Re: Interpreting shared_buffers setting - Mailing list pgsql-performance

From Jerry Sievers
Subject Re: Interpreting shared_buffers setting
Date
Msg-id 87fttb9uke.fsf@jsievers.enova.com
Whole thread Raw
In response to Re: Interpreting shared_buffers setting  (Jerry Sievers <gsievers19@comcast.net>)
List pgsql-performance
Jerry Sievers <gsievers19@comcast.net> writes:

> Bob Jolliffe <bobjolliffe@gmail.com> writes:
>
>> Excuse me if this is a silly question.  I am trying to fiddle with
>> shared_buffers setting on postgresql 10.6 on ubuntu 18.04 server.
>>
>> I have this at bottom of my config file:
>> shared_buffers = 1GB
>>
>> Yet when I check the setting from pg_setting I see something quite different:
>>
>> postgres=# SELECT name, setting FROM pg_settings where name = 'shared_buffers';
>>       name      | setting
>> ----------------+---------
>>  shared_buffers | 131072

Aaaaay!  Pasted junk...  Here's what I meant.


meta_a:postgres# select setting from pg_settings where name='shared_buffers';
 setting 
---------
 131072
(1 row)

meta_a:postgres# show shared_buffers;
 shared_buffers 
----------------
 1GB
(1 row)


>
> Why not use the show command which is good about output in human
> terms...
>
> psql (11.1 (Ubuntu 11.1-1.pgdg16.04+1))
> Type "help" for help.
>
> meta_a:postgres# select name, setting from pg_settings where name = 'shared_buffers');
> ERROR:  syntax error at or near ")"
> LINE 1: ...me, setting from pg_settings where name = 'shared_buffers');
>                                                                      ^
> meta_a:postgres# 
>
>>
>> Is this a question of units?  It looks like 128M.  Note when I change
>> the setting to 2GB in conf file I see 262144 from pg_setting.  I am
>> now unsure what the actual shared_buffers allocation is.  I cant see
>> anything in the docs which tells me how to interpret the integer.
>>
>> Any clarification welcome.
>>
>> Regards
>> Bob
>>
>>

-- 
Jerry Sievers
Postgres DBA/Development Consulting
e: postgres.consulting@comcast.net


pgsql-performance by date:

Previous
From: Saurabh Nanda
Date:
Subject: Re: How can sort performance be so different
Next
From: Tom Lane
Date:
Subject: Re: How can sort performance be so different