On Thu, 21 Mar 2013 13:45:05 -0500, Mark Zelden wrote: > >See a thread with subject "JES2 DD Concatenation issue" from >2008 with lots of information. Brian Peterson authored a requirement >to allow DSI to be used for JES2 (HASJES20) in the PPT so PROCLIBs >could be ENQed. I forgot when that happened, but I've been running >that way for years. > >Paul, you even contributed to that thread with a job stream scenario >to do the rename with the ENQs in place. > Thanks for supplying the search keys. There you wrote:
>>On Fri, 28 Mar 2008 09:44:20 -0500, Mark Zelden wrote: >>> >>>... Either way you have to recycle JES2 to pick up the new extents. > >> ... > >How would you ever re-allocate a proclib if it was ENQed? At least >before TSO. Shutdown JES2, then run a batch job to do the rename. >Wait... you can't run a batch job, JES2 is down! > The crucial question. But: Allocate a new DSNAME. Copy. change JESPARMS to reflect the new DSNAME. Recycle JES2. (Repeat if desired to copy back to the original DSNAME.) Tedious, but perhaps preferable to what happened to the OP. It's a long thread; I forget. After HSM recalled to a different volume or different extents, would simply recycling JES2 have set things aright? Any process that reads extents with no ENQ ought to at least ensure that the data set is unmovable and unmigratable. Perhaps a health check. But there are already too many warnings from health checks. "Boy that cried 'wolf!'" >>Just curious: can the PROCLIB concatenation contain PDSEs? > >Yes. Amen. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
