|
At 08:20 PM 1/12/98 -0500, you wrote: > > >Al Barsa, Jr. wrote: > >> At 06:48 PM 1/9/98 -0500, you wrote: >> >on 01/09/98at 04:32 PM, our friend Dave Mahadevan <mahadevan@fuse.net> >> >said: >> > >> > >> >>Please remove the restriction on *LDA being only 1024 bytes. Please make >> >>it 1MB. 1024 is too little....:-( >> >>-- >> > >> >Even nicer, if possible, would be several *LDAs, all 1024 long. >> >> The System/38 did not have *LDA initially. S/34 folks conplained, so it >> was added for compatability. > >One of the reasons, S/34 people refused to buy oversold S/38. IBM was forced to >introduce S/36 in 1983 (maybe the LDA did the trick!) > >> >> >> If you need more LDA like space, create a convention to create data areas >> in QTEMP. >> > >Unfortunately, you cannot pass an existing QTEMP to a job being invoked in batch. Dave: Your comment about the *LDA not being passed to a batch program is absolutely valid. When *LDA was implemented (if I remember) it was not passed to submitted programs, this was a latter addition. One circumvention would be dedicate a library to the submission of batch jobs. Submit the batch job, and have the submitted job do some sort of an immediate wait (this is to insure that timing issues prevent it from taking off immediately, before the step below completes). Use the completion message of the submitted job to create a data area in said library with the name DAVE123456, where DAVE is some sort of system defined constant, and 123456 is the job number of the submitted job, putting whatever data you want to pass in the in the data area. Create a shop standard for the length of the data area, but the maximum length is 2000 bytes. This will accomplish passing the data, however it does beg the issue of cleanup of the library. If you are ABSOLUTELY POSITIVE that every batch job will complete before the next IPL, you can use QRPLOBJ. This will make the system clean itself up nicely, as OS/400 doesn't use QRPLOBJ for data areas. (Caution, some vendors do, but this is relatively easy to check.) The problem with QRPLOBJ is that this is a private library, and if you try to make it public, IBM will change it back on you when you least expect it. To use QRPLOBJ, you need to create programs that adopt *ALLOBJ to create the data area and to retrieve it. If this is unpalitable, or if you can't count on the batch jobs completing prior to the next IPL, you need to use your own library. You can put the proper clean up code in your batch CL program, but make sure that it's also in your standard error handling routine. Still this is not foolproof, and will need to be hand checked periodically or with some other more sophisticated processing, based on your own business rules. (For example, if the batch jobs cannot be hanging around before end-of-month, and in your EOM processing, you determine that the job queue is empty and no other batch is processing, clear the library.) Hope this helps. Again, I reiterate, don't look for IBM to increase the size of the LDA. Al Al Barsa, Jr. - Account for Midrange-L Barsa Consulting, LLC. 400 > 390 Phone: 914-251-9400 Fax: 914-251-9406 Private mail should be sent to barsa@ibm.net +--- | This is the Midrange System Mailing List! | To submit a new message, send your mail to "MIDRANGE-L@midrange.com". | To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com. | Questions should be directed to the list owner/operator: david@midrange.com +---
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2025 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.