From: | elein <elein(at)varlena(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Cc: | elein <elein(at)varlena(dot)com> |
Subject: | Vacuuming of indexes on tables. |
Date: | 2006-04-18 04:26:04 |
Message-ID: | 20060418042604.GR20430@varlena.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Running version 8.1.2, 1-2G RAM. Configuration
set up to use available RAM. Running autovacuum.
I have a table with 8500000 rows. The table
gets only inserts (constantly). Once a day
a range of the rows is deleted and an explicit
vacuum analyze of the table is done.
The table has 4 indexes.
Much later in the day, a vacuum analyze of the
db showed that all of the indexes for that table
required significant vacuuming, although the
table did not.
A bit later, further vacuuming showed more indexes
required vacuuming and again the table did not.
In what situation would these indexes require the
further vacuuming? Due to the insert only activity
on the table (and the db, mostly), I cannot understand
why the indexes require further vacuuming.
Ideas?
--elein
elein(at)varlena(dot)com
From | Date | Subject | |
---|---|---|---|
Next Message | lash | 2006-04-18 05:51:27 | Postgresql and Creator2 commitChanges() |
Previous Message | Peter Wilson | 2006-04-18 00:13:13 | Re: plpgsql replication stored procedure |