Re: [ADMIN] Excessive growth of pg_attribute and other system tables

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Steve Crawford <scrawford(at)pinpointresearch(dot)com>
Cc: "Matthew T(dot) O'Connor" <matthew(at)zeut(dot)net>, Jan Wieck <JanWieck(at)Yahoo(dot)com>, pgsql-hackers(at)postgreSQL(dot)org
Subject: Re: [ADMIN] Excessive growth of pg_attribute and other system tables
Date: 2005-03-31 20:06:13
Message-ID: 14686.1112299573@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin pgsql-hackers

I wrote:
> The light just went on ... system catalog updates don't generate
> statistics reports. Hence, autovacuum doesn't know any work is needed.

The above claim is too strong --- they do normally generate stats
updates. However, in a simple test I observed that
pg_stat_all_tables.n_tup_del did not seem to increment for the deletes
that occur when a temp table is dropped during backend exit. (Most
likely we aren't flushing out the final stats messages...)

Steve, is your app in the habit of creating lots of temp tables that are
not dropped explicitly? That would explain why you are getting bit more
than other people.

regards, tom lane

In response to

Responses

Browse pgsql-admin by date

  From Date Subject
Next Message Steve Crawford 2005-03-31 20:44:15 Re: [ADMIN] Excessive growth of pg_attribute and other system tables
Previous Message Tom Lane 2005-03-31 19:04:08 Re: [ADMIN] Excessive growth of pg_attribute and other system tables

Browse pgsql-hackers by date

  From Date Subject
Next Message Alvaro Herrera 2005-03-31 20:18:28 Re: Notes on lock table spilling
Previous Message Peter Eisentraut 2005-03-31 19:11:36 Re: [COMMITTERS] pgsql: Translation updates