IDMS

  • 1.  Any Issues with CA-IDMS R19.0 and COBOL V6R2 on z/OS?

    Posted Jun 19, 2018 10:13 AM

    We are converting all of our load libraries to PDS-E format. Anything else needed for CA-IDMS? Particularly, DC COBOL.



  • 2.  Re: Any Issues with CA-IDMS R19.0 and COBOL V6R2 on z/OS?

    Posted Jun 19, 2018 10:56 AM

    Oh, we're especially interested in DC-COBOL as we have quite a few executables that run online in the CV. We have our 31-bit reentrant program pools tuned to near perfection where we only do one load per week each and every program. So we will be watching for any increase in overall load module size.



  • 3.  Re: Any Issues with CA-IDMS R19.0 and COBOL V6R2 on z/OS?
    Best Answer

    Broadcom Employee


  • 4.  Re: Any Issues with CA-IDMS R19.0 and COBOL V6R2 on z/OS?

    Broadcom Employee
    Posted Jun 19, 2018 11:10 AM

    The only thing I found related to COBOL release 6 or later is PTF SO00266

     

    PROBLEM DESCRIPTION:                                                   
    A new cobol compiler warning IGYCB7310 for newer versions of COBOL     
    starting with 6.1 can cause the CSM install process to halt due to a   
    return code of 4 from the EMPLOAD step. This correction does not       
    alter program functionality. The changes implemented will ensure a zero
    return code when compiling with a 6.1 or newer version of COBOL.       

     

    https://support.ca.com/us/download-center/solution-detail.html?docid=653258&os=OS&aparno=SO00266 



  • 5.  Re: Any Issues with CA-IDMS R19.0 and COBOL V6R2 on z/OS?

    Broadcom Employee
    Posted Jun 20, 2018 08:07 AM

    Jim,

     

    Discussed with SE/DEV and we recently encountered a client who went to Cobol release 6 and we discovered that additional Cobol runtime modules were loaded so the resolution was to increase the XA REENTRANT POOL by 20%



  • 6.  Re: Any Issues with CA-IDMS R19.0 and COBOL V6R2 on z/OS?

    Posted Jun 20, 2018 08:18 AM

    Brian,

     

    That is good to know. We are monitoring any changes in the size of OUR DC-COBOL load modules. We are seeing some differences, mostly increased. The compiler itself seems to have gone under a major upgrade and there are new compile-time switches. We're using the technique where the only PARM= value we pipe-in to the compiler at compile-time is OPTFILE. Tells compiler to look to DD name SYSOPTF for the real compile-time switches.

     

    We have our 31-bit reentrant pools sized optimally. Our CV life cycle is 7-days with a pretty heavy-duty group of online users and programs. Sunday-to-Sunday. By Wednesday, usually all programs have been loaded in all CVs. And loaded only once.

     

    James B. Moore

    (317) 212-9130 (DSN 699)

    James.b.moore238.ctr@mail.mil

    DCPS CA-IDMS DBA - Security+ Certified