RE: is the definition of an SQL store procedure stored in the program object? Or what?

["Needles,Stephen J" <SNEEDLES@xxxxxxxxxxxxxxxx>]

We are on 7.1 and the latest CUME...not sure of the TR#.

The difference is that EXTERNAL NAME is different on the "source" machine than it is on the "target" machine.  It can have '*LIBL/program' or 'library/program' and be wrong.  Ideally, we are looking to have them all '*LIBL/program'.

QSYS2 is never replicated between the boxes.

Steve Needles

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of CRPence
Sent: Wednesday, February 20, 2013 11:41 AM
To: midrange-l@xxxxxxxxxxxx
Subject: Re: is the definition of an SQL store procedure stored in the program object? Or what?

   The definition of the Stored Procedure is stored in the Program 
Associated Space [can be seen in DMPOBJ output] of the program object. 
When the object is restored [and that PAS with the definition exists], 
the definition in the catalog should be updated.  The Catalog Tables are 
journaled to QSYS2/QSQJRN so in the same IPL as the recovery, what 
happened to the data is available.

   The specific difference between the SYSROUTINES data on the two 
systems was not described.  Are we to infer that the EXTERNAL_NAME shows 
'*LIBL/thename' on the system where the CREATE PROCEDURE was issued and 
showing 'somelib/thename' on the DR system?  Something else?  The claim 
was that the replication functioned as expected, but there again, 
nothing explicit was mentioned about the EXTERNAL_NAME; we are to assume 
'*LIBL/thename' was replicated correctly as part of restore or create 
actions implemented by the replication software?

   FWiW several releases ago I was very closely involved in correcting 
how the catalogs were being delivered with the OS Extended Base 
OPTION(01), for their installation, and also their handling during DR 
[at least in part, because that is in-part, Database Restore].  The DR 
additionally has the SQL participate as an extra object handler for 
program restore.  When the restore of QSYS2 occurs versus when the 
program objects restore is a nuance that should not matter, but if 
differences exist, that could be relevant.  Beyond knowing a "DR test" 
was done, knowing when QSYS2 was restored relative to other restore 
activity in this specific DR may be helpful.

Regards, Chuck