Обсуждение: turning of pg_xlog

Поиск
Список
Период
Сортировка

turning of pg_xlog

От
Jonny
Дата:
Hi,

I have installes Postgres 8.0.15 on a embedded Linux and have only 130
MB for Postgres.
Is it possible to turn off the comlete (Wal) pg_xlog? Because this is
the biggest part.
I found an Entry how to minimize the count of stored xlogs.
Is it possible to store it  to /dev/null or something else?
Something like ln -s /dev/null  pg_xlog/ (I know this does not work :-) ....


regards, Jonny

Re: turning of pg_xlog

От
Andrew Sullivan
Дата:
On Mon, Sep 29, 2008 at 01:00:41PM +0200, Jonny wrote:
> Is it possible to turn off the comlete (Wal) pg_xlog?

No.

A

--
Andrew Sullivan
ajs@commandprompt.com
+1 503 667 4564 x104
http://www.commandprompt.com/

Re: turning of pg_xlog

От
Chander Ganesan
Дата:
Jonny wrote:
> Hi,
>
> I have installes Postgres 8.0.15 on a embedded Linux and have only 130
> MB for Postgres.
> Is it possible to turn off the comlete (Wal) pg_xlog? Because this is
> the biggest part.
> I found an Entry how to minimize the count of stored xlogs.
> Is it possible to store it  to /dev/null or something else?
> Something like ln -s /dev/null  pg_xlog/ (I know this does not work
> :-) ....
You should also understand that if you set checkpoint_segments to a
small number, its still possible that PostgreSQL might use more than
that number for a large transaction.  As such, be careful about limiting
the space available to your xlogs, lest it cause transactions to fail
due to insufficient space for the xlogs.

--
Chander Ganesan
Open Technology Group, Inc.
One Copley Parkway, Suite 210
Morrisville, NC  27560
919-463-0999/877-258-8987
http://www.otg-nc.com


Re: turning of pg_xlog

От
Tom Lane
Дата:
Chander Ganesan <chander@otg-nc.com> writes:
> You should also understand that if you set checkpoint_segments to a
> small number, its still possible that PostgreSQL might use more than
> that number for a large transaction.

"Large transactions" have nothing to do with this.  You are confusing
Postgres' implementation with Oracle's.

            regards, tom lane

Re: turning of pg_xlog

От
Chander Ganesan
Дата:
Tom Lane wrote:
> Chander Ganesan <chander@otg-nc.com> writes:
>
>> You should also understand that if you set checkpoint_segments to a
>> small number, its still possible that PostgreSQL might use more than
>> that number for a large transaction.
>>
>
> "Large transactions" have nothing to do with this.  You are confusing
> Postgres' implementation with Oracle's.
>
>             regards, tom lane
Okay.  I guess my use of the word "transaction" here was erroneous -
probably comes from my oracle background :-( .  However, the fact
remains that its possible to exceed checkpoint_segments in certain
scenarios (I think the docs refer to it as a peak in log creation or
something like that), which - if it fills up the disk that the WALs are
on, can result in issues...  And one should be wary of that when trying
to limit the number of WALs in order to meet limited disk space
requirements.

--
Chander Ganesan
Open Technology Group, Inc.
One Copley Parkway, Suite 210
Morrisville, NC  27560
919-463-0999/877-258-8987
http://www.otg-nc.com


Re: turning of pg_xlog

От
"Plugge, Joe R."
Дата:
I might have to disagree here.  I personally experienced a scenario
where I had 500 MB of file space defined for my WAL log files and then
attempted (not thinking it through thoroughly) to perform a COPY of a
very large (1.8 GB) dataset and kept adding WAL files until my file
system filled up and then the instance crashed.

-----Original Message-----
From: pgsql-admin-owner@postgresql.org
[mailto:pgsql-admin-owner@postgresql.org] On Behalf Of Tom Lane
Sent: Tuesday, September 30, 2008 1:45 PM
To: Chander Ganesan
Cc: Jonny; pgsql-admin@postgresql.org
Subject: Re: [ADMIN] turning of pg_xlog

Chander Ganesan <chander@otg-nc.com> writes:
> You should also understand that if you set checkpoint_segments to a
> small number, its still possible that PostgreSQL might use more than
> that number for a large transaction.

"Large transactions" have nothing to do with this.  You are confusing
Postgres' implementation with Oracle's.

            regards, tom lane

--
Sent via pgsql-admin mailing list (pgsql-admin@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin