Hi Sergey,
On 12/23/08 10:48 AM, Sergey Glukhov wrote:
> #At file:///home/gluh/MySQL/mysql-5.0-bug-41131/ based on
> revid:sergefp@stripped
>
> 2719 Sergey Glukhov 2008-12-23
> Bug#41131 "Questions" fails to increment - ignores statements instead stored
> procs(5.0 ver)
Please add a comment that the scope of this variable is always global.
[..]
> +FLUSH STATUS;
Is this flush necessary? Will it have any effect on the Queries status?
> # End of 5.0 tests
>
> === modified file 'sql/mysqld.cc'
> --- a/sql/mysqld.cc 2008-11-28 14:32:04 +0000
> +++ b/sql/mysqld.cc 2008-12-23 13:50:01 +0000
> @@ -6578,6 +6578,7 @@ struct show_var_st status_vars[]= {
> {"Qcache_queries_in_cache", (char*)&query_cache.queries_in_cache,
> SHOW_LONG_CONST},
> {"Qcache_total_blocks", (char*)&query_cache.total_blocks,
> SHOW_LONG_CONST},
> #endif /*HAVE_QUERY_CACHE*/
> + {"Queries", (char*) 0, SHOW_QUERIES},
Wouldn't it be better to always return global_query_id here? This way we
can avoid potential problems with saving/restoring thd::query_id.
What do you think?
-- Davi
