From: Volkan YAZICI <yazicivo@ttmail.com>
To: pgsql-performance@postgresql.org
Subject: Unexpectedly Long DELETE Wait
Date: Thu, 07 Aug 2008 09:09:39 +0300
Message-ID: <874p5xwe3g.fsf@alamut.mobiliz.com.tr>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.0.60 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii

Hi,

Below command has been running since ~700 minutes in one of our
PostgreSQL servers.

  DELETE FROM mugpsreglog
        WHERE NOT EXISTS (SELECT 1
                            FROM mueventlog
                           WHERE mueventlog.eventlogid = mugpsreglog.eventlogid);

   Seq Scan on mugpsreglog  (cost=0.00..57184031821394.73 rows=6590986 width=6)
     Filter: (NOT (subplan))
        SubPlan
             ->  Seq Scan on mueventlog  (cost=0.00..4338048.00 rows=1 width=0)
             Filter: (eventlogid = $0)

Here is some information about related tables:

  # SELECT pg_relation_size('emove.mueventlog') / pow(1024, 2);
   ?column?
  ----------
      11440
  (1 row)
  
  # SELECT pg_relation_size('emove.mugpsreglog') / pow(1024, 2);
    ?column?
  -------------
   631.8046875
  (1 row)

And there isn't any constraints (FK/PK), triggers, indexes, etc. on any
of the tables. (We're in the phase of a migration, many DELETE commands
similar to above gets executed to relax constraints will be introduced.)

Here are related postgresql.conf lines:

  shared_buffers = 512MB
  max_prepared_transactions = 0
  work_mem = 8MB
  maintenance_work_mem = 512MB
  max_fsm_pages = 204800
  max_fsm_relations = 8192
  vacuum_cost_delay = 10
  wal_buffers = 2MB
  checkpoint_segments = 128
  checkpoint_timeout = 1h
  checkpoint_completion_target = 0.5
  checkpoint_warning = 1min
  effective_cache_size = 5GB
  autovacuum = off
  
And system hardware & software profile is:

  OS        : Red Hat Enterprise Linux ES release 4 (Nahant Update 5)
  PostgreSQL: 8.3.1
  Filesystem: GFS (IBM DS4700 SAN)
  CPU       : 4 x Quad Core Intel(R) Xeon(TM) CPU 3.00GHz
  Memory    : 8GB

Does anybody have an idea what might be causing the problem? Any
suggestions to improve the performance during such bulk DELETEs?


Regards.