(I posted this to ISPF-L a few days ago and got no replies. But I remain
curious, so I'll try here. Perhaps IBMTCP-L or even MVS-OE would be
better?)
I was mildly surprised that when I create PDSE members via the z/OS 2.2
NFS server, then display them with "ls" on a Solaris workstation, the
Time Changed appears with a granularity of microseconds.
More surprised because:
z/OS
z/OS 2.2.0
ISPF
z/OS V2R2 ISPF Messages and Codes
Diagnostic Tools and Information
Diagnostic information
ISPF statistics entry in a PDS directory
... gives no hint of where this information resides. Does PDSE rather
than ISPF maintain it? The granularity is seconds for a PDS.
And, when I try to edit such a member I am warned, apparently:
ISRE548 -WARNING- Extended statistics exist for this member but will be
Explanation
The site does permit extended statistics to be recorded in the PDS
directory but they already exist.
System programmer response
Use the ISPF configuration dialog (ISPCCONF) to enable Extended Statistics
to be stored.
User response
Contact the Systems Programmer
(copied and pasted with apparent typos)
However, the CHANGED time is updated, still with microsecond granularity and
bit 1 of byte 3 remains set. Doesn't look exactly like OA30217 or OA31634.
It looks as if ISPCCONF sets global options, not controlled by any RACF profile.
Is NFS server breaking a rule? Is Edit breaking a rule by not obliterating the
extended stats as it threatened?
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
