Skip site navigation (1) Skip section navigation (2)

WARNING: pgstat wait timeout

From: Anders Steinlein <anders(at)steinlein(dot)no>
To: PG Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: WARNING: pgstat wait timeout
Date: 2011-12-03 23:07:20
Message-ID: 4EDAABA8.6020909@steinlein.no (view raw or flat)
Thread:
Lists: pgsql-hackers
While doing some pgbench testing on a new server with 9.1.1, I noticed 
quite a lot of $subject in the logs. I did some Googling and found this 
previous thread: 
http://archives.postgresql.org/pgsql-hackers/2010-01/msg02897.php It 
doesn't seem like the issue was resolved?

I did:
# pgbench  -i -s 1000 bench1
# pgbench -c 8 -j 4 -T 1000 bench1

The server is:
PostgreSQL 9.1.1
Ubuntu 10.04 2.6.32-35-generic #78-Ubuntu SMP x86_64
Quad Core Xeon, 22GB RAM, 2x SATA RAID1

The warning seems to come in average 3-4 times a minute:
2011-12-03 23:07:31 CET::@:[13728]: WARNING:  pgstat wait timeout
2011-12-03 23:07:37 CET::@:[13728]: WARNING:  pgstat wait timeout
2011-12-03 23:07:47 CET::@:[13556]: WARNING:  pgstat wait timeout
2011-12-03 23:07:52 CET::@:[13729]: WARNING:  pgstat wait timeout

Some non-default postgresql.conf settings:
shared_buffers = 6GB
effective_cache_size = 12GB
work_mem = 16MB
synchronous_commit = off
checkpoint_segments = 16
checkpoint_completion_target = 0.9
autovacuum_max_workers = 6

Is this a bug, something wrong with my setup, or simply harmless noise 
(although annoying in that case)?

Regards,
-- a.

pgsql-hackers by date

Next:From: Andrew DunstanDate: 2011-12-03 23:12:38
Subject: Re: [PATCH] PostgreSQL fails to build with 32bit MinGW-w64
Previous:From: Tomas VondraDate: 2011-12-03 23:02:20
Subject: Re: cannot read pg_class without having selected a database / is this a bug?

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group