Re: rman recover/restore from backup

"Richard Foote" <richard.foote_at_bigpond.com> wrote in message news:<hPZ_b.75578

Inspired by your advice, Richard, I actually wasted precious bandwidth on downloading dear Don's popwerpoint material for RMOUG. Jeez, that boy's written 17 books according to his powerpoint. Must just churn 'em out. And what a lot of work for the quality control department.

I nearly fell of my chair when the picture of dear Don with his seeing eye horse in the aeroplane appeared.

But I recovered my composure until (9 slides in) we get to the very first slide that has anything factual to say about Oracle. Except it's not factual: "ALL Oracle parameters can now be dynamically modified! No more init.ora file". With a slight sinking feeling, I ploughed on. Apparently there's no such thing as the PGA in 9i.

I particular enjoyed this one: "Indexes love large blocks because the b-trees are better balanced". Uh huh.

Sure enough, on slide 23, we have the infamous "db_keep_cache - a sub-pool of db_cache_size". The error repeated for db_recycle_cache.

On slide 27 he appears to have forgotten that db_cache_advice does NOT report predictions on 10% of the current cache size if 10% of the current size would be an illegal value (ie, less than 1 granule).

Astonishingly, on slide 62 (this must have been some presentation!) he says that a low library cache hit ratio means increase the size of your shared pool, with not a whisper of the magic words "bind variables".

I just love the way it takes 72 slides to get across the message that
"automatic tuning" in 9i is so easy! Some new meaning of the word
"automatic", I am sure.

It would be pathetic if it was meant to be taken seriously, but I think I discerned dear Don's true purpose with this material. It's obviously a script for a stand-up Comic Relief red nose day skit. And as such, it is a work of art. Ron would be proud.

Regards
HJR Received on Wed Feb 25 2004 - 12:42:20 CST