Bug#16228 (RBR: Slave SQL thread retries infinitely):

  Transaction aborted on slave should be retries. The OPTION_BEGIN
  bit was cleared prematurely. Removed dependence of code on value
  of OPTION_BEGIN bit when executing retries.

mysql-test/r/rpl_ndb_basic.result

@@ -71,13 +71,47 @@ CREATE TABLE `t1` ( `nid` int(11) NOT NULL default '0',
 PRIMARY KEY USING HASH (`nid`)) 
 ENGINE=ndbcluster DEFAULT CHARSET=latin1;
 INSERT INTO t1 VALUES(1,"XYZ1","ABC1");
+**** On Slave ****
 BEGIN;
 UPDATE t1 SET `nom`="LOCK" WHERE `nid`=1;
 set GLOBAL slave_transaction_retries=1;
+**** On Master ****
 UPDATE t1 SET `nom`="DEAD" WHERE `nid`=1;
-SHOW SLAVE STATUS;
-Slave_IO_State	Master_Host	Master_User	Master_Port	Connect_Retry	Master_Log_File	Read_Master_Log_Pos	Relay_Log_File	Relay_Log_Pos	Relay_Master_Log_File	Slave_IO_Running	Slave_SQL_Running	Replicate_Do_DB	Replicate_Ignore_DB	Replicate_Do_Table	Replicate_Ignore_Table	Replicate_Wild_Do_Table	Replicate_Wild_Ignore_Table	Last_Errno	Last_Error	Skip_Counter	Exec_Master_Log_Pos	Relay_Log_Space	Until_Condition	Until_Log_File	Until_Log_Pos	Master_SSL_Allowed	Master_SSL_CA_File	Master_SSL_CA_Path	Master_SSL_Cert	Master_SSL_Cipher	Master_SSL_Key	Seconds_Behind_Master
-<Slave_IO_State>	127.0.0.1	root	MASTER_PORT	1	master-bin.000001	<Read_Master_Log_Pos>	<Relay_Log_File>	<Relay_Log_Pos>	master-bin.000001	Yes	No				<Replicate_Ignore_Table>			146	Error in Write_rows event: error during transaction execution on table test.t1	0	<Exec_Master_Log_Pos>	<Relay_Log_Space>	None		0	No						<Seconds_Behind_Master>
+**** On Slave ****
+SHOW SLAVE STATUS;;
+Slave_IO_State	<Slave_IO_State>
+Master_Host	127.0.0.1
+Master_User	root
+Master_Port	MASTER_PORT
+Connect_Retry	1
+Master_Log_File	master-bin.000001
+Read_Master_Log_Pos	<Read_Master_Log_Pos>
+Relay_Log_File	<Relay_Log_File>
+Relay_Log_Pos	<Relay_Log_Pos>
+Relay_Master_Log_File	master-bin.000001
+Slave_IO_Running	Yes
+Slave_SQL_Running	No
+Replicate_Do_DB	
+Replicate_Ignore_DB	
+Replicate_Do_Table	
+Replicate_Ignore_Table	<Replicate_Ignore_Table>
+Replicate_Wild_Do_Table	
+Replicate_Wild_Ignore_Table	
+Last_Errno	146
+Last_Error	Error in Write_rows event: error during transaction execution on table test.t1
+Skip_Counter	0
+Exec_Master_Log_Pos	<Exec_Master_Log_Pos>
+Relay_Log_Space	<Relay_Log_Space>
+Until_Condition	None
+Until_Log_File	
+Until_Log_Pos	0
+Master_SSL_Allowed	No
+Master_SSL_CA_File	
+Master_SSL_CA_Path	
+Master_SSL_Cert	
+Master_SSL_Cipher	
+Master_SSL_Key	
+Seconds_Behind_Master	<Seconds_Behind_Master>
 set GLOBAL slave_transaction_retries=10;
 START SLAVE;
 select * from t1 order by nid;

mysql-test/t/disabled.def

@@ -25,7 +25,7 @@ rpl_ddl                 : Bug#15963 SBR does not show "Definer" correctly
 rpl_ndb_2innodb         : Bugs#17400: delete & update of rows in table without pk fails
 rpl_ndb_2myisam         : Bugs#17400: delete & update of rows in table without pk fails
 rpl_ndb_auto_inc        : Bug#17086
-rpl_ndb_basic           : Bug#16228 [IN REVIEW]
+#rpl_ndb_basic           : Bug#16228 [IN REVIEW]
 rpl_ndb_charset         : Bug#17246
 rpl_ndb_ddl             : Bug#17400: delete & update of rows in table without pk fails
 rpl_ndb_delete_nowhere  : Bug#17400: delete & update of rows in table without pk fails

mysql-test/t/rpl_ndb_basic.test

@@ -98,6 +98,7 @@ INSERT INTO t1 VALUES(1,"XYZ1","ABC1");
 # cause a lock on that row on the slave
 --sync_slave_with_master
 --connection slave
+--echo **** On Slave ****
 BEGIN;
 UPDATE t1 SET `nom`="LOCK" WHERE `nid`=1;
 
@@ -107,6 +108,7 @@ set GLOBAL slave_transaction_retries=1;
 # now do a change to this row on the master
 # will deadlock on the slave because of lock above
 --connection master
+--echo **** On Master ****
 UPDATE t1 SET `nom`="DEAD" WHERE `nid`=1;
 
 # wait for deadlock to be detected
@@ -119,14 +121,14 @@ UPDATE t1 SET `nom`="DEAD" WHERE `nid`=1;
 # replication should have stopped, since max retries where not enough
 # verify with show slave status
 --connection slave
+--echo **** On Slave ****
 --replace_result $MASTER_MYPORT MASTER_PORT
 --replace_column 1 <Slave_IO_State> 7 <Read_Master_Log_Pos> 8 <Relay_Log_File> 9 <Relay_Log_Pos> 16 <Replicate_Ignore_Table> 22 <Exec_Master_Log_Pos> 23 <Relay_Log_Space> 33 <Seconds_Behind_Master>
-SHOW SLAVE STATUS;
+--query_vertical SHOW SLAVE STATUS;
 
 # now set max retries high enough to succeed, and start slave again
 set GLOBAL slave_transaction_retries=10;
 START SLAVE;
-
 # wait for deadlock to be detected and retried
 # should be the same sleep as above for test to be valid
 --sleep 5

sql/slave.cc

@@ -3093,6 +3093,7 @@ static int exec_relay_log_event(THD* thd, RELAY_LOG_INFO* rli)
           else
           {
             exec_res= 0;
+            end_trans(thd, ROLLBACK);
 	    /* chance for concurrent connection to get more locks */
             safe_sleep(thd, min(rli->trans_retries, MAX_SLAVE_RETRY_PAUSE),
 		       (CHECK_KILLED_FUNC)sql_slave_killed, (void*)rli);
@@ -3110,8 +3111,16 @@ static int exec_relay_log_event(THD* thd, RELAY_LOG_INFO* rli)
                           "the slave_transaction_retries variable.",
                           slave_trans_retries);
       }
-      if (!((thd->options & OPTION_BEGIN) && opt_using_transactions))
-         rli->trans_retries= 0; // restart from fresh
+      else if (!((thd->options & OPTION_BEGIN) && opt_using_transactions))
+      {
+        /*
+          Only reset the retry counter if the event succeeded or
+          failed with a non-transient error.  On a successful event,
+          the execution will proceed as usual; in the case of a
+          non-transient error, the slave will stop with an error.
+         */
+        rli->trans_retries= 0; // restart from fresh
+      }
      }
     return exec_res;
   }