Brian, An interesting question. The data sets were placed in a NOMIG MGMTCLAS for a reason, and that original reason may be lost. Regardless, the data sets were migrated through user initiated command migration to get them to ML1. The data sets have not been recalled, which begs the question, are they still needed? First, look to the owner of the data. The data belongs to an application group or a BU at your location. Understanding whether the data is required from a Sarbanes-Oxley or other regulatory requirement will determine if the data is needed for the long term. Once you have determined the usefulness of the data, change the MGMTCLAS to an appropriate MGMTCLAS befitting the type of data it is. Once the MGMTMCLAS is changed, update your ACS routines to reflect the MGMTCLAS changes. The ACS changes will keep DFSMS in sync with what you have performed manually. If you change the ACS code first, the recall of the data set will redrive the MGMTCLAS and you will not have to manually perform an IDCAMS ALTER.
Michael Spencer BMC Software -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Brian Westerman Sent: Monday, March 30, 2009 4:11 AM To: [email protected] Subject: ML1 datasets in limbo Hi, I have some datasets on ML1 that have been there for apparently several years without going to ML2. Nothing really big and only about 30 of them. It looks like they are datasets that were not supposed to have migrated at all (in our NOMIG MGMTCLS), but because of some space management issues a few years back, they were migrated to ML1, and they have just sat there without being used. I realize that this points to a underlying problem that they should NOT have been in NOMIG if no one is using he for that long, but it brings up a question. What is the best way to handle them? A) Hrecall them back to DASD where they will obviously just sit forever. B) force them to go to ML2 (where they probably belong). I'm kind of wondering what other sites do in this kind of case. I am leaning towards "B", but I'm not sure if the best way to handle it is via FREEVOL or MIGRATE ML1, (if that would even work). Would it be better to recall them and change the MGMTCLAS to something else? I would like to establish a rule of thumb for these types of datasets, but I wanted to get some input to see what other sites were doing to handle this type of issue. Thanks for your input. Brian ---------------------------------------------------------------------- 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 ---------------------------------------------------------------------- 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
