James> Hi.

James> Firstly, let me give you some background. James> I am running MySQL 3.22.22 (was 3.22.19b but I upgraded, and it didnt fix James> my problem) on a Slackware 3.6 Linux box running linux-threads and
glibc1. James> I am having problems with a select from a large (165Mb) table. James> I am trying to find a way to cut out duplicate entries.

James> This table is backed up configurations for Ascend Max boxes, and they are James> about 20k each, they are inserted each night, at midnight, and I am
trying James> to create a perl script to remove all the dup's (the configs dont change James> that much) from the table.

James> Here is the query I run:

James> select distinct configuration, hostname, type, timestamp, status from
configs

James> what happens then is that mysqld dies, and I get no errors or core's it James> just dies:

James> mysqld started on Mon May 10 10:54:35 NZST 1999 James> /usr/local/libexec/mysqld: ready for connections James> 990510 10:54:35 Aborting James> mysqld ended on Mon May 10 10:54:35 NZST 1999

James> I am not sure what the story is, but if anyone can help I would really James> appreciate it.

James> I have attached the perlscript to this mail, in the hope that someone can James> find a better way of doing it.

James> Thanks in advance.

Hi!

In this case its very likely that MySQL is waiting on a disk full condition!

- Check what MySQL is doing with 'mysqladmin proc' - Use 'top' to check that MySQL is running. - Check the MySQL log files for any recent entries. - Check the MySQL documentation, Section 'Debugging MySQL'.

Regards, Monty