From: | Robert Treat <rtreat(at)webmd(dot)net> |
---|---|
To: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
Cc: | "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: db grows and grows |
Date: | 2002-06-20 19:57:38 |
Message-ID: | 1024603058.24091.129.camel@camel |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Thu, 2002-06-20 at 11:40, Bruce Momjian wrote:
> Bjoern Metzdorf wrote:
>
> Yes, but the problem is that we don't have a plan on how to fix the
> index growth problem right now, so if people want to prevent index
> growth, reindex is the only solution.
>
> The TODO item says:
>
> * Certain indexes will not shrink, e.g. indexes on ever-increasing
> columns and indexes with many duplicate keys
>
> but in fact that wording is misleading. >80% of index are on
> ever-increasing columns, so it isn't really 'certain index' but more
> accurately 'most indexes'.
>
> I am planning to add the reindex script to /contrib, document its need
> in the maintenance docs, and add an FAQ item. If it gets fixed in 7.3,
> great. If not, we will have communicated to users and given them the
> tools then need.
>
Would you say the reindex command/script is the recommended way of
dealing with the issue, rather than the create/rename method some have
suggested? Or maybe the difference is negligible? Does either method
have an upside in regards to the query planner statistics generated via
routine vacuum analyzes?
Robert Treat
From | Date | Subject | |
---|---|---|---|
Next Message | Andrew Sullivan | 2002-06-20 20:11:57 | Re: selecting all records where a column is null |
Previous Message | Dave Page | 2002-06-20 19:57:21 | Re: selecting all records where a column is null |