Re: VACUUM warning message.

From: <mallah(at)trade-india(dot)com>
To: <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: <pgman(at)candle(dot)pha(dot)pa(dot)us>, <mallah(at)trade-india(dot)com>, <pgsql-admin(at)postgresql(dot)org>
Subject: Re: VACUUM warning message.
Date: 2003-01-28 03:13:17
Message-ID: 1057.219.65.252.149.1043723597.squirrel@mail.trade-india.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

Hmm i do the below daily to this table. And this is pgsql 7.3

~~~~~~~~~~~~~~~~~~~~~~~
BEGIN WORK;
INSERT INTO archives.site_search SELECT * from public.site_search;
COMMIT;
TRUNCATE TABLE public.site_search;
~~~~~~~~~~~~~~~~~~~~~~
and then it is VACUMMED within 24 hrs.

The table public.site_search only under goes INSERTS thruout the day and
no updates and archives.site_search only undergoes a bulk insert once in 24 hrs
and "SELECTS" most of the time.

public.site_search starts from 0 records and ends up to 4000-5000 records
before the truncate.

archives.site_search contains 6,50,000 apprx.

Related Problem [ OT ] : this approach was taken because i was getting
very poor performance when i was inserting and reading data from the same
table within a transaction. only one table used to exists containg apprx
.6 million records. I feel there was some kind of locking going on
using this approach inserts and reads are from different table and
perforamance is much better.

Regds
Mallah.

> Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
>> I assume this is 7.2.X. I know this is fixed in 7.3.1 but I am not sure what release we fixed
>> it in. The HISTORY file doesn't have a specific mention of the fix. It is not something to
>> be worried about, as I remember.
>
> Mmm, I'm not sure. The known variant of this problem only affects certain system tables (the
> nailed-in-cache ones):
> http://archives.postgresql.org/pgsql-hackers/2002-11/msg00486.php
>
> This is evidently a user table:
>
>> WARNING: Rel site_search: Uninitialized page 1 - fixing
>
> I'd wonder about hardware failures, myself. Does the table appear to have lost a lot of rows
> that should have been there?
>
> Other interesting questions are whether the table has recently undergone a TRUNCATE, or
> deletion of a large fraction of its rows followed by VACUUM (probably VACUUM FULL).
>
> regards, tom lane
>
> ---------------------------(end of broadcast)--------------------------- TIP 6: Have you
> searched our list archives?
>
> http://archives.postgresql.org

-----------------------------------------
Get your free web based email at trade-india.com.
"India's Leading B2B eMarketplace.!"
http://www.trade-india.com/

In response to

Responses

Browse pgsql-admin by date

  From Date Subject
Next Message Tom Lane 2003-01-28 04:05:16 Re: VACUUM warning message.
Previous Message Andrew Sullivan 2003-01-27 23:18:20 Re: Replication