The following JCL will create a PAX archive file in a disk data set. If
this disk data set is on shared DASD, it can be used (if catalogued) to
restore the directory on the second system.
//STEP010 EXEC PGM=IKJEFT01,
// REGION=4096K,
// DYNAMNBR=40
//SYSTSPRT DD SYSOUT=*
//SYSEXEC DD DISP=SHR,DSN=SYS1.SBPXEXEC
//SYSTSIN DD *
LISTC ENT('TSH009.HOME.XFER.Z')
OSHELL cd /home/tsh009; +
pax -W "seqparms='blksize=27998,recfm=u,space=(cyl,(100,10))'" +
-wvzf "//tsh009.home.xfer.z" .
LISTC ENT('TSH009.HOME.XFER.Z')
/*
//
Note that the pax command used __does not__ prefix the DSN specified in the
parameters with the user's TSO profile! The JCL below shows a method to
restore this on the second system.
//STEP010 EXEC PGM=IKJEFT01,
// REGION=4096K,
// DYNAMNBR=40
//SYSTSPRT DD SYSOUT=*
//SYSEXEC DD DISP=SHR,DSN=SYS1.SBPXEXEC
//SYSTSIN DD *
MKDIR /some/directory MODE(7,5,5)
OSHELL cd /some/directory; +
pax -rvzf "//tsh009.home.xfer.z"
/*
//
On Wed, Oct 16, 2013 at 10:45 AM, Paul Gilmartin <[email protected]>wrote:
> On Wed, 16 Oct 2013 09:22:25 -0500, Ron Wells wrote:
>
> >Thanks...but in Diff lpar..
> >I can..because it is diff. named uss file...mount it...but was wanting to
> >get away from extra steps.
> >thought there was a simple..way to backup..from Dir..file/sub dir..to a
> >fiole then upload it back to another/other uss file system..
> >
> That may just regress the problem: What putative backup file can be
> mounted on both systems? Tape, certainly.
>
> pax (tar on steroids) can create the backup file you need (that's what
> motivated its invention), but then you need to move it to the other
> system.
>
> Mark mentioned "copytree" for its facilities. <RANT> Why did IBM feel
> the need to take something as simple and elegant like UNIX and taint
> it with extended attributes and such? They just couldn't resist making
> UNIX resemble MVS?
>
> "tar" is an acronym for Tape Archive. It's beyond ironic that z/OS's tar
> doesn't support tapes.</RANT>
>
> -- gil
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN
>
--
This is clearly another case of too many mad scientists, and not enough
hunchbacks.
Maranatha! <><
John McKown
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
