I apologize for not including the server information in the last e-mail. I
knew I would forget something. :)
PostgreSQL 8.1.4 (compiled from source)
Debian GNU/Linux 3.1
Intel(R) Pentium(R) D CPU 3.00GHz
1GB RAM
We have not be able to test the triggers anywhere else that has any load on
it. If we run it on a test set of data, it works fine because it is not
being hit by that many users. However, if we disable the triggers,
everything on the live server appears to work fine and PostgreSQL never dies.
Also, we are using the PostgreSQL ODBC driver from a Windows application we
have developed. It seems that if the user tries to access the server while
it restarts after this error, they get an error. However, if the user is
not doing anything when the server restarts and then tries to access data
later, it appears to automatically do the reconnect and everything is fine.
Thanks again for your assistance.
Thank you,
Carl M. Nasal II
BMA Enterprises, Inc.
Ian Barwick wrote:
2006/7/11, Carl M. Nasal II <cnasal@xxxxxxxxxxxxxxxxxx>:
(...)
Any ideas of what is causing the server to crash will be helpful.
Below are the lines from the PostgreSQL serverlog file when the crash
occurs:
----------------------------------------------------------------------------
LOG: server process (PID 29153) exited with exit code 255
LOG: terminating any other active server processes
WARNING: terminating connection because of crash of another server
process
(...)
It would be helpful if you provided the PostgreSQL version you're
using as well as details of the OS and possibly the hardware.
I've seen this kind of error in connection with hardware errors
(typically bad RAM or severe hard disk errors). Have you attempted
replicating this problem on another system to confirm / exclude this
as a possible cause?
Ian Barwick
