One possibility would be to use 3.4. After listing the datasets that match the mask, you can sort on volume and use a block command to process just the migrated ones.
Another possibility would be to use the LMDDISP ISPF service. It supports masks just like 3.4. Both of the above would support a data set list filter so you only see the migrated datasets and don't have to perform any extra checks. Five thousand at a time? How many ML2 volume mounts does this require? (If ML2 is on tape, you really want to issue the commands as fast as possible so HSM can optimize the mounts and tape travel.) How long does it take? It's probably better now but on an old system I supported HSM would drive the CPU usage to 100% until it finished the restores. > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of Lizette Koehler > Sent: Thursday, May 21, 2015 11:13 AM > To: [email protected] > Subject: HSM Recalls and MASKING > > I am fairly certain I know this answer, but thought I would ask it anyway. > > We have a need from time to time to do a large number of recalls (5000 or > more) at a time. > > The datasets have a fairly consistent naming convention. But it looks like > the RECALL > command does not use masks but requires the entire name filled in. > > I am thinking of creating a CSI (catalog search) function to take the mask > and build the > commands. > > Just wondering if I going to the right path or not. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
