Re: [GENERAL] Service outage: each postgres process use the exactamount of the configured work_mem - Mailing list pgsql-general

From Moreno Andreo
Subject Re: [GENERAL] Service outage: each postgres process use the exactamount of the configured work_mem
Date
Msg-id d5188cc0-5c39-a2de-4537-4051162073a9@evolu-s.it
Whole thread Raw
In response to Re: [GENERAL] Service outage: each postgres process use the exactamount of the configured work_mem  (Melvin Davidson <melvin6925@gmail.com>)
List pgsql-general
Il 14/04/2017 20:55, Melvin Davidson ha scritto:


On Fri, Apr 14, 2017 at 2:42 PM, Chris Mair <chris@1006.org> wrote:
you should increase shared_memory to 40GB. General philosphy is to allocate 80% of system memory to shared_memory

Uhm...

80% is too much, likely:

https://www.postgresql.org/docs/9.5/static/runtime-config-resource.html

Bye,
Chris.



Sorry, I had a brain fart, shared_buffers should be 25% system memory, or 12GB
The original set was 13 GB (I lowered to 3 GB to take a test), but probably I never acknowledged it with top...


--
Melvin Davidson
I reserve the right to fantasize.  Whether or not you
wish to share my fantasy is entirely up to you.


pgsql-general by date:

Previous
From: Moreno Andreo
Date:
Subject: Re: [GENERAL] Service outage: each postgres process use the exactamount of the configured work_mem
Next
From: Igor Neyman
Date:
Subject: Re: [GENERAL] Service outage: each postgres process use the exactamount of the configured work_mem