If a user has to do an EXPORT, SET SYMBOL=xxx That takes effort
If they can code //DDLBL DD DSN=D&DDMMYY..T&HHMMSS Then much less effort. And the other concerns presented are still true 1) Where is the JCL converted 2) Is the JCL converted on the submitting system or a different system? 3) How is the JCL Converted 4) Are the symbols in the system where the JCL is converted? 5) When the job is submitted it will convert the symbolic with the time of the conversion, but if it runs much later (hours) then the dataset name will not necessarily reflect when it was actually created. Example, JOBA is submitted at 000000 but is on Hold. It is later released at 020000. The DSN would be D&ddmmyy..T000000. User will assume the dataset was created at Midnight and not 02am when the job actually ran. I think the bottom line is people very familiar with the converter, symbol usage, and how things work will not have issues with the process. End users not as familiar will be creating lots of datasets with expectations that might not be valid. And even though we can migrate off datasets not used, what process will "know" when that file is no longer needed by the application and then can be deleted. I think on the open system/midrange/server (not mainframe server) side are used to adding Dates and Times to files with the expectation it is the actual date and time the file was created. I do not know if they have a clean-up process once those files are no longer needed. Lizette > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of Steve Smith > Sent: Friday, May 05, 2017 9:37 AM > To: [email protected] > Subject: Re: JCL System symbols > > @Lizette: If that's your only concern, then you have no concerns. There's no > connection between JCL symbols and users creating lots of datasets. > > What I do is use an IEBEDIT (IEBGENER also fine) job to submit the actual job > from //SYSUT1 DD DATA,SYMBOLS=EXECSYS to the INTRDR. Voila, full symbol > support. And there's not a damn thing you can do about it. :-) > > sas > > On Fri, May 5, 2017 at 12:06 PM, Lizette Koehler <[email protected]> > wrote: > > > Only concern would be users creating datasets with DATE/TIME entries > > and there is no cleanup. > > > > So you could have a user create 365 Date datasets * hhmmss entries and > > then not have a process to DELETE the things no longer needed. > > > > I will use the inappropriate term - ga-zillions of datasets in your > > DASD Volumes and no clean up. > > > > > > > > Lizette > > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
