Hi Sachin, Andrei, I'm looking at test cases for START ALTER / Lag-free Alter on Slave. And I happened to notice something in test case rpl.rpl_start_alter_bugs: --connection master set global binlog_alter_two_phase=true; # ... ALTER TABLE t1 DROP PRIMARY KEY; ALTER TABLE t1 ADD UNIQUE KEY ui (i); ALTER TABLE t1 ADD PRIMARY KEY (i); Note the set *global* binlog_alter_two_phase. This will not affect the testcase at all, as the ALTER TABLE queries are using the *session* value (which defaults to false). This can be verified by checking the binlog after running the testcase, mysql-test/var/mysqld.1/data/master-bin.000001 . It has only normal ALTER TABLE query events, no START ALTER / COMMIT ALTER. So it looks strange that a start_alter test doesn't use START ALTER, and I just wanted to point this out to you in case it is a mistake and maybe the test coverage of START ALTER is less than intended? A quick grep shows similar SET GLOBAL in other start_alter tests, but I did not investigate if they have the same issue. Hope this helps, - Kristian.