Question about BPEL cookbook "Making BPEL Processes Dynamic" again

Hi,all.
About the last example which is getting loan serveces endpoint info from db.
There is no receive activity in the example code. I wanna add it and make it run. Does the receive activity from the LoanService should be in or out of the while loop. If it should be out of the while, it seems as if it needs another while loop to contain the receive activity to get returning results according to the number of loan provider it has sent applications to. In this situation, how should I manipulate the returning variables? And if all of this had been done, how could I select the best offer of such a lot of returning offers?
I need your help. Thanks a million. This is very import to me.
Peter

Marc,
Sorry I didn't see your question sooner. Answers to your questions:
Yes "i" is a variable (a counter for the loop). If you look a little further up in the example it shows where we're initializing that variable.
Yes the receive needs its own while loop as stated in the article -- but this is only if you're running the child services in parallel. Better yet, with 10.1.3 we have scoped partnerlinks per bpel 2.0 so you can use flowN to call a bunch of dynamic services in parallel, each with its own receive.
As for handling the results you may want to create a nodeset to store them then use xpath to select the best offer. This would be after the receive while loop (10.1.2.x) or the flowN (10.1.3) receives all the results back. Another thing that's nice in the 10.1.3 version is that you can put timers around the receive so you ignore results that don't respond in a reasonable amount of time. In 10.1.2 you can do this too by putting a timer around the while loop but it's not quite as clean as the 10.1.3 way.
hth. Sean

Similar Messages

How to assign a parent or child process to BPEL process Dynamically

Hi friends,
root_id (varchar) - the conversation id of the instance at the top of the invocation tree. Suppose A -> B -> C, root( B ) = A, root( C ) = A, parent( B )= A, parent( C ) = B. This instance, instance at the top of the tree will not have this set. There will be relation between ROOT_ID, PARENT_ID and CIKEY in the cube_ instance at dehydration data base.
Here I want to assign root_id or parent_id dynamically to my bpel process dynamically. Let me know if anybody knows about that.
Thanks in advance...
Hari Mandadapu

Hi,
I'm more puzzled as to why you would want to do this? As you have explained these columns provide details of the process tree. Why would you want to alter the values set by BPEL there by changing the process tree it created?

Asking about BPEL Process Manager API information

Hi,
I am looking for BEPL PM API to have better administrative control from our application, mainly want to integrate the functionality to manually recover undelivered invoke/callback messages.
I checked BPEL cookbook - http://www.oracle.com/technology/pub/articles/bpel_cookbook/blanvalet.html
and it has a section of "BPEL Process Manager API and Dehydration Store", and the Link provided<$ORABPEL$\integration\orabpel\docs\apidocs\index.html> is for Java API.
I have two questions here:
1. Is there Java API which does recovery the undelivered messages?
1. Is there PL/SQL API available? Is there one to handle recovery the undelivered messages?
Thanks!
Grace

Hi me_sun, me_oracle :)
The bpel api can be found here
http://download.oracle.com/docs/cd/B31017_01/integrate.1013/b28986/toc.htm
workflow found here
http://download.oracle.com/docs/cd/E12483_01/integrate.1013/b28985/toc.htm
when us install BPEL you do get BPELs human workflow component. This is different to the Oracle Oracle workflow which was based on the database and PL/SQL.
cheers
James

A question about BPEL data formats

Hi everyone,
I am new to BPEL and have a question regarding the format of data "inside" BPEL variables. Perhaps someone can clear this up?
When defining a variable in BPEL, I can either give it an XML Schema simple or complex type - or a "WSDL message type". I am confused with regard to the last one. A WSDL message has parts, and the parts have types themselves. These types, however, can be from an arbitrary type language. XML Schema is one possibility, but there might be others. I have two questions with regard to this:
1) If I use some proprietary schema instead of XML Schema, can I handle data in this format in the BPEL process? I.e., can I assign "literal" data to the parts in my own format, and (if it is XML-based) manipulate it with XPath?
2) WSDL allows the definition of styles and encodings in the concrete part. How does the conversion happen between what's on the wire (in the SOAP body) and inside the BPEL variables? I.e. when using RPC/Encoded, does the Oracle BPEL engine automatically convert everything into plain-XML-Schema-validated-XML and back or does RPC/Encoded XML show up inside the variables? What if the BPEL engine doesn't understand the concrete format?
Thanks a lot!
Reto

Fabio,
I am not sure I understood you clearly. But ODI uses the underlying technology of the target database to perform the CDC. There are several KMs for each technology that let you achieve the same.
eg. For Oracle, a trigger based CDC is available which creates triggers on the underlying tables in the database to capture the changes.
Also for Oracle, a Logminer based CDC is available which reads the Oracle logs to capture the changes made over to tables.

Question on BPEL Process Performance

Hello,
We have a BPEL process reading the datafile through file adapter and upserting into DB using DB Adapter. Our requirement is
If there are 10 records to process and two records (record 5 and 9) fail while inserting/updating for some reason(i.e data type mismatch, column length mismatch etc..), at the end of the process you should see 8 records in the destination table and two records in error table.
I know there are solutions of this :
*1) Multiple calls to DB:* Use a While loop in a BPEL process and Invoke DB adapter for each record and use exception handling(Catch all block).
*2) Invoke Store Procedure:* to prevent multiple calls to DB, create a stored proc on DB side to iterate and insert the records and the stored proc should also return the IDs of failed records back as error response so that you can insert those failed records to a log table or in log files.
Can you suggest which solution is best in terms of performance and why ??
Also we need to perform some business validation (i.e NOT NULL check, date format check etc..), Where should we perform this.. at DB level or BPEL process level?? and why..
Thanks,
Buddhi

BPEL is a slow performer.
Always call a stored procedure to do complex data processings.
Hence go with the second approach.
Error records:
If your going to log errors in the same database, insert the error details direcly into the error table. Dont go back to BPEL.
Application specific validations should be handled in the application itself.
--Prasanna

Question about EJB and thread safe coding - asking again!

sorry everyone, no one seems to be interested in helping me on this, so i post it again.
i have a question about the EJB and thread safe coding. when we build EJBs, do we need to worry about thread safe issue? i know that EJBs are single threaded, and the container manages the thread for us. maybe i am wrong about this. if i wasnot, how can we program the EJB so that two or more instance of EJB are not going to have deadlock problem when accessing data resources. do we need to put syncronization in ours beans, do we even need to worry about creating a thread safe EJB? thanks in advance. the question really bothers me a lot lately.

sorry everyone, no one seems to be interested in
helping me on this, so i post it again.Excellent plan. Why not search a little bit on your own instead of waiting for your personal forum slaves to answer your call. See below.
i have a question about the EJB and thread safe
coding. when we build EJBs, do we need to worry about
thread safe issue? Read this: http://forum.java.sun.com/thread.jsp?forum=13&thread=562598&tstart=75&trange=15
i know that EJBs are single
threaded, and the container manages the thread for us.
maybe i am wrong about this. if i wasnot, how can we
program the EJB so that two or more instance of EJB
are not going to have deadlock problem when accessing
data resources. do we need to put syncronization in
ours beans, do we even need to worry about creating a
thread safe EJB? thanks in advance. the question
really bothers me a lot lately.
Java's Thread Tutorial
JavaWorld: Introduction to Java threads
IBM: Introduction to Java threads
Google: java+threads+tutorial

Question about Illustrator CS5 making eps files in Mac OS 10.9.5

Hello, I am using Illustrator CS5 on a Mac. I just updates to Mac OS 10.9.5
Now when I save an simple Illustrator vector file as an eps, and place it in another program (such as Quark), I get code instead of the preview image. I tried older vector eps files, and they still work. i tried older Quark files, and they still work. I ruled out font problems. I tried importing new Illustrator ai file, same result. Any thoughts?

Ask in the AI forum and include proper details about your export settings and your artwork. EPS parsing errors usually mean you chose an unsuitable combination of settings and/or the embedded previews are botched.
Mylenium

Question about Tables, New Territory for me once again.

Adding a Table to a scrollPane is easy enough but what I am trying to do is add Label1, Table1, Label2, Table2, Label3, and Table3 to a single scrollPane. Is this possible?

Sorry, Yeah I do mean JTables.
Also I was hoping to have it setup so that I could have them all in a single scrollPane and have additional rows in a table would just drop everything else down a little more in the layout. I have only seen one way to put a table into a scrollPane and that's by doing:JScrollPane scrollPane= new JScrollPane(table); Is it possible to do Label, Table, Label, Table, Label, Table all controlled by a single scrollPane and show all of the data in each? These tables are all related material and it would be a major pain to have to scroll down each of the tables individually.

Questions about web services and BPEL in ALBPM

Hi all,
We are currently evaluating various BPM products in order to decide which one to use. Currently I am evaluating ALBPM Studio 5.7 and I am running my processes on the embedded engine provided with the studio. I have some questions:
1. We want to be able to use complex types for the web services that are generated for processes. Will this limitation be solved any time in the future?
2. We want to be able to use the generated web services directly (that is, without calling the startSession method first). Is that possible?
3. I could not find extensive documentation about developing BPEL processes. There is only one small chapter in the developer documentation about BPEL processes. Is there any more documentation or sample about how BPEL is used in ALBPM?
4. I have developed a very simple BPEL process, but when I try to deploy it, I get the error "Activity BPEL receive does not have any incoming transitions". How can I resolve this issue? I guess it may be a problem about my PartnerLink definition.
5. When I try to catalog a wsdl which imports another wsdl (and when the imported wsdl also imports some xsds from another namespace), I get an error which says that a web service xxx with namespace yyy is not found, but actually the thing named xxx is not a web service but a soap fault. What could be the problem?
6. Is it possible to write a BP method in PBL for a BPEL activity?
Thanks in advance and best regards..

look for anwsers in the other newsgroup posting:
http://forums.bea.com/bea/message.jspa?messageID=600047258
MAriano
Benitez/BEA/BPM/Beautiful

Business Activity Monitoring for BPEL Processes

Hello,
IEP is capable of monitoring real time events, Real time business event collection and processing.
In that case, can we use IEP to monitor BPEL process activities with activity data.
One very big part, the ability to provide functionality for third parties (clients like Me) to register some sort of observer to capture live BPEL process activity, is missing.
I'm trying to cover that big hole in my project. That's the reason I'm posting about BPEL process management capabilities and events.
But I got no concrete answers for that. Not every business use can be provided by any generic system.
I looked at CAM (which comes with Alaska build). But I couldn't get much info about entire picture.
Other BPEL engines, the one I have used, has the functionality to do activity monitoring and process management. (IBM Process server, Apache ODE)
Could any IEP, BPEL component developers answer any sort of solution to my use case?
Thank you
Raja

I agree that this would be very useful. If anyone can provide additional insight it would be much appreciated.

How to enable https in Oracle BPEL Process Manager 10.1.2

Hi,
I have a few security related questions surrounding BPEL process manager.
1. Does the BPEL engine have the capability to invoke a web service using https (HTTP over SSL)? Does it automatically do that if partner link URI starts with https:// ?
2. If not, what needs to be done to enable accessing a https based web service?
Thanks,
Vidya

Eric,
I had applied the steps specified in the URL and modified the files default-web-site.xml, secure-web-site.xml and server.xml in directory Ora_Home\j2ee\home\config. Able to invoke the application deployed in Ora_Home\j2ee\home\applications successfully using https.
For securing BPEL process I had followed the same steps by modifying files default-web-site.xml, secure-web-site.xml and server.xml in directory Ora_Home\j2ee\OC4J_BPEL\config. While invoking the BPEL Processes using https protocol I am facing error “HTTP 500 - Internal server error”, after the security alert popup.
Do I need to do any other configurations for securing BPEL processes deployed in Ora_Home\integration\orabpel\domains\default\deploy directory?
Please assist me on this.
Thanks,
Vidya

BPEL process cancellation API?

is there any way to cancel bpel process dynamically?
some API which by cikey will cancell the process? (wf_engine.abortprocess equivalent)
thanks in advance

Hi me_sun, me_oracle :)
The bpel api can be found here
http://download.oracle.com/docs/cd/B31017_01/integrate.1013/b28986/toc.htm
workflow found here
http://download.oracle.com/docs/cd/E12483_01/integrate.1013/b28985/toc.htm
when us install BPEL you do get BPELs human workflow component. This is different to the Oracle Oracle workflow which was based on the database and PL/SQL.
cheers
James

SOA Suite 10.1.3.1.0 BPEL Process, no domain drop down.

I've installed SOA Suite 10.1.3.1.0 with an Oracle 10g Enterprise Dehydration Database on a separate server. Currently we have two BPEL Domains, however, when trying to launch the BPEL Console, I do not get a drop down list to select between the two domains. I've searched and so far none of the answers seem to fit.
I'm using RHEL 4. Thanks!

Hi ,
Even , I am facing the same problem .
If I go to BPEL Admin , there also I am not getting the Domain List .
I reinstalled all the process and again I instllaed the 10.1.2 BPEL Process .
Again did the same thing that is creating user, tablespace etc in Linux Machine .
Still not getting the default Domain ...
Please suggest me where I am wrong ....
Thanks
Prashant Dwivedi

BPEL process crashes randomly

Iam working on softwareproject based on JAVA Web Services, deployed on a AXIS server, which are orchestrated in a BPEL process. The BPEL process was created with JDeveloper and deployed on the Oracle BPEL PM 10.1.3.1.0.
The main WMYPC BPEL process invokes several other BPEL processes, which represent functionality of lower layers. The lower BPEL processes invoke again lower BPEL processes. So there exist three layers of BPEL process.
Each BPEL process was tested and worked correctly, but when the main BPEL process is called, which involves the other BPEL processes, the main process crashs randomly on different stages of progress, after a lower process is getting invoked with the following error message:
Your test request was processed synchronously. It took 83.464seconds to finish and generated the following output:
Value:
<Faulthttp://schemas.xmlsoap.org/soap/envelope/>
<faultcode>env:Server</faultcode>
<faultstring>java.lang.IllegalStateException: Cannot call setRollbackOnly() current thread is NOT associated with a transaction</faultstring>
</Fault>
Beside the error above, a second error occures sometimes.
Your test request was processed synchronously. It took 64.891seconds to finish and generated the following output:
Value:
<Faulthttp://schemas.xmlsoap.org/soap/envelope/>
<faultcode>env:Server</faultcode>
<faultstring>com.oracle.bpel.client.delivery.ReceiveTimeOutException: Waiting for response has timed out. The conversation id is 70976ad86304fa5b:-18ed62ca:111af3d9f30:-7ff0. Please check the process instance for detail.</faultstring>
</Fault>
The development enviroment runs on a virtual machine running MS Windows Server 2003 SP1 with performance problems. So i thought there is a timeout problem and changed the syncMaxWaitTime for the BPEL domain to 120, but i didnt saw any improvements.
I guess its a PM configuration problem and hope somebody can help me to solve it.
Thanks for your help in advance.
Regards Ronbee
Message was edited by:
ronbee

Hi Ronbee,
I guess you are using VMWare for Windows 2003 Server.
If that's the case , then its an obvious performance issue.
Cheers
Anirudh Pucha

Questions about the load processing of OpenSparc T1 Dcache

Hi,
I have some questions about OpenSparc T1 Dcache load processing.
During load processing, subsequent loads to the same address need to search the store buffer for a valid store to that address. If there is a CAM hit, data is sourced from the store buffer, not from the D-cache, and no load request will be sent to the L2.
What if there is no CAM hit. Would the load request be sent to L2? Or would Dcache be checked for the requested data?
If the load request would be sent to L2, what next? Would the Dcache be updated?
Thanks

Store buffer is checked for Read after Write (RAW) condition on loads. If there is full RAW - i.e. full data exists in the store buffer - then the data is bypassed and no D cache access happens.
If RAW is partial (e.g. word store followed by a double word load) then load is treated as a miss. Store is allowed to complete in L2 cache and then load instruction is completed.
For the miss in STB, D cache is accessed. If hit, data is fetched from D$. If miss, data is fetched from L2$ and allocated in D$.

Question about BPEL cookbook "Making BPEL Processes Dynamic" again

Similar Messages

Maybe you are looking for