ForecastJobs() - optional List of UniqueIds is mandatory?

Hi everyone,

Build CPS 8.0 M33.104

It all started with not being able to export the job forecast in a csv format. Please let me know if it is possible.

Now i wanted to utilize the JobForecaster class, however i am facing a serious of problems:

My code snippet:

java.util.TimeZone tz = java.util.TimeZone.getTimeZone("PST8PDT");
DateTimeZone begin = new DateTimeZone(tz,2015,3,14,1,0,0,0);
DateTimeZone end = new DateTimeZone(tz,2015,3,14,2,0,0,0);
JobForecaster jfc = new JobForecaster();
List l = new ArrayList();
l.add(Long.valueOf(50774835));
jcsOut.println(l.size());
Iterator it = jfc.forecastJobs(jcsSession,begine,end,l);
while(it.hasNext())
{
ForecastJob fcjob = (ForecastJob)it.next();
DateTimeZone estdtz = fcjob.getExpectedStartTime();
DateTimeZone eetdtz = fcjob.getExpectedEndTime();
jcsOut.println(fcjob.getJobDefinition().getName()+","+estdtz.toString()+","+eetdtz.toString());
}

1. as per documentation, List jobUniqueIds is optional, however if i don't pass it, i get the error:

symbol : method forecastJobs(com.redwood.scheduler.api.model.SchedulerSession,com.redwood.scheduler.api.date.DateTimeZone,com.redwood.scheduler.api.date.DateTimeZone)

2. so i created a list and added one element 50774835, which is a unique id of a job that runs every 15 minutes.(above snippet),

Caused by: java.lang.NullPointerException: while trying to invoke the method com.redwood.scheduler.api.model.JobDefinition.getName() of a null object returned from com.redwood.scheduler.api.model.ForecastJob.getJobDefinition()

3. when  i removed the jobdefinition part from line 15 above, i am able to print the estimated start time/end time/runtime (i assume in milliseconds), however i do not see an option to validate whether i am forecasting the job i want, because as per the results, the job had an estimated run time of a minutes, however, the average run time of the job is just 6.5 seconds with a standard deviation of 13.5 seconds.

4. that brings to me to my main question, how do forecast all jobs that fall withing a time frame, if i have to pass the list of uniqueids as a mandatory value?

any help is appreciated.

I do understand that forecast is not persisted and resource intensive, but how else can we build a dynamic forecast based on the most latest runtime statistics.

thanks

Nanda