On Mon, 20 Apr 2009 14:24:14 -0700, Ron Hawkins wrote:
>
>This was the source of some incredible HSM thrashing a few years ago when a
>group of users decided they did not want any of their datasets migrated. The
>clever dicks would run an IEFBR14 to allocate them every morning, and HSM
>would happily migrate them again every night.
>
Gulp.
So it was recalled because it was allocated, but not marked accessed
because it wan't opened.
Better, in Rexx, e.g.:
EXECIO 0 DISKR SYSUT1 (OPEN FINIS
... repeat over ddnames until weary.
I'd strongly counsel the OP to change her HSM rules rather than
rely on pretending to access the data set.
But I've done the latter, long ago when I didn't trust HSM.
I no longer do so -- one of us has matured.
But there remains a point of irritation. We don't run HSM on
all our systems: some don't mount the needed volumes or are
test systems not qualified for HSM. So I'm logged on to a
test system and attempt to allocate a data set. I get the
message that it's migrated and I should start HSM. Not my job.
I log on to a production system and recall the data set, and
back on the test system reply RETRY. It tells me to start
HSM. You *** moron DYNALLOC! Look around you! The data set
is back; no need for HSM!
Bad design. Sigh.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
