Overview of SDR‎ > ‎ODD‎ > ‎

Submitting ODD jobs from CMS

NOTE: All regularly scheduled IE "production" jobs should be scheduled under the HUM_IRP_BATCH account.

On Demand Data jobs that return data from the Student Data Repository may now be submitted from CMS (aka PS or PeopleSoft).  (In the past, these processes were started from GJAPCTL in Banner.)  Output from these jobs will be sent to the email account of the person who submits the job.  The process involves using a Run Control, so may be familiar to staff who use Run Control IDs for other reasons.

To get started, you will need to log into Campus Solutions and navigate to Humboldt CS Cusomization > HUM – Records & Enrollment > HUM On Demand Data Request.  You may now Search for an existing Run Control ID or Add a new one.  You will only see your own Run Control IDs, so everyone will need to create at least one Run Control in order to submit jobs.  (If you only have a single Run Control ID in your Account, clicking Search will automatically select and open it.)


Setting up a Run Control
To create a new Run Control ID, click on the Add a New Value tab, type in a name for your Run Control, and press Add.  You may create a single Run Control and reuse it each time you want to submit a job.  If you submit several different jobs frequently or on a schedule, you may wish to create a Run Control for each job and name them appropriately.  Once created, a Run Control ID cannot be renamed or deleted, so you may wish to try out the first approach before creating multiple Run Controls.

Submitting a Job to Run
After selecting or adding a Run Control ID, you will see a job setup form.  A Run Control ID that you have used before will remember the last Job Name you selected.  You may either resubmit this job or click on the magnifying glass to see a list of jobs you are allowed to run.  


Entering Parameters
After selecting a job, you will then be shown a list of parameters that apply.  (Note: Term parameters may often be left blank to indicate the current term.)  You may click the magnifying glass next to each parameter to see a list of acceptable values.  Some parameters (e.g. majors) have so many values that you may need to use the Look Up button to find the correct value.  

 
Running the Job
After entering parameter values, click the Run button to submit your job.  If you have trouble entering a value that you think should be correct, your run control may have gotten corrupted.  To fix this, change the job name to a different job and then change it back.

You will now see the Process Scheduler Request. Click OK to submit your job. 
 
Choosing a Time to Run or Repeat
By default, your job will run right away.  You may change the Run Date/Time to run your job later or change the value in Recurrence to run your job on a regular schedule. You probably will need to create a special Run Control ID for any job that you are not running right away.  If you want your job to run automatically on a schedule, keep in mind that the SDR gets refreshed every night and do not schedule your job to run before 7 AM.  Recommended Recurrence values are M-F at 7:30 AM or   Mondays at 7:30 AM.  If you need a different schedule, please send an email to reports@humboldt.edu.

After setting the recurrence, be sure to set the Run Date before pressing OK. By default, PS will stick in a Run Date in the past and your job will then get fired off repeatedly for every day since the run date. This can be VERY annoying. If it happens to you, click the green recurrence arrow in the Process Manager (see next section) and choose Stop. Then kill your job (see below) and try again. 

After submitting your job, you will return to the Requests screen.  The length of time your job takes to run will vary.  You will receive an email when it has completes.

Tracking Your Job
You may click on Process Monitor link (to the left of the Run button) to see the jobs you have submitted, but this will only tell you the progress of the job submission, and not the progress of the job itself.  Click Refresh to update the screen.  Within a few minutes, the Run Status will change to Success and the Distribution Status will change to Posted.  This means that your job has successfully been sent to the SDR and you may now reuse the Run Control ID to submit another job.  Click Go back to Hum ODD Request to return to your Run Control ID.

Killing Repeating Jobs
As it relates to On Demand Data, the Process Monitor is mostly useful to track and change repeating jobs that have a Recurrence set, shown by the green circle-arrow. (You may need to change the number of days to 7 and click Refresh in order to see all your scheduled jobs.) Click the Details link to the right of an Instance that says Queued.  Since all the ODD request jobs will have the same Description (HUM_ODD_AE), you may need to click Parameters to verify that you are looking at the instance you wish to kill.  Click Cancel Request to stop a job that you no longer want to run.  

Forwarding Results to Other Users
ODD datasets are always emailed to the person who submits the request in CMS.  There is no direct way to also send them to other users.
  But the person who submits a dataset can set up filtering/forwarding in gMail to automatically forward a dataset on to others.  Before you can forward email to someone else, they have to give you permission.  Click on the gear in gMail and choose Settings/Forwarding.  Then click Add a forwarding address.  Once someone else has accepted your request to forward, you will be able to create a filter that forwards specific messages to them.

To set up the filter, first run the ODD job.  Then put a check by the output in gMail and choose “Filter messages like this” from the More menu.  Identify the message you want to forward--usually, you will leave the from address alone, add the name of the dataset in the Contains box, and check the Has attachment box.  Click “create filter with this search”.  On the next page, check forward and select an address to forward the message to.  If you want to forward to multiple addresses, you will need to set up multiple filters.  Luckily, they are easy to create.


Comments