On Wednesday, 05/20/2009 at 10:01 EDT, David Boyes <[email protected]> 
wrote:
> On 5/20/09 4:29 AM, "jose raul baron" <[email protected]> wrote:
> 
> > - Does VSAM still exist in z/VM 5.4 ?
> 
> > - Is it perhaps included in the z/VM 5.4 code (e.g. as TCPIP) ?
> 
> No, it's under the same terms (and prices) as of old.

- The CMS VSAM feature of VSE/VSAM is no longer available. If you don't 
have it, you can't get it.
- Someone who already has it cannot give it to you.
- If you already have it, you can keep using it as long as you keep paying 
the monthly license charge.
- It isn't licensed for use on IFLs.
- If it has become incompatible with current z/VSE VSAM support, all you 
will receive from us are notes of sympathy and regret.
- OTOH, we haven't consciously done anything to break it.  If *z/VM* 
changes something that breaks *it*, we would likely undo whatever broke it 
if that doesn't in turn break something else.  (i.e. the reason for the 
change in the first place.)

I can't believe that no one remembers the SPE from around 1985 called the 
"Alternate VSAM Emulator" that enables a program to intercept the VSAM 
open/close/tclose calls.  That means the program provides the addresses of 
the read & write entry points as well since the program fills in the ACB. 
See 
http://publib.boulder.ibm.com/infocenter/zvm/v5r3/index.jsp?topic=/com.ibm.zvm.v53.dmsa5/hcsd2b00628.htm.

It was originally exploited by DB2 (SQL/DS) on VM, allowing it to emulate 
VSAM.

I can envision that the VSAM calls and data could be exported via a 
custom-built "connector" to z/VSE or z/OS.  Or Linux.  There are those who 
might be willing to pay someone to provide a solution that would leave 
their programs intact but sever the connection to CMS/VSAM.  I wonder: 
Perhaps the emulator could simply use the BPX1xxxx POSIX routines to mount 
remote NFS server and read/write data that way.

Alan Altmark
z/VM Development
IBM Endicott

Reply via email to