Re: bzr commit into mysql-5.1 branch (ingo.struewing:2624) Bug#35807

[Mats Kindahl]

Ingo Strüwing wrote:
> Hi Mats,
> 
> Mats Kindahl, 10.07.2008 21:15:
> ...
>> Ingo Struewing wrote:
>>> #At file:///home2/mydev/bzrroot/mysql-5.1-bug35807/
>>>
>>>  2624 Ingo Struewing	2008-07-09
>>>       Bug#35807 - INSTALL PLUGIN replicates row-based, but not stmt-based
>>>       
>>>       INSTALL PLUGIN and UNINSTALL PLUGIN worked with statement-based and
>>>       mixed-mode replication only, but not with row-based replication.
>>>       
>>>       There is no statement-based replication of these statements.
>>>       But there was row-based replication of the inserts and deletes
>>>       to and from the mysql.plugin table.
>>>       
>>>       The fix is to suppress binlogging during insert and delete to
>>>       and from the mysql.plugin table.
> ...
>>> +SHOW BINLOG EVENTS;
> ...
>> If you want to verify that *nothing* is written to the binary log, you
>> can use the following code to see that you get zero.
>>
>> let $before_pos = query_get_value("SHOW MASTER STATUS", Position, 1);
>> INSTALL PLUGIN example SONAME 'ha_example.so';
>> let $after_pos = query_get_value("SHOW MASTER STATUS", Position, 1);
>> eval SELECT $after_pos - $before_pos AS Delta;
> ...
> 
> Thank you. I will do that change. Trying a weak excuse: "SHOW BINLOG
> EVENTS" came from a suggestion of the last review. ;-)

Ah. :)

My main objection is to use the plain "SHOW BINLOG EVENTS", since it
holds the version of the server in the response. You could equally use
"include/show_binlog_events.inc", but I think the above is more robust.

Just my few cents,
Mats Kindahl
-- 
Mats Kindahl
Lead Software Developer
Replication Team
MySQL AB, www.mysql.com