Hello Jon Olav!
* Jon Olav Hauglid <jon.hauglid@stripped> [10/09/15 13:17]:
> #At file:///export/home/x/mysql-5.5-runtime-bug56595/ based on
> revid:jon.hauglid@stripped
>
> 3137 Jon Olav Hauglid 2010-09-15
> Bug #56595 RENAME TABLE causes assert on OS X
>
> The problem was that RENAME TABLE caused an assert if the system variable
> lower_case_table_names was 2 (default on Mac OS X) and the old table name
> was given in upper case. This caused lowercase_table2.test to fail.
>
> The assert checks that an exclusive metadata lock is held by the connection
> trying to do RENAME TABLE - specificially during updates of table triggers.
> The assert was triggered since the check is case sensitive and the lock
> was held on the normalized (lower case) version of the table name.
>
> This patch fixes the problem by making sure a normalized version of the
> table name is used for the metadata lock check, while using a non-normalized
> version of the table name for the rename of trigger files. The same is done
> for ALTER TABLE ... RENAME.
>
> No test case added since this problem is already covered by
> lowercase_table2.test
>
...
> === modified file 'sql/sql_table.cc'
> --- a/sql/sql_table.cc 2010-09-03 07:42:51 +0000
> +++ b/sql/sql_table.cc 2010-09-15 09:08:18 +0000
> @@ -6555,8 +6556,8 @@ bool mysql_alter_table(THD *thd,char *ne
> (need_copy_table != ALTER_TABLE_METADATA_ONLY ||
> mysql_rename_table(save_old_db_type, db, table_name, new_db,
> new_alias, NO_FRM_RENAME)) &&
> - Table_triggers_list::change_table_name(thd, db, table_name,
> - new_db, new_alias)))
> + Table_triggers_list::change_table_name(thd, db, alias, table_name,
> + new_db, new_alias)))
I suspect that alignment of the above code is wrong.
Please check.
> {
> /* Try to get everything back. */
> error=1;
>
> === modified file 'sql/sql_trigger.cc'
> --- a/sql/sql_trigger.cc 2010-08-31 09:52:56 +0000
> +++ b/sql/sql_trigger.cc 2010-09-15 09:08:18 +0000
> @@ -1874,6 +1874,7 @@ Table_triggers_list::change_table_name_i
>
> @param[in,out] thd Thread context
> @param[in] db Old database of subject table
> + @param[in] old_alias Old alias of subject table
> @param[in] old_table Old name of subject table
> @param[in] new_db New database for subject table
> @param[in] new_table New name of subject table
> @@ -1890,6 +1891,7 @@ Table_triggers_list::change_table_name_i
> */
>
> bool Table_triggers_list::change_table_name(THD *thd, const char *db,
> + const char *old_alias,
> const char *old_table,
> const char *new_db,
> const char *new_table)
> @@ -1910,8 +1912,11 @@ bool Table_triggers_list::change_table_n
> DBUG_ASSERT(thd->mdl_context.is_lock_owner(MDL_key::TABLE, db, old_table,
> MDL_EXCLUSIVE));
>
> - DBUG_ASSERT(my_strcasecmp(table_alias_charset, db, new_db) ||
> - my_strcasecmp(table_alias_charset, old_table, new_table));
> + DBUG_ASSERT((lower_case_table_names &&
> + (my_strcasecmp(table_alias_charset, db, new_db) ||
> + my_strcasecmp(table_alias_charset, old_table, new_table))) ||
> + (!lower_case_table_names &&
> + (strcmp(db, new_db) || strcmp(old_alias, new_table))));
Sorry. It seems that I have misled you with this assert. When
lower_case_table_name is 0 table_alias_charset points to
my_charset_bin and thus in the original assert case-sensitive
comparison is performed. So original assert is correct.
So I think you should revert this change.
Also probably it makes sense to add test case which will do
"RENAME TABLE t1 TO T1" to lowercase_fs_off.test.
What do you think?
...
Otherwise I am OK with this patch and think that it can be
pushed after addressing/considering above points.
--
Dmitry Lenev, Software Developer
Oracle Development SPB/MySQL, www.mysql.com
Are you MySQL certified? http://www.mysql.com/certification
