Apparently there is a bit of both going on here.  HSM knows that the
VSAM components are in fact components of a cluster, but if you do a
LISTCAT of a migrated VSAM dataset, whether you LISTCAT the CLUSTER name
or one of the components, it shows the entry is of a nonvsam dataset
type.  LISTCATting the CLUSTER name doesn't show anything about any of
the underlying components.  

Rex 

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Paul Gilmartin
Sent: Thursday, March 13, 2008 9:18 AM
To: IBM-MAIN@bama.ua.edu
Subject: Re: Easy way to covert IEFBR14 and IDCAMS deletes to HDELETE

On Thu, 13 Mar 2008 04:17:40 -0700, Edward Jaffe wrote:

>Dave Cartwright wrote:
>> No, a VSAM cluster is turned into a flat file on migration and is 
>> cataloged as a NONVSAM dataset on volume MIGRAT.  If its device type 
>> is tape it shows as MIGRAT2, else it's MIGRAT1.
>
>Too bad. That seems like a poor design choice in hindsight given the 
>current topic.
>
Yah.  One would think it would be a single flag bit in either the
catalog or the MCDS well invested.

Why is there a NONVSAM option on HDELETE?  Wouldn't some programmers
find it similarly useful to have NONPDS and NONHFS, etc.?

But, needing to delete several VSAM data sets last night, I HMIGRATED
them first.  I find:

o All 3 components continue to appear in DSLIST, now as volume MIGRAT1.

o An attempt to HDELETE the DATA or INDEX component fails, telling me
  that the operation is prohibited for a component of a VSAM data set.

o HDELETEing the primary entry deletes all 3 components.

So, in deleting a VSAM data set HDELETE apparently exploited information
which this thread earlier alleged is available only after the data set
is recalled.  So, must all VSAM data sets be recalled to be deleted;
and, in fact, all data sets recalled in order to determine whether
they're VSAM on NONVSAM?  I earnestly hope not.  Perhaps inspecting the
first very few blocks suffices to determine whether the data set is
VSAM, in which case HDELETE NONVSAM should be able to exploit a similar
shortcut.

-- gil

----------------------------------------------------------------------
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

Reply via email to