From: | Greg Stark <gsstark(at)mit(dot)edu> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: replacements for vacuum? |
Date: | 2004-12-18 07:25:43 |
Message-ID: | 87k6rggj9k.fsf@stark.xeocode.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:
> Lonni J Friedman <netllama(at)gmail(dot)com> writes:
>
> > I'd like to be able to run vacuum in a 'test' or read-only mode where
> > i'd see what it would do before actually running it.
>
> Er ... what possible value would that have? ISTM it would expend 80% of
> the effort to achieve 0% of the result.
Just a guess, maybe you mean "analyze" when you say "vacuum"? People often
conflate them since they often run both together with "vacuum analyze". But
there wouldn't be much point in running a test vacuum, they're might be some
point in running a test analyze.
If so, one little known feature: you can run analyze inside a transaction. The
new statistics are only used by that session until you commit. I started a
script to explain a set of queries, run analyze, then re-explain the queries
and compare the plans before either committing or rolling back. I think it
would be a useful DBA tool for a high availability production system, but I
haven't finished it.
--
greg
From | Date | Subject | |
---|---|---|---|
Next Message | Michael Fuhr | 2004-12-18 08:27:15 | Re: Multi row sequence? |
Previous Message | Greg Stark | 2004-12-18 07:20:41 | Re: Scheduler in Postgres |