Control instance cretion

Hello experts,
I have a scenario here. I have a singleton bpel process which listens to a JMS queue.
The singelton process picks messages from queue and processes and if there are any errors in the processing its taken over the Fault management framework and marked for manual intervention. Which mean this process is halted till some person manually does some action.
But meanwhile the messages in the queue are consumed by the JMS adapter and i can see the status as running in the em console. Though the messages are ye to be received by the bpel they are consumed from the JMS queue. Is there a way to avoid this ? as in to stop the jms adapter to consume message till BPEl is available ?
Is retiring the process when a fault occurs a better idea?
Thanks,
Prakash

What have you set for adapter.jms.receive.threads ?
http://docs.oracle.com/cd/E25178_01/integration.1111/e10231/adptr_propertys.htm#CHDCDAHF
Cheers,
Vlad

Similar Messages

ESB Control Instances Tab is Empty

Hi,
I have working on SOA Suite 10G. I have deployed ESB CustomerData Tutorial sucessfully. But this ESB is not activated meaning ESB should read customer XML from input directory every second and write into database but that is not happening. If you login to ESB Control Instances TAB is empty. Meaning ESB is not getting invoked at all.
PLease let me know what can be the issue.
Regards,
PB

Hi.
Probably your File Adapter is not being triggered by the polling activity. This can be for several reasons. You need to check all your adapter configuration like file name pattern, and that your physical path where your XML file resides matches the one you configured in the adapter.
Also check for the logs under ORACLE_HOME\opmn\logs.
Denis

Auto Update X Control Instances

My LV version is 8.6.1 Developer
Are there any XControl experts out there? Every time you make changes to the Xcontrol GUI but previously had utilized that XControl instances in 100 different VIs, is there a way to automatically force LV to update all those 100 VIs along with its GUI dimension? Right now, I have to manually open all those 100 files then delete the old instances, drag and drop to create a new instance ....this is really annoying!

mishkin wrote:
I don't think you need an XControl expert on this one. As far as replications, XControls should behave just like every other control. If you want to make a change to the control and have it be replicated to previous instances, make a typedef or a strict typedef out of it. Otherwise, you'll have to go back and edit old instances of your controls manually...
No, I don't think we're on the same page here buddy. You're thinking of the normal standard control but my question was xcontrol specific. With a normal control with strict type def enabled, all your future instances will reflect all the changes. Xcontrol doesn't have it. What it does have, however, is when moving from xcontrol child data to child facade file using strict type def. That's the best you will get out of using strict type def. However, once you created an instance of that xcontrol, it becomes finalized. If you try to right click on the new xcontrol instance, try to locate the feature "disconnect from type def" ...you won't see it. It doesn't exist. Whereas with a normal control instance, you will see it.
Another interesing point too even when using a normal control, after you created a new instance but accidentally disconnect it from type def, good luck. There is no way to reconnect them. I thought NI should look into fixing all these quirky issues especially with all the other issues associating with xcontrol.

CTS + error during deployment There is no cluster control instance running on host

We are running our QA portal in a cluster (3 hosts..one for MS, 1 for ERS, 1 for JC). The alias for the JC is <jcalias> and the real name of the host is <clustername1). So depending on which component is one which of the 3 cluster hosts wil dictate which host <jcalias>.
But when trying to deploy a portal archive via CTS+, we get error:
Jul 24, 2014 11:18:12... Error: An error occured while starting a cluster instance.
Jul 24, 2014 11:18:12... Error: There is no clutser control instance running on host sapepqci which is described in SecureStorage .
The instances, returned by MessageServer [MS host: <msalias>; MS port: 3912], are :
Name:JM_T1405698721103_0_<clusterhost1>.viacom.com Host:<clusterhost1> HostAddr: aaa.bbb.ccc.ddd State:5
Name:JC_<jcalias>_EPQ_14 Host:_<clusterhost1>. HostAddr: aaa.bbb.ccc.ddd
State:5
Name:JC_<appsrvr1>_EPQ_00 Host:<appsrvr1_IP> HostAddr: <appsrvr1_IP>
State:5
Name:JC_<appsrvr2>_EPQ_00 Host:<appsrvr2_IP> HostAddr: <appsrvr2_IP>
State:5
Please check if there is an appropriate running cluster instances. For more inforation check CSN Note 1276366
Jul 24, 2014 11:18:12... Info: ***** End of SAP J2EE Engine Deployment (J2EE Application) *****
I have checked the referenced not. The profiles all use the JCALIAS as the host name, and in the configtool secure store it's also specified in admin/host/<SID>
If I change it in the secure store to the real name (clusterhost1) that the message server is giving, it will probably work. But what happens in the future when, after a cluster resart, <jcalias> gets assigned to <clusterhost2> instead? I'd rather be able to use the alias in secure store and for the message server to pick that name up and use it.

Hi David
Jul 24, 2014 11:18:12... Error: An error occured while starting a cluster instance.
Jul 24, 2014 11:18:12... Error: There is no clutser control instance running on host sapepqci which is described in SecureStorage .
You cluster resources are online?
BR
SS

There is no clutser control instance running on host

Hi
I am getting following error message when deploying packages through SDM
Jun 9, 2009 12:29:53 PM Info: Software type of SDA: J2EE
Jun 9, 2009 12:29:53 PM Info: ***** Begin of SAP J2EE Engine Deployment (J2EE Application) *****
Jun 9, 2009 12:29:53 PM Info: Starting cluster instance processes.
Jun 9, 2009 12:29:53 PM Error: An error occured while starting a cluster instance.
Jun 9, 2009 12:29:53 PM Error: There is no clutser control instance running on host SBPEPCI1 which is described in SecureStorage .
The instances, returned by MessageServer [MS host: SBPEPCI1; MS port: 3900], are :
Name:JC_CPCAPP117_EPP_04
Host:CPCAPP117
Name:JC_CPCAPP115_EPP_03
Host:CPCAPP115
Name:JC_CPCAPP114_EPP_00
Host:CPCAPP114
Name:JC_CPCAPP112_EPP_02
Host:CPCAPP112
Name:JM_T1244432033820_0_CPCAPP109
Host:SBPEPCI1
Name:JC_CPCAPP109_EPP_01
Host:CPCAPP109
Please check if there is an appropriate running cluster instances.

Issue got resolved by changin following paramter in Secure Store
admin/host/<SID>
I replaced above virtual hostname entery by actual hostname.
Message itself gives hint for not finiding server cluster name. there was mismatch between clsuter name specuified in secure store & information given by message server
Jun 9, 2009 12:29:53 PM Error: An error occured while starting a cluster instance.
Jun 9, 2009 12:29:53 PM Error: There is no clutser control instance running on host SBPEPCI1 which is described in SecureStorage .
The instances, returned by MessageServer [MS host: SBPEPCI1; MS port: 3900], are :
Name:JC_CPCAPP117_EPP_04
Host:CPCAPP117
State:5
HostAddress:172.17.134.50
Name:JC_CPCAPP115_EPP_03
Host:CPCAPP115
State:5
HostAddress:172.17.134.48
Name:JC_CPCAPP114_EPP_00
Host:CPCAPP114
State:5
HostAddress:172.17.134.47
Name:JC_CPCAPP112_EPP_02
Host:CPCAPP112
State:5
HostAddress:172.17.134.45
Name:JM_T1244432033820_0_CPCAPP109
Host:SBPEPCI1
State:5
HostAddress:172.17.131.38
Name:JC_CPCAPP109_EPP_01
Host:CPCAPP109
State:5
HostAddress:172.17.131.31

Controlling instances in dynamic forms with a strange twist.

Hello, Is there a way to program a numeric text field that controls the number of instances of a subform? For example, I would like to have a nested subform appear twice in the form if the user enters 2 in the numeric field. But more importantly, I would like to see its corresponding XML output appear like the following: <SUBUNIT_1> <LENGTH></LENGTH> <SEQUENCE> ELEESGGGLVQPGGRQSPEKGLETHYAE </SEQUENCE> <N_TERMINAL>Yes</N_TERMINAL> <C_TERMINAL></C_TERMINAL> </SUBUNIT_1> <SUBUNIT_2> <LENGTH></LENGTH> <SEQUENCE> ELEESGGGLVQPGGRQSPEKGLETHYAE </SEQUENCE> <N_TERMINAL>No</N_TERMINAL> <C_TERMINAL></C_TERMINAL> </SUBUNIT_2> If you pay close attention to the above code, you will see that I want the parent tag's ending digit incremented each time the subform is repeated. <SUBUNIT_1> <SUBUNIT_2> <SUBUNIT_3> ect . . . That is the twist! Is this programmatically possible? Regards. Harold

Harold, The code to create a specific number of instances is as follows: JavaScript Subform1.setInstances(2); or Subform.instanceManager.setInstances(2); FormCalc comments.setInstances(2); // Subform name = comments If you are looking for documentation watch out: it's listed as setInstance rather than setInstances! If your form is associated to a specific data structure via a Data Connection (typically an XSD) then a subform repeats based on a repeating element in the XML or, if the subform instances are instatiated programatically as yours would need to be, then the generated data will have repeated elements. If, however, your form fields are not bound to any specific data structure then you've got your work cut out but I think you could achieve the twist you require. I think you will have to write some code to build the XML structure that you require. Your code can refer to the fields on the form to extract the data and build the <dataset>. You'll need to know about scripting DOM (Document Object Model) to achieve this. A third option, if you are using Forms Server or LC Workflow to process the form submission, is to write Java code that processes the submitted data and transforms it to the required strucure. Or a fourth that has just crossed my mind is to associate an XSLT stylesheet to the form to process the outgoing doing. I'm afraid I don't really know any more about that to be 100% sure that would satisfy your requirement. Hope these ideas help. Neil

Controlling Instanced MovieClip timelilne from another .as file

Hi, I'm working on building a 'cover flow' similar to the scroll through iTunes uses for it's album covers. The only difference is my 'cover' needs a background MovieClip that will sit behind it and animate when teh images are scrolled to get it a 3D effect.
So I have a MovieClip i'm exporting called coverflowItem. This is the movie clip that will sit behind each 'cover' image and have the timeline i want to control from a different .as file. That file is called coverflow.as where all of the elements are assembled and animated. What i need to do now is find out how to control each instanced coverflowItem so when the arrow buttons are clicked, not only do the covers slide to the appropriate side which they are currently doing now, but also have the associated MovieClip play giving the 3D illusion.
I can call the functions from the coverflowItem.as in the coverflow.as but have had no luck actually getting the movie clips to play properly.
//Create CoverflowItem
var cover:CoverflowItem=new CoverflowItem(_data);
this code lets me assess the functions but not actually manipulate the MovieClip. I'm sure my logic if flawed. I'm new to working with packages so any help would be appreciated!

if you call it from a timeline you can pass a display list item (var whatever:MyGlobal=new MyGlobal(this) ) to the constructor and assuming you're trying to reference the parent of that timeline, use:
package
    import flash.display.MovieClip;
    public class MyGlobal extends MovieClip
        public static var _dolares:Number = 3500;
        public static var _visitantes:Number;
        public static var _retroalimentacion:String = "Info box"
        //objects
        public static var _senal:Number = 0;
        public static var _flores:Number = 0;
private var dlObj:MovieClip;
        //my function
public function MyGlobal(mc:MovieClip){
dlObj=mc;
        public function decision(valor, visitantes, mensaje, irAbra, irDibujo)
            if (MyGlobal._dolares >= valor)
                //spend money
                MyGlobal._dolares -= valor;
                //visitor effect
                MyGlobal._senal = visitantes;
                //message for info box.
                MyGlobal._retroalimentacion = mensaje;
                //show info box
                MovieClip(dlObj.root).gotoAndPlay(2);
                //////close menu ;
                MovieClip(dlObj.parent).objetoAbra.gotoAndStop(irAbra);
                //Set graphic;
                MovieClip(dlObj.parent).objetoDibujo.gotoAndStop(irDibujo);

Installing Grid Control - new instance - then db server?

Linux 64 bit system. We plan to dual purpose a Grid control instance on the same Linux server as a reporting database server, running Oracle 10g 10.2.0.4, GC at 10.2.0.5 (the reporting db may or may not be managed under this grid control instance - other servers will have agents deployed to be managed by this Grid control instance). Neither have yet been installed. I'd prefer to have Grid Control running on it's own instance completely separate from the reporting database, in separate homes. Is this possible? Also, would it be better to install Grid control first, in its own home, then install the other database instance in it's own home? Or vice-versa? Or does it matter? If there's a document or note that covers this, I would be most appreciative if someone could point me to it.
Thanks!
Kelly

The Grid Control Repository database can be in the Same HOME as the OMS, however the database version for the GC repository may not be the version you want. I almost always setup a HOME for my grid control respository database sep. from my OMS home. This serves 2 purposes. 1 the security patches for my database and my database version can be kept up to date sep. from my OMS. Also allows me to sep the OMS from the database onto different servers much easier later.
I would recommend using a sep ORACLE_HOME for GC repository database from the OMS. This has made my life simplier in the past as the database version in the OMS binaries is usually a few revisions behind and I like to keep my DBs on the latest patch set for support purposes.
The means that your server will have 3 homes
1 for GC Repoistory Database
1 for Reporting database
1 for OMS
At least this is the way I would do it, unless really stapped for disk space for the additional homes.

Multiple Instances of Sample Planning App Slowing down Server?

I am trying to work with the sample planning application that came with the installation of EPM system 11. In order to see what effect my changes have in the application and in an attempt to understand the functionality of the Planning, I have created several instances of the sample application so I can have a control instance, and several that I am changing. I currently have created and initialized 4 versions of the Sample Applicaiton, and recently the server is freezing. It will be working fine for about 5 minutes, and then essentially freeze for 10 minutes. Monitoring the Performance on the machine shows that the CPU usage spikes to about 80% during the freeze times, and then drops to below 10%. Does anyone know if having multiple Sample Applications on a single Essbase Server may be causing this problem? Thank you.

I see another thread on this subject. Thanks.

Usage of Apache Beehive JMS control inside Message Driven Bean

Hi,
I am developing a j2ee application using weblogic workshop 10.3. This application contains the following three projects.
- J2EE Utility Project
- Message Driven Bean Project
- Ear Project
In the Utility project I have created one apache beehive jms control. I am using this control inside the message driven bean application. I haven't got any issues during the development and deployment phase with the beehive control. But during runtime the apache beehive jms control is not getting instantiated and the mdb application is throwing the null pointer exception when the EJB container try to execute the methods on the control instance. I have used the declarative programming model to create the control instance inside the mdb application.
I am using the weblogic 10.3 server for testing and I have created the 10.3 weblogic+workshop domain. By default it has the beehive libraries deployed.
Can somebody help me out to resolve this issue?
Regards,
Bp

You can get the message directly with getText() method. Following is the snippet of code.
if (aMsg instanceof TextMessage) {
TextMessage msg = (TextMessage) aMsg;
String testMessage = msg.getText());
Kishore.

Issue with Task Control Call back and Event Choice in WLI BPM 8.1

Hi,
I am using a Task Control Factory to create Task Control Instances in a loop
and then
create tasks and assign them to a predefined group. I have an event choice
for two events
waiting for the onTaskCompleted and onTaskAborted event call backs.
I have two such operations (of task creation) in the same loop. For
simplicity I have used two separate task factory controls for each of the
tasks (because I even had worse problems using a single control factory for
both tasks in the same loop).
When I instantiate the process, through work list I am able to claim, start
and complete the first task. When I try the same procedure (claim, start and
complete) on the second task, the the event choice for the corresponding
task receives 2 event call backs. First it receives the onTaskCompleted as
expected, but then it also received another call back of the same type which
displays the error on the server log as:
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <on process
completion, 1 undelivered messages to
'sampleRequestTaskFactoryCtrl_onTaskCompleted'>
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <ejbLoad:
Conversation 1064411008439:null not found>
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <A message was unable
to be delivered from a WLW Message Queue. Attempting to deliver the
onAsyncFailure event>
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <ejbLoad:
Conversation 1064411008439:null not found>
Does anybody know why two events are fired for the same work list action?
Thanks in advance.
Henry M

Henry,
Please post this question on the weblogic.developer.interest.integration
newsgroup.
Regards,
Raj Alagumalai
Backline Workshop Support
"Henry Mugasha" <[email protected]> wrote in message
news:[email protected]..
Hi,
I am using a Task Control Factory to create Task Control Instances in aloop
and then
create tasks and assign them to a predefined group. I have an event choice
for two events
waiting for the onTaskCompleted and onTaskAborted event call backs.
I have two such operations (of task creation) in the same loop. For
simplicity I have used two separate task factory controls for each of the
tasks (because I even had worse problems using a single control factoryfor
both tasks in the same loop).
When I instantiate the process, through work list I am able to claim,start
and complete the first task. When I try the same procedure (claim, startand
complete) on the second task, the the event choice for the corresponding
task receives 2 event call backs. First it receives the onTaskCompleted as
expected, but then it also received another call back of the same typewhich
displays the error on the server log as:
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <on process
completion, 1 undelivered messages to
'sampleRequestTaskFactoryCtrl_onTaskCompleted'>
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <ejbLoad:
Conversation 1064411008439:null not found>
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <A message wasunable
to be delivered from a WLW Message Queue. Attempting to deliver the
onAsyncFailure event>
<Sep 24, 2003 9:43:56 AM EDT> <Warning> <WLW> <000000> <ejbLoad:
Conversation 1064411008439:null not found>
Does anybody know why two events are fired for the same work list action?
Thanks in advance.
Henry M

Running multiple instances from same oracle home

Hi Gurus,
I am preparing for OCA 11g examination. I had a question about running multiple instances from same oracle home. While reading the book I came across a statement about database control which says "If there are several database instances running off the same Oracle home, then each will have its own Database Control instance".
Now I assume that this statement implies multiple instances accessing the same physical database (which as per my knowledge can happen only in case of a RAC).
I am not able to understand how exactly can you have multiple database instances in a single oracle home. I have installed Oracle 11g on my computer whose oracle home is +"D:\oracle\product\10.2.0\db_1"+; now this directory houses all the binaries (executables and folders like admin,bin, etc) for the oracle instance that I have installed (the instance name is ocp1). Now if I am to install or create another Oracle Instance (say ocp2) in the same home, wont there be a conflict of binaries of the two instances in the same home ?
Basically,it makes sense if db_1 folder is further split into multiple folders, one for each instance (for example, one for ocp1 & one for ocp2); each of these folders will then contain the binary files, libraries,etc for its own instance; but, this is not the case, here db_1 houses all the binaries directly under it ...
The other question which stems from the statement I mentioned above is that if having multiple instances running from same oracle home accessing the same physical database, can this arrangement be called RAC at all; or does RAC dictate that you cannot have multiple instances that are accessing the same database have to be installed on physically separate machines and so you cannot have two RAC instances (accessing same physical database) installed on the same computer ??
I hope my questions are not too much confusing ... !!
Cheers,
Sudripta

Hello,
"If there are several database instances running off the same Oracle home, then each will have its own Database Control instance".This is true, and also means each instance will have its separate database (i.e. dbf files etc.) & separate DB Control. Usually in this case you create several instances on the same node.
Now I assume that this statement implies multiple instances accessing the same physical database (which as per my knowledge can happen only in case of a RAC).You are correct about RAC where several instances (usually running on separate nodes) access the RAC.
More info:
1 Oracle Home - 1 Database - 1 SID is possible.
1 Oracle Home - Many Databases - Many SID is possible.
1 Oracle HOme - 1 Database - Multiple SID is possible and this is called RAC.

Standby Database for GRID instance

Hi All,
I have a requirement from a client to provide a Standby database (Physical or Logical) for the GRID control instance. Baffled by this, I tried to explain that the GRID instance is very small and may not require a standby. The client is adamant.
So, has anyone had any experience (experimental or otherwise) of configuration of a standby for GRID?. If so, how? and which would be a feasible option, Physical or Logical?
Any feedback is appreciated.
Regards,
Timothy.

The search function on here isn't great, it took me longer than I care to admit to find you this:
OMS question on Physical Standby database
If you need clarification on anything, shout. Like I said, its a pain to do.

Trouble with 10g after Installing Grid Control

Main problem: after installing Grid Control and rebooting, I can no longer access my databases.
I have 10g installed on a fairly decent Windows XP machine. It was running okay, but I had two Oracle database instance processes running at 200mb of RAM each (of a total of 1 GB of RAM), so it was a bit bogged down, as is to be expected when running Oracle as a server on Windows.
I then installed Grid Control, which seemed to go okay, and another database as well. But once the install completed, and I gained access to the Grid Control version of the OEM for the first time, that's when things got wacky. Connecting to any OEM instance started taking anywhere from 2-10 minutes, everything was bogged down considerably, etc. I tried shutting everything down and rebooting, but now things are worse. OEM is timing out and not finding servers, listener is timing out, etc. I've tried manipulating a few things in Net Manager (I noticed that my LISTENER section didn't seem to contain any data at all, and my databases weren't listed in there either) with no result.
I've also had trouble with "host credentials" and being able to administer the databases within the OEM, even after implementing a fix I had found to add my user name to a "log on as batch job" local security policy.
Finally, I noticed that after installing Grid Control, when I run "sqlplus", I now see "Oracle91 Enterprise Edition" instead of the "Oracle10g" it used to show (and I never had anything but 10g installed on this machine!).
Here's what happens when I try to access my dBs:
-When I try to launch the OEM connected to any of the 3 databases I created (which were previsouly running), the browser just times out with "page cannot be displayed".
-When I run "emctl start <ANYTHING>" I get file/path not found errors, and a file path with both forward and backward slashes (starting after the "agent" directory):
C:\oracle\product\10.1.0\agent/<...etc.>
-When I try to test a connection to a service in Net Manager, I get this error:
ORA-12557: TNS:protocol adapter not loadable
I had to manually add my listener info and databases info(under Service Naming) in Net Manager, but everything appeared fine in Oracle Directory Manager (unfortunately, that doesn't help much, since this is just a general info tool).
I am nothing near a DBA and my knowledge is limited, so I apologize for not being more consise, but I'd be glad to elaborate if someone could tell me what they need to know.
Thank you,
John

Yes; thank you both. Although I had 1 GB physical and 1.5 GB virtual memory configured, running Grid Control seemed to completely kill all available resources.
I have removed GC and its related database and applications, and I can again conncet to my databases via Database Control. Thank you.
By the way, another issue has arisen, seemingly in the aftermath of installing GC: my OEM Database Control instances now time out after about 60 seconds, logging me out each and every time. I have cookies enabled, but this doesn't seem to be the trouble. Is there a Web setting somewhere that will kick users out of the instance at timed intervals?

Shared data controls and Ctrl+N

Hi,
From the oracle docs page mentioned below
http://docs.oracle.com/cd/E14571_01/web.1111/b31974/taskflows_parameters.htm#CHDIEDEH
I have a question regarding the line "However, if an end user uses the Ctrl+N keys to open a new browser window, the two windows have the same server-side state and share data control instances." from shared data control instances topic.
If we issue Ctrl+N, it opens a new window and the default being a blank page we would be required to navigate to our application once again. Wouldnt that create a new server session anyways?
Under what circumstance would Ctrl+N reuse the same server session from the old window and share the same instance of data control? Can someone tell me a test case to verify if the new window is using the same data control instance?

Hi,
the doc basically says that you work in a separate data control frame, which means you are in the same session but a new transaction. read this for best practices and some more background information
http://one-size-doesnt-fit-all.blogspot.com/2011/10/pageflowscope-with-unbounded-task-flows.html
Frank

Control instance cretion

Similar Messages

Maybe you are looking for