OK, to answer all your questions:

1. I am able to debug the code and I can confirm that the parameters are correctly set up.
2. When I retrieve the working copy, I do request the full state.
3. I am not able to verify the savestate object that the operation returns because when saveWorkItem is invoked, the participant is called again and then again and again in a recursive way until the server throws an internal error that has no connection to the save of the workitem but rather to the way too many recursive calls that are going on.

Now, you will ask why am I allowing the infinite recursive calls to go on. As a matter of fact, my participant has a guard against this condition and it checks for the workitem state. If the state changes, then the second call ends before anything else is executed. However, the 2nd call shows that the state of the workitem is not changed, so the guard is not triggered.

Since nothing else changes in the workitem, I do not have any other attribute to check for in the guard to avoid recursive calls.

You want to use saveWorkItem3 and use the technique described here: https://rsjazz.wordpress.com/2012/07/31/rtc-update-parent-duration-estimation-and-effort-participant/ (see Communication Between Operations to Avoid Recursions) to prevent the recursion from happening. I think the same code is in the other downloads, but commented out.

For completeness, another technique to prevent recursion when triggering on state changes is to compare the workflow state of the oldState and the workflow state of the newState and exit, if they are the same (no state change happened). But there are cases when this might not be appropriate and you want to use saveWorkItem3 and communication between the participants and advisors.

Hello Ralph,

I did use your recommended solution and now I am invoking saveWorkItem3 with the additional parameter that prevents recursion from happening.

Now I can say that the operation does not report an error, in the ccm.log file there is no error, but the state of the item is not changed.

HELP!

Dan

Dan, as long as I have no indicators what could be wrong, I am unable to help. It has worked for me in the past I can't tell what you are doing wrong.

What kind of information could I get from RTC and give you that would help you understand what might be wrong?

Thanks,
Dan

I would expect to see at least an exception if something went wrong - transition not applicable, and the save status should show the save failed. Like in the code I publish:

    IStatus saveStatus = fWorkItemServer.saveWorkItem3(parentworkingcopy, null, null,   additionalParams);
if (!saveStatus.isOK()) {
        // generate an error or do it in the calling method
        return false;
}

My code does the following:

Set<String> additionalParams = new HashSet<String>();
additionalParams.add(ManageServiceRequestStateParticipant.SERVICE_REQUEST_STATE_ID);
IStatus saveStatus = fWorkItemServer.saveWorkItem3(workItemWorkingCopy, null, workflowAction, additionalParams);

if (!saveStatus.isOK())
{
     IReportInfo info = collector.createInfo("Unable to change workitem state", "Error saving workitem with id : " + workItemWorkingCopy.getId());
     info.setSeverity(IProcessReport.ERROR);
     collector.addInfo(info);
}

So, if anything would be wrong with the execution of the saveWorkItem call, I would see the error message in the UI.

That does not happen.