I remember the issue of the log size limit coming up a while back.
I thought that limit was supposed to be lifted in some release of V3?
Isn't IBM aware that the limit is an issue for large sites?
(Although you are one of the largest I remember reading about)
We only need a 3GB log so far (for a 10GB data base), but I'll eventually
catch up to you!

> ----Original Message-----
> From: Steven P Roder [SMTP:tkssteve AT REXX.ACSU.BUFFALO DOT EDU]
> Sent: Wednesday, April 21, 1999 10:23 AM
> To:   ADSM-L AT VM.MARIST DOT EDU
> Subject:      Re: Internal error LOGSEG871
>
> On Wed, 21 Apr 1999, Richard Sims wrote:
> > Steve - The way that I approach DB backups is to run them daily: a full
> backup
> >         at 23:00, before the nightly backup cycle; and an incremental in
> the
> > morning after the backups complete but before first shift begins.  The
> > incrementals are fast and so not intrusive, so could readily be
> performed more
> > often if needed.  I should think you'd be able to do something similar
> to
> > periodically reset your Recovery Log and thus live happily with
> Roll-Forward,
> > avoiding triggered backups.
>
> Yes, this would work, if it didn't take 3-4 hours for a full.  My DB is
> 45GB (8.2 million pages).  A triggered incremental of 4-5GB worth of log
> entires can take a long time when clients are backing up.  I would have to
> run an incremental just before backups, and then perhaps another one after
> backups (before migration/copy pool/expiration).  I have a 7-tape DLT
> stacker for DBBackups, and then managing the tapes in it becomes an
> operational issue.  If the Log could be larger, so I only had to run one
> dbbackup/day, things work our more nicely....
>
> In a perfect world....
>
> Steve (unVMix Systems Programmer/Dude) Roder
> (tkssteve AT ubvm.cc.buffalo DOT edu | tkssteve AT acsu.buffalo DOT edu | 
> (716)645-3564
> ,
>    | http://ubvm.cc.buffalo.edu/~tkssteve)