bzr push into mysql-trunk branch (andrei.elkin:3710 to 3711) Bug#13893310

[Andrei Elkin]

 3711 Andrei Elkin	2012-04-20
      BUG#13893310
      
      Using the recent push's bug id to tag fixing a race condition in
      rpl_parallel_change_master.
     @ mysql-test/suite/rpl/t/rpl_parallel_change_master.test
        further improving the gaps test. 
        The previous version appeared to test select count(*) = 1 from d2.t1
        having race condition with SQL thread. 
        Consistent-read isolation of innodb can't change the first found count(*).
        So other than the blocking (13) connection should be used to select
        in autocommit on mode.

    modified:
      mysql-test/suite/rpl/t/rpl_parallel_change_master.test
 3710 Mattias Jonsson	2012-04-20
      Bug#13907676: HA_ARCHIVE::INFO
      In WL#4305 the refactoring of the archive writer,
      it could flush the writer when it was not yet open.
      
      This was due to if bulk insert was used but no
      rows was actually inserted (write_row was never called),
      the writer was marked dirty even if it was not open.
      
      Fix was to only mark it as dirty if it was opened.

    modified:
      mysql-test/r/archive.result
      mysql-test/t/archive.test
      storage/archive/ha_archive.cc
=== modified file 'mysql-test/suite/rpl/t/rpl_parallel_change_master.test'
--- a/mysql-test/suite/rpl/t/rpl_parallel_change_master.test	2012-04-19 16:30:09 +0000
+++ b/mysql-test/suite/rpl/t/rpl_parallel_change_master.test	2012-04-20 16:10:18 +0000
@@ -71,13 +71,14 @@ INSERT INTO d1.t1 VALUES (13);
 INSERT INTO d2.t1 VALUES (8);  # this worker will race over one inserting (13)
 INSERT INTO d2.t1 VALUES (9);
 
---connection slave
+--connection slave1
 # make sure workers doing d2.t1 raced the one that occupied  with d1.t1
 --let $count= 1
 --let $table= d2.t1
 --let $wait_condition= select count(*) = 1 from $table where a = 8
 --source include/wait_condition.inc
 
+--connection slave
 # make worker executing (13) to error out
 COMMIT; 
 

No bundle (reason: useless for push emails).