + Hibernate 4 / Java configuration / rest service example with gradle

In this post I'll explain the required work to create a rest API utilizing both spring and hibernate version 4, and the configuration will be using java configuration classes not XML. I'll use gradle to build and for dependency management, it's way easier than maven and keeps you focused on the application, if you are not familiar with gradle and interested in it see my previous post about it. The first part which is dependency management is covered in gradle post mentioned above. I'll skip to explain each tier of the project and its configurations: As you can see we have 4 tiers: 1) DAO tier / data tier In this tier we configure the datasource and hibernate, I used HSQL in memory db it can be easily substituted with other db engine providing the right dependencies The DaoConfig defines the data source, transaction manager, session factory and  hibernate properties The most important part is the annotations : 1)  @Configuration  : to tell spring th

Android RecyclerView - Adding Empty View

So RecyclerView was introduced to replace List view and it's optimized to reuse existing views and so it's faster and more efficient as stated in the documentation: While using it, I faced the issue of missing a useful functionality that is implemented in ListView. that feature is setting an empty view in case there was no records. In ListView it was as simple as this View emptyView = findViewById(; ListView  myList = .... myList.setEmptyView(emptyView); but this method doesn't exist for recycler view so we need a work around until android team fixes this. and here are the screen shots of both list view and recycler view fix List view : Recycler view : here is how I fixed it: here is the content of empty_view, it can be anything. Enjoy.

[PART 5] NuTracker ReactJS app - Add Login & Profile using Router

In the previous part we finished the dashboard read functionality, now we want to add the skeleton for other pages: - Login   In this page the user will be able to login to their account and the dashboard won't show unless the user is logged in. - Profile In this page the user will be able to update their daily nutrition goals that they can track in the dashboard. to be able to have multiple 'pages' in react and navigate from one to one, we need something that can switch the rendered content based on what we want, we can do that with if statements in the App components and store some location state, but why invent the wheel. React Router every major single page app web framework has the routing concept and functionality to interact with the usual browser urls and switch the content based what user should see. for example on the profile page I want the url path to be /profile, and for login to be /login and so on. in more advanced cases you want the users