RE: SQL index authority

Thanks, Elvis.

Using QCMDEXC as a stored procedure would work except that I do not want
to hardcode the library name in the GRTOBJAUT cmd. Also, I cannot use the
libl that is active when the RUNSQLSTM cmd is executed. Looks like I'll
just have to put some comments in the source telling the next person to be
sure and execute the GRTOBJAUT after running the RUNSQLSTM cmd. Thanks
for the idea, though.




"Elvis Budimlic" <ebudimlic@xxxxxxxxxxxxxxxxxxxxxxxxx>
Sent by: midrange-l-bounces@xxxxxxxxxxxx
10/03/2007 12:28 PM
Please respond to
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>


To
"'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx>
cc

Subject
RE: SQL index authority






I don't think sql GRANT command is valid against sql INDEX objects. It
doesn't make sense in SQL as you can never reference an index directly.
Only query optimizer is support to use it for alternate access plan
implementation and/or source of real time statistics.

If you're on the other hand using an sql INDEX as a regular keyed LF, I
guess only recourse you have is the GRTOBJAUT command as you've already
discovered.

If you want to stick with RUNSQLSTM compatible statements, you can call
QCMDEXC as a stored procedure and embed the GRTOBJAUT command string in
it.
It should work fine.

Elvis

RPG & DB2 Summit | Minneapolis | October 1-4
Mike Cain - DB2 for i5/OS Temporary Indexes - The Good, The Bad, The Ugly
October 16
2007 System i Fall Technical Conference | Orlando | November 4-7
Celebrating 10-Years of SQL Performance Excellence on IBM System i,
eServer
iSeries and the server affectionately known as the AS/400

-----Original Message-----
Subject: SQL index authority

I have replaced a couple LFs with SQL indexes. However, I can't seem to
figure out how to grant authority to the indexes as I would using
GRTOBJAUT. I know how to do it with tables and views using the GRANT
stmt, but not indexes. I want the equivalent of GRTOBJAUT OBJ(INDEXA)
OBJTYPE(*FILE) USER(*PUBLIC) AUT(*CHANGE). Or am I looking at this in the

wrong way? Thanks.

Michael Kirkpatrick