Bug #45214 get_master_version_and_clock does not report error when queries fail

        
The "get_master_version_and_clock(...)" function in sql/slave.cc ignores 
error and passes directly when queries fail, or queries succeed 
but the result retrieved is empty.
  
The "get_master_version_and_clock(...)" function should try to reconnect master
if queries fail because of transient network problems, and fail otherwise.
The I/O thread should print a warning if the some system variables do not 
exist on master (very old master)

mysql-test/extra/rpl_tests/rpl_get_master_version_and_clock.test:
  Added test file for bug #45214
mysql-test/suite/rpl/r/rpl_get_master_version_and_clock.result:
  Added test result for bug #45214
mysql-test/suite/rpl/t/rpl_get_master_version_and_clock.test:
  Added test file for bug #45214
sql/slave.cc:
  The 'is_network_error()' function is added for checking if the error is caused by network.
  Added a new return value (2) to 'get_master_version_and_clock()' function result set 
  to indicate transient network errors when queries fail, and the caller should 
  try to reconnect in this case.

2 files changed, 100 insertions, 0 deletions

diff --git a/mysql-test/suite/rpl/r/rpl_get_master_version_and_clock.result b/mysql-test/suite/rpl/r/rpl_get_master_version_and_clock.result
new file mode 100644
index 00000000000..263896b884a
--- /dev/null
+++ b/mysql-test/suite/rpl/r/rpl_get_master_version_and_clock.result
@@ -0,0 +1,40 @@
+stop slave;
+drop table if exists t1,t2,t3,t4,t5,t6,t7,t8,t9;
+reset master;
+reset slave;
+drop table if exists t1,t2,t3,t4,t5,t6,t7,t8,t9;
+start slave;
+SELECT IS_FREE_LOCK("debug_lock.before_get_UNIX_TIMESTAMP");
+IS_FREE_LOCK("debug_lock.before_get_UNIX_TIMESTAMP")
+1
+SELECT GET_LOCK("debug_lock.before_get_UNIX_TIMESTAMP", 1000);
+GET_LOCK("debug_lock.before_get_UNIX_TIMESTAMP", 1000)
+1
+set global debug= 'd,debug_lock.before_get_UNIX_TIMESTAMP';
+stop slave;
+start slave;
+SELECT RELEASE_LOCK("debug_lock.before_get_UNIX_TIMESTAMP");
+RELEASE_LOCK("debug_lock.before_get_UNIX_TIMESTAMP")
+1
+Slave_IO_Errno= 2013
+SELECT IS_FREE_LOCK("debug_lock.before_get_SERVER_ID");
+IS_FREE_LOCK("debug_lock.before_get_SERVER_ID")
+1
+SELECT GET_LOCK("debug_lock.before_get_SERVER_ID", 1000);
+GET_LOCK("debug_lock.before_get_SERVER_ID", 1000)
+1
+set global debug= 'd,debug_lock.before_get_SERVER_ID';
+stop slave;
+start slave;
+SELECT RELEASE_LOCK("debug_lock.before_get_SERVER_ID");
+RELEASE_LOCK("debug_lock.before_get_SERVER_ID")
+1
+Slave_IO_Errno= 2013
+set global debug= '';
+reset master;
+include/stop_slave.inc
+change master to master_port=SLAVE_PORT;
+start slave;
+*** must be having the replicate-same-server-id IO thread error ***
+Slave_IO_Errno= 1593
+Slave_IO_Error= Fatal error: The slave I/O thread stops because master and slave have equal MySQL server ids; these ids must be different for replication to work (or the --replicate-same-server-id option must be used on slave but this does not always make sense; please check the manual before using it).
diff --git a/mysql-test/suite/rpl/t/rpl_get_master_version_and_clock.test b/mysql-test/suite/rpl/t/rpl_get_master_version_and_clock.test
new file mode 100644
index 00000000000..3084bfc32fc
--- /dev/null
+++ b/mysql-test/suite/rpl/t/rpl_get_master_version_and_clock.test
@@ -0,0 +1,60 @@
+#
+# BUG#45214 
+# This test verifies if the slave I/O tread tries to reconnect to
+# master when it tries to get the values of the UNIX_TIMESTAMP, SERVER_ID,
+# COLLATION_SERVER and TIME_ZONE from master under network disconnection.
+# The COLLATION_SERVER and TIME_ZONE are got only on master server version 4.
+# So they can't be verified by test case here.
+# Finish the following tests by calling its common test script:  
+# extra/rpl_tests/rpl_get_master_version_and_clock.test. 
+# And meanwhile this test checks that the slave I/O thread refuses to start if slave
+# and master have the same server id (because this is a useless setup,
+# and otherwise SHOW SLAVE STATUS shows progress but all queries are
+# ignored, which has caught our customers), unless
+# --replicate-same-server-id.
+#
+
+source include/master-slave.inc;
+
+#Test case 1: Try to get the value of the UNIX_TIMESTAMP from master under network disconnection
+connection slave;
+let $debug_saved= `select @@global.debug`;
+
+let $debug_lock= "debug_lock.before_get_UNIX_TIMESTAMP";
+eval SELECT IS_FREE_LOCK($debug_lock);
+eval SELECT GET_LOCK($debug_lock, 1000);
+
+set global debug= 'd,debug_lock.before_get_UNIX_TIMESTAMP';
+source extra/rpl_tests/rpl_get_master_version_and_clock.test; 
+
+#Test case 2: Try to get the value of the SERVER_ID from master under network disconnection
+connection slave;
+let $debug_lock= "debug_lock.before_get_SERVER_ID";
+eval SELECT IS_FREE_LOCK($debug_lock);
+eval SELECT GET_LOCK($debug_lock, 1000);
+
+set global debug= 'd,debug_lock.before_get_SERVER_ID';
+source extra/rpl_tests/rpl_get_master_version_and_clock.test;
+
+eval set global debug= '$debug_saved';
+
+#Test case 3: This test checks that the slave I/O thread refuses to start
+#if slave and master have the same server id.
+connection slave;
+reset master;
+# replicate ourselves
+source include/stop_slave.inc;
+--replace_result $SLAVE_MYPORT SLAVE_PORT
+eval change master to master_port=$SLAVE_MYPORT;
+start slave;
+
+let $slave_param= Last_IO_Errno;
+let $slave_param_value= 1593;
+source include/wait_for_slave_param.inc;
+--echo *** must be having the replicate-same-server-id IO thread error ***
+let $last_io_errno= query_get_value("show slave status", Last_IO_Errno, 1);
+let $last_io_error= query_get_value("show slave status", Last_IO_Error, 1);
+echo Slave_IO_Errno= $last_io_errno;
+echo Slave_IO_Error= $last_io_error;
+
+# End of tests