RE: V5R3 Performance Issue -- MIDRANGE-L

Rick:

Based only on what you've said, yes, many (most?) of us have run into this. Not 
only with V5R3 but also with V5R1, V5R2, V4R5, etc., at various times.

Given the condition that no useful indexes have been deleted and similar 
conditions, the answer is almost always "Get current on your DB2 PTFs."

Rob started at the correct point for V5R3 -- WRKPTFGRP.

What might be missed is that Rob is focusing on the DB2 'group' PTF, 
potentially very different from current cumulative PTFs. Without knowing what 
your DB2 group level is, the only advice that's possible is the same SQL 
performance advice that is always given. The archives are loaded with it.

Beyond group PTFs, numerous PTFs are _always_ available from IBM that are 
almost guaranteed not to be on your system regardless of how current it is, 
even up-to-the-minute current. These are commonly for specific environments, 
one of which might match yours. These don't get included in cume PTFs, they 
don't get included in HI-Pers, they don't get included in groups -- they must 
be requested by PTF number. Research through the PSP reports might show they're 
available; often they're only found by reporting a problem to IBM and being 
told they're there.

And sometimes the PTF doesn't exist yet. There could be something unusual in 
the very particular combination of hardware and software and configuration on 
your system that uncovers a bug that no one has reported yet. Or someone else 
might have run into it and developed a workaround, but it just wasn't reported. 
I know of one stupid bug that's been in every release of OS/400 since V3R1; I 
guess no one cares enough to report it. (DMPCLPGM from ILE CL, not OPM CL, will 
print page headings split over two physical pages for every logical page of 
output on most if not all printers. But, big deal.)

Note that answers or valuable hints are often in the PTF cover letters (as well 
as "VxRx Memo to Users"). Sometimes it isn't exactly the PTF but a changed 
behavior of DB2/SQL that can be adapted to. (Application developers _need_ 
access to cover letters and Memos to Users for this reason. That makes internet 
access useful even when authority to DSPPTF is withheld.)

It might be unfortunate, but that's just the way it is.

In your case, you can investigate from an application view point and make the 
decision that it isn't an application problem. Hand it off to operations and 
have them work with IBM to get fixes if needed. Sounds like that's where you 
ended up.

Tom Liotta


midrange-l-request@xxxxxxxxxxxx wrote:

>   7. V5R3 Performance Issue (Rick.Chevalier@xxxxxxxxxxxxxxx)
>
>We installed V5R3 a couple of weeks ago.  Since then one of our nightly 
>critical path jobs has more than doubled in run time.  Operations has been 
>working with another team member on finding the issue but they haven't been 
>successful yet.  It appears to be in an RPG program with embedded SQL.  I know 
>there is at least one other job using QM queries that has also seen a two fold 
>increase in run time.  It's not in the critical path so it isn't an issue at 
>this time.  Has anyone else experienced anything like this with V5R3?

-- 
Tom Liotta
The PowerTech Group, Inc.
19426 68th Avenue South
Kent, WA 98032
Phone  253-872-7788 x313
Fax    253-872-7904
http://www.powertech.com


__________________________________________________________________
Switch to Netscape Internet Service.
As low as $9.95 a month -- Sign up today at http://isp.netscape.com/register

Netscape. Just the Net You Need.

New! Netscape Toolbar for Internet Explorer
Search from anywhere on the Web and block those annoying pop-ups.
Download now at http://channels.netscape.com/ns/search/install.jsp