Re: Exact use of Log file - why so big?

Hi,

Look up "Recovery Model" in the docs. You describe the "Simple" model, but
the default is "Full"

Cheers
Doug Forster

"Sky" <public.skysigal{*AT*}xact-solutions.com> wrote in message
news:7D4573A7-C59D-49C1-9EDA-E529A377F4CC@xxxxxxxxxxxxxxxx
Coming from a background of simpler file access, and simple databases for
that matter, I'm having a little trouble wrapping my mind around the log
file
concept.

If I understand this correctly, a log file, is a queue of pending
operations
writen to *before* writting to the database, and cleared *after* writting
to
the db, so that if there is a sudden shut down and restart, that pending
operations can be written to the database.

What I don't understand is why it remains biggish - or even there.
a) If the log file queue is processed, shouldn't the file size go back to
0?
I mean, all operations that were in it are now in the db, right?
b) And if filesize 0, why not delete it?

It just seems, from a newbie's point of view, a royal distraction....its
the
db we want, no, and once the log file has caught up with it/been zeroed
out,
why does it remain important?

Thanks for any help with understanding its importance, and its proper
place
in the role of things.

Sky


.

Loading