Postgres Pro
Downloads
Home   >   mailing lists

Re: Checkpoint_segments optimal value - Mailing list pgsql-general

From Prabhjot Sheena
Subject Re: Checkpoint_segments optimal value
Date
Msg-id CAOf_bQYBKzZohRD8R5_ig=6H2zv4X2gJuBsAVfedsSsN8Hd+5w@mail.gmail.com
Whole thread Raw
In response to Re: Checkpoint_segments optimal value  (John R Pierce <pierce@hogranch.com>)
Responses Re: Checkpoint_segments optimal value  (John R Pierce <pierce@hogranch.com>)
List pgsql-general
Tree view
The overall load of system is good. It is a vm but the other database is running normal and not doing much activity

 vmstat 5 10
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 0  2  22048 296804   5016 10829360    0    0     1     3    0    0  5  1 75 18  1
 9  1  22048 255184   5024 10868884    0    0     0    60 8955 10518 33  3 41 15  8
11  3  22048 197664   5024 10925944    0    0     0     0 7866 8767 37  2 44 12  6
12  1  22048 140000   5048 10976620    0    0     0    10 9181 10632 39  3 42 11  6
 7  1  22048 148068   5076 10970104    0    0     1    33 7456 7715 34  2 46  8 10
11  3  22048  87408   5076 11027832    0    0     0     0 9846 12631 31  4 38 14 14
 8  1  22048  48380   5084 11067456    0    0     0    35 7887 8902 33  3 43 11 10



On Thu, Jul 17, 2014 at 5:10 PM, John R Pierce <pierce@hogranch.com> wrote:
On 7/17/2014 5:01 PM, Prabhjot Sheena wrote:
i just did explain analyze and currently database is running slow coz of the query

explain ANALYZE SELECT account.id, account.organization_id, run.application_id, work_unit.script, work_unit.id, work_unit.start_time, run.id, work_unit.priority  FROM work_unit, run, account  WHERE  work_unit.status = 3 AND work_unit.run_id = run.id AND work_unit.type != 1 AND run.status = 1 AND run.account_id = account.id;

what does the overall workload on the system look like when this is so slow?   taking over a minute to do 1820 iterations of a 3138 row index scan seems way out of bounds for anything other than a very heavily overloaded server.

is this running on a virtual machine where there are other virtual servers contending for the same resources ?

btw, here's your analyze formatted pretty:  http://explain.depesz.com/s/XRI




-- 
john r pierce                                      37N 122W
somewhere on the middle of the left coast

pgsql-general by date:

Previous
From: Adrian Klaver
Date:
Subject: Re: Windows Installation User account - Correct database for us
Next
From: John R Pierce
Date:
Subject: Re: Checkpoint_segments optimal value