I should state that the overflow log path would be the path specified
in the logtarget of the restore comand

So is NORETRIEVE required in that case?

I don't have an exact answer to that question, but if you don't want to use
all the logs in the logpath, I would only rollforward to the minimum
rollforward time (which will use only the necessary included logs). To find
the minimum rollforward time, use a bogus time such as
1950-01-01-00:00:00.000000 (not sure about that format) and then DB2 will
tell you that it cannot use that time and give you the minimum rollforward
time. Then substitute the minimum rollforward time DB2 gives you and
resubmit the rollforward (with the NORETREIVE option). This will prevent
using all the more recent logs in the logpath.

Scott:
Yes, it took me a while to get my brain wrapped around this...

Assuming you want to roll-forward only the logs stored w/ the online DB
backup:

Do the backup like so:
-backup database $db online to <wherever> include logs

Then:
-restore db $db taken at $pit logtarget /tmp
-db2 rollforward db $db to end of logs and stop overflow log path /tmp
noretrieve

$pit is the point in time referenced in the online backup file, ie:
<DB>.0.<instance>.NODE0000.CATN0000.20071228021205.001 would be
0071228021205

I happen to be using /tmp to store the logs that are extracted from
the backup, choose whatever dir you fancy.

I'm not familar w/ AIX, but on Linux this script will automate it.  Feed
it the name of the database you want to restore+roll forward as the arg.

db=`echo $1 | tr '[a-z]' '[A-Z]'`
pit=`ls $<backup location>/$db.0.*.001 | awk -F. '{print $6}'`
cd $<backup location>
echo -e "Restoring database $db...\n"
db2 restore db $db taken at $pit logtarget /tmp
db2 rollforward db $db to end of logs and stop overflow log path
\(/tmp\) noretrieve

Careful w/ the left and right parens - they will trip up the shell
interpreter.  Notice how I escaped them..

Good luck.  HTH

cheers
aj