On Tue, 8 Oct 2019 16:28:30 +0000, Jesse 1 Robinson wrote:
>This is a classic problem endemic to traditional PDS since the dawn of MVS.
>IBM's solution to this problem is PDSE. If you convert the data set to Library
>(PDSE) you may reduce problem occurrence to negligible. OTOH PDSE may
>introduce its own set of problems, chief of which is the inability to share
>the data set across sysplex boundaries. If that restriction is unacceptable,
>you might try adding in your RYO enqueue program as you suggest. Be aware,
>however, that DISP=SHR is unlikely to provide much if any protection against
>concurrent update.
>
I have used ISPF LM services for this purpose. But that requires a TSO
environment. Otherwise, you may simulate the ISPF technique, as in:
https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.4.0/com.ibm.zos.v2r4.f54pc00/isppcmmnamen.htm
Whatever you do, all players must follow the same rules. Any
nonconformists must be re-coded.
I have tested using LMPUT to write multiple members concurrently.
It works, but only within a single job because ISPF LMPUT reqiuires
ENQ EXC on the DSN.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
