Yes, the problem *only* occurs ... ... with z/OS V2R1, ... with PDSEs, ... in case at least one empty member exists in the source PDSE of the copy operation (IEBCOPY), irrespective of how the empty member was created.
We do have SMSPDSE and SMSPDSE1 set up and running. "guaranteed synchronous write" was not the culprit. Thanks. Karl -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Bernd Oppolzer Sent: Friday, March 20, 2015 5:10 PM To: [email protected] Subject: Re: Intermittent, not consistently reproducible problems with PDSEs on z/OS V2R1 (incl. infrequent S0F4-20 RSN 1C0752EE ABENDs) Am 20.03.2015 um 17:02 schrieb Henn, Karl: >> Have you tried with - directly after each IEBGENER step - do an IEBGENER >> from the member to SYSOUT ? > Yes. Makes no difference. Same problem > >> Have you tried with only non-empty members? > Yes. No empty members - no problem. > >> Have tried with DD * instead of DD DATA ? > Yes. Makes no difference. Same problem. > >> Have you tried with something other than NULLFILE for empty members? >> (E g an IEBCOPY from a third dataset with empty members.) > Yes. Makes no difference. Same problem. that is: the problem is not limited to IEBGENER ... if you build the empty member with other tools, the problem remains? So this is a PDSE problem ... from your posts I'm almost sure that you don't have this problem with normal PDSs. PDSEs are supported by a special address space, AFAIK. Is there a possibility to have site-specific exits installed in the PDSE supporting subsystem; do you have any? > Thanks. > > Karl > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
