BUG #1641: Server crash at 182 connections - Mailing list pgsql-bugs

The following bug has been logged online:

Bug reference:      1641
Logged by:          Joek Hondius
Email address:      Joek@rem.nl
PostgreSQL version: 8.0.1
Operating system:   Windows XP pro
Description:        Server crash at 182 connections
Details:

Hi, i get server crashes when connecting more than 182 clients. This is very
consequent behaviour, i get it every time at the same number.

I could not find any previous message about this.

I could help with more verbose logs if you want.

-------------------------------------------------
Using: pg win 8.0.1 @ Win XP pro:

postgresql.conf:
max_connections = 500
shared_buffers = 8000

-------------------------------------------------

** start app to open 200 connections to server **
    ( the server crashes at 182 connections )

-- insert 1-180 connections here --


2005-05-03 11:05:49 LOG:  connection received: host=192.168.85.103
port=3732

2005-05-03 11:05:49 LOG:  connection authorized: user=big1
database=stresstest

2005-05-03 11:05:49 LOG:  connection received: host=192.168.85.103
port=3733

2005-05-03 11:05:49 LOG:  connection authorized: user=big1
database=stresstest

2005-05-03 11:05:49 LOG:  server process (PID 4516) exited with unexpected
status 128

2005-05-03 11:05:49 LOG:  terminating any other active server processes

2005-05-03 11:05:49 WARNING:  terminating connection because of crash of
another server process

2005-05-03 11:05:49 DETAIL:  The postmaster has commanded this server
process to roll back the current transaction and exit, because another
server process exited abnormally and possibly corrupted shared memory.

2005-05-03 11:05:49 HINT:  In a moment you should be able to reconnect to
the database and repeat your command.


-- insert 180 terminating connections here --

2005-05-03 11:05:50 WARNING:  terminating connection because of crash of
another server process

2005-05-03 11:05:50 DETAIL:  The postmaster has commanded this server
process to roll back the current transaction and exit, because another
server process exited abnormally and possibly corrupted shared memory.

2005-05-03 11:05:50 HINT:  In a moment you should be able to reconnect to
the database and repeat your command.

2005-05-03 11:05:50 LOG:  all server processes terminated; reinitializing

2005-05-03 11:05:50 LOG:  database system was interrupted at 2005-05-03
10:44:16 W. Europe Daylight Time

2005-05-03 11:05:50 LOG:  checkpoint record is at 0/1164D5C0

2005-05-03 11:05:50 LOG:  redo record is at 0/1164D5C0; undo record is at
0/0; shutdown TRUE

2005-05-03 11:05:50 LOG:  next transaction ID: 404465; next OID: 887898

2005-05-03 11:05:50 LOG:  database system was not properly shut down;
automatic recovery in progress

2005-05-03 11:05:50 LOG:  record with zero length at 0/1164D600

2005-05-03 11:05:50 LOG:  redo is not required

2005-05-03 11:05:50 LOG:  database system is ready

2005-05-03 11:05:50 LOG:  connection received: host=192.168.85.103
port=3735

2005-05-03 11:05:50 FATAL:  the database system is starting up

2005-05-03 11:05:50 LOG:  connection received: host=192.168.85.103
port=3736

2005-05-03 11:05:50 LOG:  connection authorized: user=big1
database=stresstest

2005-05-03 11:05:50 LOG:  connection received: host=192.168.85.103
port=3737

2005-05-03 11:05:50 LOG:  connection authorized: user=big1
database=stresstest