Re: How to analyze a slowdown in 9.3.5?

From: Joe Van Dyk <joe(at)tanga(dot)com>
To:
Cc: "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org>
Subject: Re: How to analyze a slowdown in 9.3.5?
Date: 2015-01-12 22:04:43
Message-ID: CACfv+pJhPtk5GeLK0Q_Q9UpBBLDazefnZFJF5XfJONFP8rCNYw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Sun, Jan 11, 2015 at 8:07 AM, Michael Nolan <htfoot(at)gmail(dot)com> wrote:

>
>
> On Sat, Jan 10, 2015 at 8:54 PM, Melvin Davidson <melvin6925(at)gmail(dot)com>
> wrote:
>
>> Just curious. Have you checked that the tables are being vacuum/analyzed
>> periodically and that the statistics are up to date? Try running the
>> following query to verify:
>>
>>
> A vacuum analyze runs every night and there would not have been many
> inserts or updates to the tables used by the lookup function since the
> latest vacuum analyze. I think I may have even done a vacuum analyze on
> the two largest tables after the first DB shutdown.
>

One thing to check (I ran into this two weeks ago) -- even though vacuums
were happening, a query running on a standby machine was preventing the
vacuum process from removing the dead rows. You may want to check for bloat
or use 'vacuum verbose' to see if there's many dead rows not being cleaned
up.

Joe

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Antony Gelberg 2015-01-12 22:16:13 Re: Stuck trying to backup large database - best practice?
Previous Message Bosco Rama 2015-01-12 17:38:53 Re: Stuck trying to backup large database - best practice?