Thank you, Rafal. See https://issues.apache.org/jira/browse/ODE-760
-Jon -----Original Message----- From: Rafal Rusin [mailto:[email protected]] Sent: Thursday, February 04, 2010 5:37 PM To: [email protected] Subject: Re: Correlation set problem I tried your process on ODE 1.3.3 and it shows this error. Please create a Jira entry for it, because it's a bug. On 5 February 2010 01:07, Jonathan Coogan <[email protected]> wrote: > I recently ran into this problem also (Ode 1.3.3, war distribution). I'd > like to know if any of the Ode devlopers can comment on this. > > In my tests it fails even without a loop. If you have more than 2 Receive > activities (with correlation) the server will throw an internal error on the > 3rd one. The first 2 work. > > Here's a stupid simple example. I created a process that you can call 4 > times to add 4 different numbers together. Each time you call it you pass in > an int value and a String that is used to identify the instance. Each time > you call it, the process will return the running sum of the previous inputs. > After the 4th input the process will complete. It works the 1st 2 times and > hangs on the 3rd run. The server console says, "INTERNAL ERROR: No ENTRY for > RESPONSE CHANNEL 20" and prints this stacktrace: > > java.lang.IllegalArgumentException: INTERNAL ERROR: No ENTRY for > RESPONSE CHANNEL 20 > at > org.apache.ode.bpel.engine.OutstandingRequestManager.associate(Outstan > dingRequestManager.java:145) > at > org.apache.ode.bpel.engine.BpelRuntimeContextImpl.inputMsgMatch(BpelRu > ntimeContextImpl.java:934) > at > org.apache.ode.bpel.engine.PartnerLinkMyRoleImpl.invokeInstance(Partne > rLinkMyRoleImpl.java:220) > at > org.apache.ode.bpel.engine.BpelProcess.invokeProcess(BpelProcess.java: > 241) > at > org.apache.ode.bpel.engine.BpelProcess.handleWorkEvent(BpelProcess.jav > a:408) > at > org.apache.ode.bpel.engine.BpelEngineImpl.onScheduledJob(BpelEngineImp > l.java:439) > at > org.apache.ode.bpel.engine.BpelServerImpl.onScheduledJob(BpelServerImp > l.java:441) > at > org.apache.ode.scheduler.simple.SimpleScheduler$4$1.call(SimpleSchedul > er.java:411) > at > org.apache.ode.scheduler.simple.SimpleScheduler$4$1.call(SimpleSchedul > er.java:405) > at > org.apache.ode.scheduler.simple.SimpleScheduler.execTransaction(Simple > Scheduler.java:218) > at > org.apache.ode.scheduler.simple.SimpleScheduler$4.call(SimpleScheduler > .java:404) > at > org.apache.ode.scheduler.simple.SimpleScheduler$4.call(SimpleScheduler > .java:401) > at > java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303) > at java.util.concurrent.FutureTask.run(FutureTask.java:138) > at > java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecu > tor.java:886) > at > java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor. > java:908) > at java.lang.Thread.run(Thread.java:619) > > Thank you. > -Jon > > > > -----Original Message----- > From: Brian Lund [mailto:[email protected]] > Sent: Monday, January 18, 2010 9:18 AM > To: [email protected] > Subject: Re: Correlation set problem > > Hi Dave, > I had a bit of trouble getting the correlations to work inside loops as well. > Your BPEL(s) didn¹t come across for me so i¹m not sure if you¹re already > tried this, but have you tried to break out the sequence/flow inside your > loop into It¹s own process (i.e. The first activity inside the loop would be > an invoke, and the other process would instantiate a new process on each > receive)? Perhaps it would be an easier correlation to create since ode > would be spawning an entirely new process to send message(s) to? > I would be very interested in your solution when you get this to work. > > Thanks, > Brian > > On 1/17/10 5:57 PM, "Dave Chodos" <[email protected]> wrote: > >> Hi, >> I have been trying to get a long-running "counter" process up and >> running on ODE, but have run into timeout errors when I make repeated >> calls to the process. Specifically, my process has a receive to >> initiate the process, and then another receive in a while loop that >> increments the counter. The initial call, and the first call to the >> increment operation appear to work fine. >> Subsequent calls to the increment operation, however, simply time >> out, and nothing is returned. >> >> Initially I was working with my own process, but then decided to >> try out a sample "counter" process (which I found >> here:http://svn.apache.org/repos/asf/ode/trunk/bpel-test/src/test/res >> o >> urces/bp >> el/2.0/TestCounter/) to 1) make sure that this process executed >> properly, and then 2) try and figure out what was going wrong with my >> own code. To my surprise, this process exhibited the same behaviour* >> (timeouts on increment calls after the first one), which makes me >> think that either I'm passing the wrong parameters into the process, >> or that there's a problem with my ODE setup. >> >> To test out whether the problem was with all correlation-based >> processes, I wrote another process that has a sequence of different >> correlated receive operations, and this one works fine. However, >> calling different operations within a loop (using a pick within a >> while loop, for instance) results in the same timeout problem. Thus, >> it seems that the problem is with calling >> receive(s) multiple times within a loop (while, do-while, for-each, >> etc.) >> >> I am running Apache Tomcat 5.5.27 on Unix, and using ODE 1.3.3. >> For what it's worth, I'm using NetBeans to write my BPEL processes, >> but I haven't had problems with other processes I've written, so I >> don't think it's a NetBeans issue. >> >> I've attached the code for my counter process, as well as the code >> for the working sequential correlation process. I also pasted my SOAP >> calls to the counter process (generated using SoapUI) so that you can >> see what I've tried there. I suspect that either there is something >> wrong with my Apache/ODE setup or I am misunderstanding something >> fundamental about how correlation sets work >> -- I'm relatively new to BPEL and web services, so this is a definite >> possibility :) >> >> Thanks in advance for any help you can provide. >> >> Dave Chodos >> >> * When doing one last test this morning, I found that the counter >> process is now timing out on the first attempt to call >> getAndIncrement, rather than the second, like it was doing last night. Not >> sure what to make of this... >> However, it is definitely not working as expected :( >> >> *** call to init operation *** >> <soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/"; >> xmlns:coun="http://example.com/bpel/counter";> >> <soapenv:Header/> >> <soapenv:Body> >> <coun:init> >> <name>testProc</name> >> <alias>testProcAlias</alias> >> </coun:init> >> </soapenv:Body> >> </soapenv:Envelope> >> >> *** call to getAndIncrement operation *** <soapenv:Envelope >> xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/"; >> xmlns:coun="http://example.com/bpel/counter";> >> <soapenv:Header/> >> <soapenv:Body> >> <coun:getAndIncrement> >> <name>testProc</name> >> <alias>testProcAlias</alias> >> </coun:getAndIncrement> >> </soapenv:Body> >> </soapenv:Envelope> > > Regards, -- Rafał Rusin http://rrusin.blogspot.com http://www.touk.pl http://top.touk.pl
