Re: File Last Changed -- MIDRANGE-L

Me? No. Did the Cume install? I don't know. But in that regard (i.e., if the Cume did do some kind of change) I do find it odd that there are *File objects in another library with Change dates going back to 2006.
From the Info Center: "To detect an unused object, look at both the last-used date and the last-changed date. Change commands do not update the last-used date unless the commands cause the object to be deleted and created again, or the change operation causes the object to be read as a part of the change." It doesn't go into specifics, but I figured that 'change commands' were things like CHGOBJOWN, CHGOBJD, and GRTOBJAUT.

* Jerry C. Adams
*IBM System i Programmer/Analyst
B&W Wholesale Distributors, Inc.* *
voice
615.995.7024
fax
615.995.1201
email
jerry@xxxxxxxxxxxxxxx <mailto:jerry@xxxxxxxxxxxxxxx>

Glenn Birnbaum wrote:

If you change an attribute (i.e. authority, description, ownership,
etc.), it updates the change date, but not the used date, since you
really didn't use it.

Did you do a mass change on that Saturday of authority, start
journaling, change owners, or whatever......

HTH,
Glenn

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Jerry Adams
Sent: Thursday, November 15, 2007 1:06 PM
To: Midrange-L
Subject: File Last Changed

I have been trying to analyze the *File objects in a library to determine if I really, really need to keep them. One of the tools I'm using on the questionable ones (i.e., no one claims responsibility and I

can't find anything that references them) is DSPOBJD.

The weird part (to me) is that a lot of them have a "Last Changed Date" of 26 May 2007 (a Saturday) at about 20:00. I thought that might be because of a PTF (Cume or Group) because I would have done that on a Saturday; in fact, my time log indicates that I did work that day and, other than a natural disaster, applying PTF's is about the only reason that I would have worked on a Saturday. However, I know applied a CUME and Groups to that system on 20 October 2007 (a Saturday, naturally); so

that idea doesn't seem to hold water.

And most of those files exhibit a "Last Used Date" earlier (sometimes by

a couple of years) than the "Last Changed Date". But, if I (or anyone) actually changed the file in any way, then the "Last Used Date" should get updated, too. Right? Well, that's what my tests (simple DFU on a file) showed, anyway. A simple query against a file affects the "Last Used" but, naturally, not the "Last Changed" date.

I have not been able to find anything in the Info Center (V5R4) or the Archives that would explain why the "Last Changed Date" would be later than the "Last Used Date." Anyone on the List know (or simply want to hazard a SWAG)?