BUG #4941: pg_stat_statements crash

From: "" <alr(dot)nospamforme(at)gmail(dot)com>
To: pgsql-bugs(at)postgresql(dot)org
Subject: BUG #4941: pg_stat_statements crash
Date: 2009-07-24 13:14:40
Message-ID: 200907241314.n6ODEeMm078336@wwwmaster.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs pgsql-hackers


The following bug has been logged online:

Bug reference: 4941
Logged by:
Email address: alr(dot)nospamforme(at)gmail(dot)com
PostgreSQL version: 8.4.0
Operating system: windows 2008,2003
Description: pg_stat_statements crash
Details:

i was testing for issues with reattach memory patch fix
I believe that in windows 2008 , and 2003 shared_preload_libraries =
'pg_stat_statements' can crash postgres

2009-07-23 17:48:22 EDT 3820 LOG: server process (PID 4608) was
terminated by exception 0xC0000005
2009-07-23 17:48:22 EDT 3820 HINT: See C include file "ntstatus.h" for a
description of the hexadecimal value.
2009-07-23 17:48:22 EDT 3820 LOG: terminating any other active server
processes

so far have been able to reproduce this by running pgbench on mutable
systems. ( two non production and 2 clean installs on MS virtual Server )
machines have 3+ gig ram
shared_buffers = 94MB
max_connections = 300

pgbench.exe -i -s 300 pgbench
and running over and over
pgbench -c 100 -C from windows xp client
please not it may not crash on the first or second run but when it crashes
the first time
it will then crash every time after that even with reboots. it may be that
the 'pg_stat_statements' log somehow gets corrupted?
i am unable to reproduce if -C is not used ?
i have also noticed that normally task manager splits the load between a
lot of postgres processes but when it is turned on 1 postgres process uses
50% load.
I have tested both the postgres 8.4.0.9202 and 8.4.9177 postgress.exe and
both seem to have same issue .
hope this info helps
Allen

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Tom Lane 2009-07-24 13:25:05 Re: BUG #4939: error query result
Previous Message Magnus Hagander 2009-07-24 12:13:49 Re: BUG #4938: upgrade to 8.4 issue

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2009-07-24 13:31:13 Re: When is a record NULL?
Previous Message Michael Gould 2009-07-24 13:10:26 Re: When is a record NULL?