Re: R: Autovacuum and frequent pg_bulkload

From: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>
To: Job <Job(at)colliniconsulting(dot)it>, "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org>
Subject: Re: R: Autovacuum and frequent pg_bulkload
Date: 2016-11-20 16:03:57
Message-ID: 9a61a198-bb69-94dd-61ea-be025b2850f8@aklaver.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On 11/20/2016 07:20 AM, Job wrote:
> Hi Adrian,
> Thank you for your fast reply!
>
>> FYI, Postgres 8.4 is over two years past EOL.
> Yes, i am aware; we will migrate next year but for application compatibility reason we have at the moment to the remain here..
>
>
>> What are the steps in the load/delete cycle?
>
> We need to load, with pg_bulkload, log datas for reporting.
> We load every hours about one million of row and twice a day we generated aggregated report.
> Then we delete old rows (we cannot use truncate because we only delere rows older than 24 hours).
> I think, but i can be wrong, that pg_bulkload do not reuse free space marked previously by the delete..

Can you be more specific.

In other words show the actual commands you use?

Also how are you defining free space, space the OS sees or space
Postgres sees?

How are you determining the free space?

>
> Only vacuum full empty free space, but tablea became locked.

VACUUM full returns space to the OS if that is what you mean?

>
> Thank you
> Francesco
>

--
Adrian Klaver
adrian(dot)klaver(at)aklaver(dot)com

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Tom Lane 2016-11-20 16:39:39 Re: How to change order sort of table in HashJoin
Previous Message Job 2016-11-20 15:20:31 R: Autovacuum and frequent pg_bulkload