The process Bpel of JavaScript call

Similar Questions

• error in the process bpel deployiong please help

  I get an error while the process bpel deployiong please help
  
  
  BUILD FAILED
  E:\jdevstudio10132\jdev\mywork\Application4\vinayread\build.XML:79: There was a problem connecting to the server "chdsez116553d" using the "8888" port: bpel_vinayread_1.0.jar failed to deploy. Exception message is: ORABPEL-05215
  Error loading process.
  The field of process encountered the following errors when loading the "vinayread" (version '1.0') process: BPEL validation failed.
  Source BPEL validation failed, the errors are:
  [Error ORABPEL-10902]: compilation failed
  [Description]: in "bpel.xml" XML parsing failed because "undefined element.
  In WSDL to "file:/D:/product/10.1.3.1/OracleAS_8/bpel/domains/default/tmp/.bpel_vinayread_1.0_679323b0585449e9fd1887e6ee2bf444.tmp/vinayred.wsdl", part of message {http://TargetNamespace.com/vinayred} 'root element' element is not defined in one of the schemas.
  Please make sure the spelling of the element QName is correct and the WSDL import is complete.
  ".
  [Potential fix]: n/a.
  .
  .
  If you installed a hotfix on the server, verify that the ownership of the bpelcClasspath domain includes patch classes.
  at com.collaxa.cube.engine.deployment.CubeProcessHolder.bind(CubeProcessHolder.java:285)
  at com.collaxa.cube.engine.deployment.DeploymentManager.deployProcess(DeploymentManager.java:862)
  at com.collaxa.cube.engine.deployment.DeploymentManager.deploySuitcase(DeploymentManager.java:728)
  at com.collaxa.cube.ejb.impl.BPELDomainManagerBean.deploySuitcase(BPELDomainManagerBean.java:445)
  at sun.reflect.NativeMethodAccessorImpl.invoke0 (Native Method)
  at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
  at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
  at java.lang.reflect.Method.invoke(Method.java:585)
  at com.evermind.server.ejb.interceptor.joinpoint.EJBJoinPointImpl.invoke(EJBJoinPointImpl.java:35)
  at com.evermind.server.ejb.interceptor.InvocationContextImpl.proceed(InvocationContextImpl.java:119)
  at com.evermind.server.ejb.interceptor.system.DMSInterceptor.invoke(DMSInterceptor.java:52)
  at com.evermind.server.ejb.interceptor.InvocationContextImpl.proceed(InvocationContextImpl.java:119)
  to com.evermind.server.ejb.interceptor.system.JAASInterceptor$ 1.run(JAASInterceptor.java:31)
  at com.evermind.server.ThreadState.runAs(ThreadState.java:646)
  at com.evermind.server.ejb.interceptor.system.JAASInterceptor.invoke(JAASInterceptor.java:34)
  at com.evermind.server.ejb.interceptor.InvocationContextImpl.proceed(InvocationContextImpl.java:119)
  at com.evermind.server.ejb.interceptor.system.TxRequiredInterceptor.invoke(TxRequiredInterceptor.java:50)
  at com.evermind.server.ejb.interceptor.InvocationContextImpl.proceed(InvocationContextImpl.java:119)
  at com.evermind.server.ejb.interceptor.system.DMSInterceptor.invoke(DMSInterceptor.java:52)
  at com.evermind.server.ejb.interceptor.InvocationContextImpl.proceed(InvocationContextImpl.java:119)
  at com.evermind.server.ejb.InvocationContextPool.invoke(InvocationContextPool.java:55)
  at com.evermind.server.ejb.StatelessSessionEJBObject.OC4J_invokeMethod(StatelessSessionEJBObject.java:87)
  at DomainManagerBean_RemoteProxy_4bin6i8.deploySuitcase (unknown Source)
  at com.oracle.bpel.client.BPELDomainHandle.deploySuitcase(BPELDomainHandle.java:319)
  at com.oracle.bpel.client.BPELDomainHandle.deployProcess(BPELDomainHandle.java:341)
  at deployHttpClientProcess. jspService(_deployHttpClientProcess.java:376)
  at com.orionserver.http.OrionHttpJspPage.service(OrionHttpJspPage.java:59)
  at oracle.jsp.runtimev2.JspPageTable.service(JspPageTable.java:462)
  at oracle.jsp.runtimev2.JspServlet.internalService(JspServlet.java:594)
  at oracle.jsp.runtimev2.JspServlet.service(JspServlet.java:518)
  at javax.servlet.http.HttpServlet.service(HttpServlet.java:856)
  at com.evermind.server.http.ResourceFilterChain.doFilter(ResourceFilterChain.java:65)
  to oracle.security.jazn.oc4j.JAZNFilter$ 1.run(JAZNFilter.java:396)
  at java.security.AccessController.doPrivileged (Native Method)
  at javax.security.auth.Subject.doAsPrivileged(Subject.java:517)
  at oracle.security.jazn.oc4j.JAZNFilter.doFilter(JAZNFilter.java:410)
  at com.evermind.server.http.ServletRequestDispatcher.invoke(ServletRequestDispatcher.java:623)
  at com.evermind.server.http.ServletRequestDispatcher.forwardInternal(ServletRequestDispatcher.java:370)
  at com.evermind.server.http.HttpRequestHandler.doProcessRequest(HttpRequestHandler.java:871)
  at com.evermind.server.http.HttpRequestHandler.processRequest(HttpRequestHandler.java:453)
  at com.evermind.server.http.HttpRequestHandler.serveOneRequest(HttpRequestHandler.java:221)
  at com.evermind.server.http.HttpRequestHandler.run(HttpRequestHandler.java:122)
  at com.evermind.server.http.HttpRequestHandler.run(HttpRequestHandler.java:111)
  to oracle.oc4j.network.ServerSocketReadHandler$ SafeRunnable.run (ServerSocketReadHandler.java:260)
  at oracle.oc4j.network.ServerSocketAcceptHandler.procClientSocket(ServerSocketAcceptHandler.java:239)
  to oracle.oc4j.network.ServerSocketAcceptHandler.access$ 700 (ServerSocketAcceptHandler.java:34)
  to oracle.oc4j.network.ServerSocketAcceptHandler$ AcceptHandlerHorse.run (ServerSocketAcceptHandler.java:880)
  to com.evermind.util.ReleasableResourcePooledExecutor$ MyWorker.run (ReleasableResourcePooledExecutor.java:303)
  at java.lang.Thread.run(Thread.java:595)
  
  
  Total time: 1 second
  
  
  
  Please give SME solutions

  Check all the XSD / WSDL / BPEL files in your project, if something is mentioned on "TargetNamespace.com".

  Marc

• Subprocess relatively to the process of hand when called programmatically.

  Hello
  
  JDeveloper 11.1.1.6, WLS 10.3.6, BPM 11.1.1.6
  
  I have 2 steps A and B. A. I want to call B.
  I see 2 approaches to do this.
  
  Approach 1. Call the sub-process B a send and receive activities.
  If we do this way, when an instance of B is created, it is created as a child of the process a. (when he is seen in the EM Console)
  
  Approach 2. Use a Web Service call in the main process. Lets say it's a Java Web Service and we use the following API to call the process.
  IMS IInstanceManagementService is Fixture.getBPMServiceClient () .getInstanceManagementService ();.
  Task task = ims.createProcessInstanceTask (bpmContext, pms.getCompositeDN()+"/"+pms.getProcessName ());
  But I think it will be created in a separate process.
  
  So my Questions are the following.
  
  is a. in approach 2, possible to make the sub-process B as a child of a process?
  
  b. the sub-process B might have to be called several times in parallel, with some settings each time.
  I want to understand what is the best approach?
  Use method A to call her go a subprocess (loop) with parallel, Multi Instance and use table to pass parameters for each sub deal call?
  Or
  Use approach B?
  
  Thanks for any help
  Sameer

  Hi Sameer,

  Your send and receive events would be the better of the two options you listed. If the two processes are in the same composite project and send and receive events are used, there is less overhead to call the sub-process.

  Think you have alluded to this, but you will also keep the audit trail information if you take the first approach, not only in Enterprise Manager, but also in the workspace.

  Dan

• API to disable the process BPEL (State = stop) in Oracle soa suite 10 g

  Hello world
  
  I am looking for a solution disable a BPEL process without using the BPEL console in Oracle SOA suite 10 g. I was thinking about an API, but I can't find in the list below those that would be useful:
  
  http://download-UK.Oracle.com/docs/CD/B31017_01/integrate.1013/b28986/index.html?com/Oracle/BPEL/client/package-summary.html
  
  Any ideas?
  
  It also crossed my mind to use Ant by disabling a BPEL process. Anyone know if there are any available custom Ant task for updating the State of a BPEL process?
  
  I look forward to hearing from you all!
  
  Thank you!!
  
  SR

  Hi SR,

  You can do this by having a BPEL process with a java embedding containing code below.

  /*
   * Input Variables
  */
  String processName   = (String)getVariableData("processName");
  String revision      = (String)getVariableData("revision");              
  
  /*
   * Declare variables.
  */
  IBPELProcessHandle procHandle = null;
  Locator loc = null;
  BPELProcessMetaData bpelProcessMetadata = null;              
  
  try{
      /*
       * get Locator Instance
      */
      loc = getLocator();
      /*
       * Lookup Process. Revision optional.
      */
      if(revision == null || revision.trim().equals(""))
      {
        procHandle = loc.lookupProcess( processName );
      }
      else
      {
        procHandle = loc.lookupProcess(processName, revision);
      }
      if(procHandle == null)
      {
        throw new Exception("Unable to find process: " + processName);
      }
      /*
       * Get Metadata of the process.
      */
      bpelProcessMetadata = procHandle.getMetaData();              
  
      /*
       * Set Lifecycle to Retired.
       * Use setState(IBPELProcessConstants.STATE_OFF) to change process state to off.
       */
      bpelProcessMetadata.setLifecycle(IBPELProcessConstants.LIFECYCLE_RETIRED);
      /*
       * Finally update the process with the modified metadata.
       */
      procHandle.updateMetaData(bpelProcessMetadata);
  }catch (Exception e)
  {
    addAuditTrailEntry(e);
  }
  

  If you wish, you can schedule this process.

  Concerning
  PavanKumar.M

• By the way the values and the process of PL/SQL call via a column binding

  Hi all
  
  I have a report and a link on one of the columns that you take you to another page and passes the data from the report to this page. Everything works well. In addition to picking up the details of the report, that I join the link to call a process page before it goes to the next page by using some of the values picked up in the report. This I can go to work by changing the link to a URL and put in some javascript that is to say javascript:doSubmit('TEST_PROCESS'); By doing this "Lose" the opportunity to pass the info according to the report, via the link.
  
  Can someone give me a clue as to how I can achieve both please? I guess I need to create my own javascript function, but I don't know how to enter the information of a line in a report to send.
  
  Concerning
  Helen

  Helen:

  JavaScript works against the current page's HTML elements. Application parts are NOT HTML elements and can therefore be manipulated directly from Javascript.

  CITY

• After updating to Firefox 15 it crashes at the end of a javascript call window.close (); opener.alert ('test'); Does work as expected or is this a bug?

  After updating to Firefox 15, it crashes after a popup called javascript window.close (); opener.alert ('test'); Does work as expected or is this a bug?

  It works better if you disable the modal prompt temporarily as a test by setting prompts.tab_modal.enabled false on the subject: config page?

• How to give multiple entries in the process BPEL syn in Jdeveloper

  Hello
  
  I'm new to BPEL. I have created programs that takes only one input and after treatment poster single output. I would like to make a contribution of more than one runtime (for the name of the employee e.g., age, salary) and want to view multiple output after their treatment instead of just I am doing now. How can I do this by using synchronous BPEL process. Can you tell what will learn time for BPEL using google and the forums, for a person to 3 years in JAVA and new on BPEL exeperienced.
  
  Yogendra Rishishwar

  Hello

  You just need to change the XSD in your process.

  Someone posted something similar last week: How to change the entry and exit of BPEL process diagram

  Hope this will help you...

  Roman.

• Failed to retrieve the workflow of the process BPEL activity attchements

  Dear Experts,
  
  I am using human activity task and using the copy operation, I am able to send attachments in worklist application. Once again worklist application I am able to join multiple files using + sign on the Attachments tab. Once approved the task assigned to the worlist portal, I'm trying to download / view attachments from BPEL process. I can able to display all the attachments, but I'm not able to recover the content of attachments, because he is missing.
  
  Here in my overall "UploadAttachment_1_globalVariable" variable, we have some
  
  < setting >
  < mimeType > application/msword < / mimeType >
  < name > TDRequition.doc < / name >
  edd36c1b2395e0ab < taskId >:-43c274e7:122a1aa09d5:-7 has 73 < / taskId > / / element content is missing here in the two attachments
  < Version > 2 < / version > / / so how cum able to retrieve the attachment
  < / fixing >
  -< setting >
  < mimeType > application/msword < / mimeType >
  < name > TDInitiation.doc < / name >
  edd36c1b2395e0ab < taskId >:-43c274e7:122a1aa09d5:-7 has 73 < / taskId >
  < version > 2 < / version >
  < / fixing >
  
  Please suggest me.
  
  Thank you
  Rajesh has

  I think that the workload does not display the contents of the task in a variable. You can use workflow service function hwf:getTaskAttachmentContents() and assign the value retrieved to a variable of type 'base64Binary '.

  Hope this will help.

  Cedric

• The process data adapter when DB is down?

  Hi, I'm trying to understand what is happening in an invoke activity that called a DB adapter if the database is out of service. I need to set up some error around the DB handeling are down, but I don't know what condition I should be looking? The process BPEL fails or t - she continue?

  Hello

  The invoke process will be error with the error remotefault.

  This is an unhandled exception. But if you want to have exception manage and want to cancel, and then add the catchBrach in your bezel or main and then add the remoteFault system error and create a faultVar and then perform your operation if not in the scope of the catchBranch.

  Kind regards
  Abdou

• Can I replace reminder in the process template engine?

  I'm confused. I guess I can put the SequenceFilePostStep and SequenceFilePostStepRuntimeError callbacks in the process template I use and so replace this engine callback for my sequence of all the files that I run using this process template. What I actually see is that I should create these reminders of engine in each of my sequence files to make them work. Those of the process template is not called. Stack of calls that I have is - - . So, I would keep my engine in one place callbacks instead of many places .

  Hello

  If you want to maintain engine in one place callbacks instead of many places , use ProcessModelPostStep and ProcessModelPostStepRuntimeError in your sequence of process template file. These fires after each step of the sequence of client. TS reference manual table 10-1, "after the engine executes each step in any client.
  sequence of drop calls to process template and each step in calls to subsequence resulting.

  The equivalent of this would be to use the SequenceFilePostStep and reminders of SequenceFilePostStepRuntimeError in all of your files to the client's sequence. BTW if you want a global error handler you can use ProcessModelPostStepRuntimeError and SequenceFilePostStepRuntimeError in your process template. This document will include all the steps in the sequences of process model and the customer test sequences.

  I think that reminders of engine are turned on and off, as opposed to the alternative.

  CC

• JavaScript is not called by the process of PL/SQL

  Hello Experts
  
  I try to call a javascript function from pl/sql using htp.p process, but javascript does not get mentioned. In fact, my requirement is as follows:-
  
  I have a report with check boxes. I use apex_application.g_X01.count to loop through the boxes and check the values of the report. If all of the checkboxes are selected, then I want to show a confirmation box so that the user can confirm that they want to go ahead and do the follow-up actions.
  
  My Javascript function that I placed it in the HTML header and the Body attribute is:-
  
  function confirm_response()
  {
  var answer = confirm ("test", "Confirm");
  If (response)
  {
  $x('P1_FLAG').value = 'Y ';
  }
  
  }
  
  Looks like my pl/sql process, which currently is trying to evoke the javascript: -.
  
  Start
  IF apex_application.g_f01.count = < < a number > > THEN
  HTP. P ("< script type =" text/javascript"> ');
  HTP. P ('confirm_response()');
  HTP. P ("< /script >" ");
  IF: P1_FLAG = 'Y' THEN
  -Continue with follow-up actions
  ON THE OTHER
  -To get out the process of
  END IF;
  END IF;
  
  end;
  
  But when I run the present that it does not work. The confirmation box never appears.
  
  It does not even if I change the PL/SQL process as below: -.
  
  
  Start
  HTP. P ("< script type =" text/javascript"> ');
  HTP. P ('confirm_response()');
  HTP. P ("< /script >" ");
  end;
  
  Can someone please advice.
  
  Thank you

  Hello
  >
  It does not meet my needs because I want first of all check if all the boxes are ticked, then only run javascript.
  >

  So how count the boxes checked in JS is the question.

  Continuing to your operation code snippet, you have box f01.

  Modify the script in the HTML header to this

  See you soon,.

• A correlation must be established at the level of the process calling or called?

  In the Oracle BPM Suite 11 g: advanced topics of BPMN book, the authors wrote:

  When you use the message-based correlation, don't the called process must be aware of the correlation, the appeal process. The runtime will take care of the details for the calling process, you don't need to include details of correlation in the process model for the calling process.

  In the called process, you must include the definition of correlation set and specify the tasks or events appropriate to use correlation. Let's look at an example in the following diagram:

  In the example above, the authors introduce a correlation on the called process.

  

  However, documentation of Oracle on the correlation, the typical design for correlation flow mentions that the correlation is initiated by the sender of the message, as shown below.

  1 design processes that communicate with each other.

  2. in the calling process, add the stream object that sends a message to the other process.

  3. define a correlation to the stream object that sends the message and configure it to initiate the property aliases.

  4. in the called procedure, add the stream object that receives the message.

  5. configure the stream object that receives the message using the correlation you defined and assign a value to the property alias.

  I also found this article - work with Custom correlation in Oracle BPM Suite 11 g, in which the author gave 2 examples of correlation in a taxi booking process. In these two examples, the author has also launched the correlation on the sender of the message.

  In the same section in the book, the authors also wrote:

  You need not set properties of correlation on the outputs of the process, for example, his task to send, or any nodes (message) or throw the message events. Only the entries have correlation properties.

  I would be very grateful if you could help me with the following questions:

  1. When is the right time to launch a correlation of the receiver / called process side?
  2. How the calling process can decide that the answer message he receives come from what conversation if the calling process is not the man who hired the conversation?
  3. Regarding the last quote, is there something wrong if I define a correlation on the output to send to a specific process instance?

  After spending some time more twisting my brain, I think I got a better understanding of the correlation now. Please correct me if I am mistaken on all parties. It is important to me.

  Where and when to define the correlation depends on 2 rules:

  1. Who needs conversation?
  2. Who is 1st to 'talk' in the conversation?

  There are a few scenarios that can occur:

  1. Only the called process needs a conversation.
  • This scenario has been well illustrated by Arun Pareek in his blog - working with Custom correlation in Oracle BPM Suite 11 g, in which he created a process for the taxi booking. As shown in the image below, there is a subprocess embedded to cancel a reservation. In this case, when another part of the system (process BPEL, BPMN process, etc.) sends a message "Cancel booking" in this process of 'CabBooking', he cares not the conversation. It is the process of 'CabBooking' who needs conversation in order to correct one of its instances running unreserve. The system of 'other', it's just a normal service with the ID of the booking must be cancelled.
    • 
  • Commercially, this kind of procedure should be the case for most popular use for correlation. Technical side, this is where Mark Nelson is correct. As only the called process needs the convesation, the calling process has no need to be aware of the correlation. According to rule 2, it is clear that the activity of "Booking request" is the place to initiate the correlation, as has been done by Arun Pareek in his blog. Subsequently, the "cancel booking" activity can make use of this initiated correlation.
• Only the calling process must a conversation.
  • This scenario was presented by Mark Nelson in his book - Oracle BPM Suite 11 g BPMN topics advanced, in which he had 2 processes: process a to calculate a square of a number and process B to submit the number of input in the process a as shown below. In this example, the correlation was launched on the 'send task' B and used by the task "receive" B processes processes. In the process, he doesn't care the conversation; his only responsibility is to calculate the square of the number of input. It's the process B who needs conversation in order to map the correct response to the Service call message.
• A.
• B.
• Technically, the motivation for this scenario is usually the case in which the called process is a reusable service that does not support WS-addressing. Generally, within the Oracle stack, as WS-addressing is supported, simply use the default conversation without having to manually create correlations. On the business end, one of the reasons for having such a correlation are to ensure that there is only one instance of execution with a special at any time correlation key. For example, for procedure B, it is not possible to have 2 instances with the same number of entries as the correlation key. After the 2nd rule, it is easy to know where to start the correlation - the activity of the 'task to send' process B.
• The appeals process and need a conversation.
  • This scenario is also confirmed in blog Arun Pareek. In the 'Get booking availability' embedded sub-process, he throwed a 'send booking confirmation' signal and then took in the primary process to the activity of «Get answer of Confirmation booking»
. In fact there are 2 ways to implement correlation in this case.
• 
  • Assuming that no correlation has been created in the process of "Taxi booking" (ignore the use case for "Cancel booking" above), sequel to the 2nd rule, as the "send Booking Confirmation" is the activity 1 to join the conversation, this is the place to initiate the correlation, as has been done by Arun Pareek. Subsequently, the 'Booking Confirmation Response Get' will make using the correlation of insiders.
  • We can push this scenario 1 scenario - "only the called process needs a conversation", by launching the correlation directly on the first activity, "Booking request", "cabin booking" in the process of. Using this approach, the 'availability reservation get' subprocess embedded as well as any other external processes that need to communicate with the 'Cabin booking' process will not be aware of the conversation. In this case, all the 'other' process is enough to call a favor. Correlation initiated on the 'Cabin booking' process will take care of the rest.
  • The 2nd approach should be the best practice, because it helps ensure that the correlation is initiated in only 1 place - the 1st activity of the process called and reused by anyone.
• Subprocess multi-instance incorporated.
  • This scenario was presented by Mark Nelson will use a process C, which is a sub-process multi-instance incorporated, to call multiple service to the process above. In this example, the correlation was launched on the 'task to send' C process and used in the 'task to receive' C processes. Once A process has no need to be aware of the conversation. In addition, a conversation of scope must be defined inside the embedded sub-process.
    • 
  • The reason why the correlation must be initiated the calling process is because only one instance of the calling process is several service call to the same process in parallel at the same time. On arrival, the calling process response messages must know what answer message is requiring the Service. Again, within the Oracle stack, as WS-addressing is supported, it is sufficient to use brought the conversation without having to manually define correlations.

This is how I understand the correlation at this stage. Please let me know if I misunderstood something.

Thanks a lot for your help!

• How a BPEL workflow can return in the middle of the process?

  I had a BPEL process, and after invoking a web service, a response back to the client immediately, and the flow continues. However, I found that he did not return to reality, he returned after the process is complete. And if I added a component to wait 3 seconds. She would not immediately respond, but during performance tests, it seems the process would not continue after 3 seconds and cause other problems. Y at - it of to way the answer immediately returned without the component of waiting? Thank you very much.

  This behavior is specified... Even if you do an action to respond, the real answer will be returned only when he meets a dehydration activity. Different options available to you are:

  1. in the PS3, there is a new Checkpoint activity, which you must do immediately after the answer.
  2. in previous versions of 11 g, you can use a business exec of java with a call to breakpoint() immediately after the answer
  3. you define a component property called idempotentReply to true. Then you don't need no activity of dehydration explicitly in the bpel code. Drawback with this approach is that this setting will affect all operations on all the links of sponsors on the component.

• The process in javascript checkbox array

  Hello
  I am new to Apex and Javascript. I use Apex 3.2.0
  
  Here's my question: I have a tabular report with a selector box. When the user clicks on the button Delete, I want to display an alert if the user selects all records.
  The array containing the boxes checked in the tabular report is Apex_Application.G_F30
  
  I know that I can verify the account in a process of pl/sql with "If apex_application. G_F30. Count = 0 then... "
  
  However, I want to do this in javascript. The script called for the delete key. If the count is zero, the alert would be "no files not selected" or something like that. If the number is greater than zero, then I submit the process to delete the records.
  
  Your help is greately appreciated.
  
  Kind regards
  
  Al
  

• Supprised to find that answer BPEL does not complete the process

  BPEL 10.1.3.1.0
  
  We are new to BPEL realitvely so we were surprised by the activity of the response behavior.
  
  We have a few synchronous BPEL process that call each other. We go back an outputVariable/object that indicates the success or failure of the called procedure. After invoking any of these processes, we check the content of the object that is returned in the response.
  
  If the object indicates an error, we the output variable and run a response to our caller.
  
  If the object indicates success, control passes to the next activity in sequence.
  
  But when we looked at the check in the console, we noticed that the execution of the process ended with the answer. Several of the following activities indicate that execution continues and (potentially) returned separate response at the end of the process.
  
  We read this right? There are actually two returned messages?
  
  We would expect this kind of behavior for an asynchronous process, but they are all synchronous processes.
  
  In our tests, the first response sent is received. But we wonder what could happen on a busy network. If the messages back for some reason, the calling process may receive a success message when there really was a problem.

  You are right

  If you model the additional activities after the response that they will be executed.
  you could do the response of your catch error handler