Need help in tuning

From: Phoenix Kiula <phoenix(dot)kiula(at)gmail(dot)com>
To: PG-General Mailing List <pgsql-general(at)postgresql(dot)org>
Subject: Need help in tuning
Date: 2014-08-07 01:21:17
Message-ID: CAFWfU=uZTt5oUY8d-2bce5v=ErSs4P9Ja1rFMfw2jfrfX0nxvg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

My PG server is still going down. After spending the weekend doing a
CLUSTER of my largest table (it's a RAID 1 system with SATA hard disks
and 4 GB memory, mostly devoted to PG) I still have this issue.

When I do a "top" command, 99% of the CPU and about 15% of the memory
is being taken by PG. When I press a "c" in the top UI, I see that
"postmaster" is doing some "CLUSTER". However, I don't do any more
clustering. The only automatic setting I can think of are autovacuum.

So, question: to the degree that my system allows for performance,
what steps can I take to find out what's happening? I see some things
mentioned: I/O, vacuum settings, pg_stats, pg_activity -- is there a
simple guide somewhere which shows me step by step what to do? Google
hasn't been much help. "Postgresql performance tunings" brings a lot
of esoteric articles.

Thanks!

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Tatsuo Ishii 2014-08-07 01:30:48 Re: The dangers of streaming across versions of glibc: A cautionary tale
Previous Message Peter Geoghegan 2014-08-07 01:12:53 Re: The dangers of streaming across versions of glibc: A cautionary tale