MySQL Bugs: #62182: Core dump

Bug #62182	Core dump
Submitted:	16 Aug 2011 19:11	Modified:	16 Sep 2011 23:58
Reporter:	Marcelo Rodrigues	Email Updates:
Status:	No Feedback	Impact on me:	None
Category:	MySQL Proxy: Scripts	Severity:	S1 (Critical)
Version:	0.8.1	OS:	Any
Assigned to:		CPU Architecture:	Any
Tags:	core dump

Description:
I have two MySQL servers, running on two machines with CentOS release 5.2.

Versions used:
[root@new ~]# rpm -q mysql
mysql-5.0.77-4.el5_4.1
[root@new ~]# rpm -q mysql-server
mysql-server-5.0.77-4.el5_4.1
[root@new ~]# rpm -q lua
lua-5.0.2-1.el5.rf

A third machine acts as webserver and Mysqlproxy. This is, too, a CentOS release 5.2, and is 32 bits.

Mysqlproxy is a tgz, version 0.8.1.
Using the script "rw-splitting.lua", provided with the instalation.
On server is declared as RW (proxy-backend-address), and the other one as RO (proxy-read-only-backend-address).

Problem: About one minute after being started, the mysqlproxy process dies with a segmentation fault. It doesn´t matter if I connect or not: it dies either way.

If I comment out the "rw-splitting.lua" script, the problem goes away.

Tried running it trough command line, no deamon, with "log-level=debug". Not much comes back. Just the initialization and the segmentation fault warning.

How to repeat:
Just startup the process, using the rw-splitting.lua script. It crashes everytime, without exeption.

If you are able to get a back trace compare it with the one printed in bug: http://bugs.mysql.com/bug.php?id=54337 (see last comment from Sveta). Thanks.

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".