Session 3081 - Tips for Implementing New DFSMSrmm Function

SHARE 90
February 22-27, 1998


I think I've talked to Nadine Hart before, when I placed a call to RMM level-2. She was really pleasant on the phone when I had to deal with her, and she seemed to know her material here.

RMM is the "Removable Media Manager" - what most of us would simply call the tape library. We've been running it at Duda since it became available, replacing a homegrown tape library written by Bruce Pocock and David Best.

Nadine Hart talked about some of the new features in the new release of RMM. By inchmeal, RMM is becoming more and more usable to the vast legions of CA tape library users out there, and I heard more than one query from TMS users at SHARE who were anxious to dump CA.

RMM now ships some sample reporting code, written for DFSORT's ICETOOL report writer. She emphasized that the sample code was actually no fooling supported, but only as written. If you modify it, it belongs to you she cautioned.

There is a new "verify" function that produces a simulated housekeeping run. After any change to installation retention policy, DFRMM now requires that you execute a verify run before doing regular housekeeping (this can be overridden, but it is frowned upon).

RMM housekeeping performance has been improved substantially, with the use of VSAM batch LSR. This becomes available in OW25329 (but watch out for VSAM APARS OW24364 and OW26054!)

Nadine commented that lots of new function is being introduced through the APAR mechanism. Because RMM is new, she explained, they can get away with this. After RMM stabilizes in a couple of years, they will move more and more toward only implementing new function in new releases.

There is a new API for RMM control. It isn't generally available just yet... but will be within a month. Basically anything you can do in the TSO commands for RMM can be done through the API. You can request output in structured fields, or as plain text, though the latter can change in format from release to release. The API is multithreaded, and you can keep several conversations going through use of TOKEN parameters.


Back to session index
Back to index of SHARE meetings
Read the disclaimer