Re: Big disappointment with Postgres

From: joel garry <>
Date: Mon, 7 Feb 2011 09:28:58 -0800 (PST)
Message-ID: <>

On Feb 5, 4:38 am, Noons <> wrote:
> Mladen Gogala wrote,on my timestamp of 5/02/2011 1:58 AM:
> > "Optimizer hints (not wanted)
> >      Optimizer hints are used to work around problems in the optimizer and
> > introduce upgrade and maintenance issues.
> Beg your pardon?  So, a hint that fixes a plan "introduces upgrade and
> maintenance issues", but a patch to fix a specific problem with the optimizer
> does not introduce such?
> Hellooooooo?

I think the relevant problem is reliance on hints. Tom Kyte's real feeling about hints is "See Jonathan:" I think we all knew that, but it's always fun to pick on noteworthy absolute statements.

> Has any of these idiots EVER had to upgrade an existing production system?  Ah
> yes, I forgot this is the open community: they don't give a fig about existing
> production, they just "rework" or "refactor" whenever needed.
> Nice racket if you can make it stick...
>  > We would rather have the
> > problems reported and fixed. We have discussed a more sophisticated
> > system of per-class cost adjustment instead, but a specification remains
> > to be developed."
> Ah, ok: the "too hard basket".  Of course.  IOW: we'll find a sucker to go and
> investigate that one of these days.
> > In essence, when I encounter a problem, I am supposed to contact one of
> > the gurus and have him develop a "custom patch" for me. Nice little
> > racket.
> LOL!   Narh, perish the thought!
> As if that hasn't been the MO of that lot since day one...

I may have missed something, but isn't the enterprisedb product supposed to address this?

I get what Mladen is saying about the developers and the racket - I'd even say it derives straight from the bazaar doc - I get what you are saying about "that lot," but I don't quite get why it doesn't count that a commercial solution for the exact problem exists. Fast, good, cheap, pick two, you know?


-- is bogus.
Received on Mon Feb 07 2011 - 11:28:58 CST

Original text of this message