>>>>> "Christopher" == Christopher Lee <leec@stripped> writes:
Christopher> Hi,
Christopher> I am running 3.22.23b on a DEC alpha w/ Digital UNIX. I get the
Christopher> "Error (1114): the table... is full" message when I try to run a big
Christopher> GROUP BY statement. However, setting SQL_BIG_TABLES to 1 doesn't seem
Christopher> to fix the problem:
mysql> set SQL_BIG_TABLES= 1;
Christopher> Query OK, 0 rows affected (0.00 sec)
mysql> delete from seqlib;
Christopher> Query OK, 0 rows affected (0.00 sec)
mysql> insert into seqlib select t1.cluster_id,t2.lib_id from br_seq
Christopher> t1,sequence t2 where t1.ug_id=t2.ug_id group by
Christopher> t1.cluster_id,t2.lib_id;
Christopher> ERROR 1114: The table 'SQL9241ed_0' is full
Christopher> What's going on here? Am I doing something wrong? I don't see any
Christopher> hints in the documentation or mailing list about what more I should
Christopher> do, so I would really appreciate it if someone could tell me why
Christopher> setting SQL_BIG_TABLES doesn't work for me...
Hi!
I agree; SQL_BIG_TABLES should work in this case.
Do you have any idea how big the 'group by' should be. I assume you
are not reaching the 2G limit with it?
(It's likely that 3.23.x should work, but I don't recommend to upgrade
yet)
Regards,
Monty
