Re: REINDEX on large DB vs. DROP INDEX/CREATE INDEX - Mailing list pgsql-general

From Gregory Stark
Subject Re: REINDEX on large DB vs. DROP INDEX/CREATE INDEX
Date
Msg-id 873asmjfpz.fsf@oxford.xeocode.com
Whole thread Raw
In response to Re: REINDEX on large DB vs. DROP INDEX/CREATE INDEX  (Wes <wespvp@msg.bt.com>)
Responses Re: REINDEX on large DB vs. DROP INDEX/CREATE INDEX
List pgsql-general
"Wes" <wespvp@msg.bt.com> writes:

> I guess I should also turn off fsync for the duration.

It shouldn't make a big difference. fsync only happens at the end of a
transaction or at a checkpoint.

Since you're concerned with very long operations the slowdown at the end of
the transaction won't make a big difference.

Checkpoints could be an issue, but you would be better off just raising
checkpoint_segments and/or checkpoint_timeout to make sure you don't get one
more often than once every few minutes when you're doing large operations like
this.

--
  Gregory Stark
  EnterpriseDB          http://www.enterprisedb.com
  Ask me about EnterpriseDB's RemoteDBA services!

pgsql-general by date:

Previous
From: Gregory Stark
Date:
Subject: Re: check constraint question
Next
From: Steve Clark
Date:
Subject: best way to query